arXiv Daily Digest - 2026-06-12

Anomaly detection in multivariate time series is challenged by four structurally distinct anomaly types -- point (isolated spikes), distributional (level shifts), temporal (rhythm changes), and collective (inter-sensor correlation breakdowns) -- each requiring different feature representations. Most unsupervised methods target only one or two types and provide limited interpretability. We present CRAFTIIF (Cross-Resolution Analytic Four-Type Interpretable Isolation Forest), a fully unsupervised framework targeting all four types without dataset-specific tuning. CRAFTIIF generates K=500 random analytic wavelet feature draws across four families (Morlet, DOG, Haar, Coiflet), each targeting a specific anomaly type, feeding five structured Isolation Forests -- one per type plus a meta-IF for compound anomalies. An adaptive Otsu/MAD threshold calibrates detection automatically across anomaly rates from 0.1% to 69.2%. Because each IF is trained exclusively on type-specific features, branch firing provides direct anomaly-type attribution by construction, without post-hoc explanation. Evaluated on all 19 datasets of the mTSBench benchmark (Zhou et al., TMLR 2026), CRAFTIIF achieves mean F1=0.228 (all 19 datasets) and F1=0.322 (13 detectable datasets), ranking first among all 25 evaluated methods on VUS-PR (0.463 vs. previous best 0.329, +40.7%). A diagnostic framework -- oracle F1, detectability limits, and branch separation ratios -- identifies 6 of 19 datasets as fundamentally undetectable by any unsupervised method. Ablation over 11 conditions confirms adaptive thresholding (+38% F1), four-branch structure (+20%), and meta-IF (+23%) are each essential. Code: https://github.com/smitswil/craftiif

Supramolecular chemistry, which includes the study of non-covalent host-guest assemblies, has advanced various applications. However, designing host-guest systems remains time-consuming, requiring days of dry-lab verification per candidate pair. Although LLMs have emerged as a fast alternative with strong performance on molecular binding tasks, no benchmark currently systematically evaluates LLMs for host-guest reasoning across fundamental supramolecular chemistry tasks, e.g., binding affinity prediction. To this end, we collaborate with domain experts to release the first Supramolecular Benchmark, called SupraBench, to evaluate LLMs in chemistry reasoning. Specifically, we design four fundamental tasks, i.e., binding affinity prediction, top-binder selection, solvent identification, and host-guest description, plus an auxiliary vision-based task for molecular identification. We also release SupraPMC, a curated 16M-token corpus of Supramolecular chemistry articles distilled from Europe PMC, to support the adaptation to the supramolecular domain. We benchmark a broad range of open and proprietary LLMs and find that LLMs leave substantial headroom across all tasks. Domain adaptation pretraining over SupraPMC transfers cleanly to in-distribution regression but trades off against strict letter-format output. Moreover, the difficulty profile differs sharply across task families, revealing distinct failure modes that indicate specific gaps in current supramolecular chemistry reasoning. Our source codes and benchmark datasets are available at https://github.com/Tianyi-Billy-Ma/SupraBench.

We present MaxProof, a population-level test-time scaling framework for competition-level mathematical proof in the MiniMax-M3 series. M3 first trains three proof-oriented capabilities -- proof generation, proof verification, and critique-conditioned proof repair -- using a defense-in-depth generative verifier engineered for low false-positive rate. These capabilities are merged into a single released M3 model. At test time, MaxProof treats the model as a generator, verifier, refiner, and ranker, searches over a population of candidate proofs, and returns one final proof through tournament selection. With MaxProof test-time scaling, the M3 model reaches 35/42 on IMO 2025 and 36/42 on USAMO 2026, exceeding the human gold-medal threshold on both.

AI coding agents are increasingly used to generate pull requests (PRs) that propose code fixes in software projects. From a first exploration of the AIDev dataset, we find that 46.41\% of the fixes proposed by the agents Copilot, Devin, Cursor, and Claude are rejected. This represents a significant amount of wasted resources that require human reviews, verifications, and running tests and validations for fixes that are merely discarded. Our goal in this paper is to understand the failure modes of AI-agents, an understanding that is crucial for better integrating AI-agents as efficient teammates. In this paper, we conduct a qualitative study on a representative sample of 306 non-merged pull requests created or co-authored by the agents mentioned earlier, followed by a quantitative analysis of the reasons for rejection. Our qualitative findings identify 14 reasons divided into four high-level categories for rejecting AI-agent fixes. We observe that developers can reject fixes due to fixes whose implementation is incorrect (e.g., incomplete, wrong approach), fixes that do not pass the continuous integration (CI) pipelines and fail tests, fixes for which the agent is unable to perform the implementation (e.g., no code generated, sessions lost), and fixes whose priority is low. Our results shed light on the importance of better guiding the model at these levels: (1) proposing hints about the approach to follow for fixing an issue, (2) outlining constraints or limitations regarding the approaches that should not be taken, and (3) instructing the agent on how to validate the implementation through CI pipelines and without introducing a breaking change. Our results suggest the need for good prioritization of tasks so that generated fixes do not lead to wasted human review efforts or wasted agent resources (e.g., tokens, compute, or allowed number of requests).

Automatic speech recognition (ASR) correction has traditionally focused on isolated utterances or short local contexts. However, as text and speech become increasingly interleaved in long interactions, ASR correction requires conversation-level contextual evidence. Existing ASR correction methods often rely on the current hypothesis or concatenate raw dialogue history. In such contexts, sparse correction evidence can be difficult to locate amid redundancy and noise. Addressing these challenges, we propose an ontology memory-augmented ASR correction framework for long text-speech interleaved conversations. The framework organizes preceding interaction history into a dynamically updatable ontology memory, where entities, terminology, surface variants, potential ASR confusions, and semantic relations are stored as retrievable nodes for context-grounded correction. To evaluate this setting, we construct RAMC-Corr, a dataset derived from MAGIC-RAMC for long-range ASR correction with grounded context. Experiments on RAMC-Corr show that our method improves over direct correction in 9 out of 10 paired backbone-setting combinations and encourages more selective and evidence-grounded corrections for context-dependent ASR errors.

Editing pretrained neural networks requires specialized algorithms tailored to specific objectives. Designing such algorithms is often time-consuming and demands significant effort. We present an exploratory framework that formulates neural model editing as a reinforcement learning problem, where agents modify models using reward feedback. We introduce two environments: MaskWorld, where agents scale weights multiplicatively, and ShiftWorld, where agents apply additive weight updates. The reward function combines a utility-preservation objective with a task-specific editing objective, enabling agents to learn targeted modifications while maintaining overall model performance. We evaluate the framework on bias mitigation in text classification and machine unlearning in image classification, both of which traditionally rely on specialized algorithms. Our results show that the learned policies reduce forget set accuracy to nearly 0% while preserving over 90% retain set accuracy on the unlearning task. In the bias mitigation setting, the learned policies improve bias-related performance by more than 5% while maintaining general classification utility. Our findings show that neural model editing can be cast as a reinforcement learning problem, allowing editing policies to be learned from reward feedback rather than manually engineered for each task.

Equilibrium Propagation offers a compelling alternative to traditional machine learning for training energy-based networks. Here we demonstrate a hybrid optical-digital implementation of EP using a Spatial Photonic Ising Machine (SPIM). The SPIM exploits the gauge transformation method to optically encode both continuous neuron states and rank-1 binary trainable patterns as phase modulations via a spatial light modulator, with inference realized using a finite difference scheme. The experimental system is evaluated on the Wine classification dataset. The potential of this approach, including the use of continuous couplings and structured coupling matrices, is evaluated numerically on the more complex MNIST dataset. Our work provides a concrete pathway toward energy-efficient physical implementations of Equilibrium Propagation.

Novelty is a crucial metric for assessing the quality of academic papers. Scholars strive to highlight the novel aspects of their work, particularly in the title, abstract, and introduction. Peer review, serving as the gatekeeper of scientific rigor, rigorously evaluates the novelty of papers, yet a cognitive gap may exist between author self-promotion and reviewer evaluation. To investigate this, we analyzed 15,328 academic papers published in Nature Communications from 2016 to 2021, along with their peer-review comments. We found that both reviewers and authors emphasize result-oriented innovation, with reviewers adopting a more comprehensive evaluation perspective. Furthermore, by examining promotional intensity against inherent paper novelty, we found that its effect depends on the paper's actual innovation level. Highly innovative papers benefit from stronger promotional language, receiving more positive evaluations. We also found that promotional language significantly correlates with reviewer disagreement on novelty specifically for papers of moderate innovativeness, whereas it has negligible impact for papers with either very high or very low novelty. This reveals how promotional language operates most prominently in the gray area of academic evaluation.

Diffusion models have seen wide adoption for 3D molecular generation, yet they offer no principled signal of when a generated molecule is likely to be of low quality. We propose a post-hoc method for estimating per-sample uncertainty in pretrained molecular diffusion models. Building on a Laplace approximation of the denoising network, we measure the variability of the noise prediction across the generation trajectory. Empirically, we show that the resulting uncertainty score is informative of sample quality, exhibiting a negative correlation with established sample-level quality metrics. We further study how the proposed uncertainty score can be used to filter generated samples, improving model performance via test-time scaling.

AI-agents (e.g., GitHub Copilot) collaborate as teammates in different software engineering tasks, including code generation proposed through pull requests (Agentic-PRs). For better agent efficiency, developers create instruction files that guide the AI-agents, including how to navigate the project, locate the right components, run tests, respect best practices, and more. In this paper, we investigate the relationship between the creation of these instructions and the performance of AI-agents in creating better pull requests, which have a higher chance of success (i.e., the merge rate), address more complex tasks (e.g., code churn), and require less effort to be merged (e.g., time to merge). To this end, we analyze 15,549 agentic PRs from 148 projects in the AIDev dataset. Using the three dimensions, we compare each project before and after the creation of the instruction files. We find that specifying instructions for AI-agents does not necessarily lead to better results. With the instruction files, 27.7\% of the projects increased their merge rate by at least 20\%, while 26.35\% decreased it. The same observation is seen with the amount of changes (e.g., code churn, number of modified files) and with the efforts to merge an agentic PR (e.g., merge time and number of comments). From a first exploration, we find that projects that managed to increase their merge rate have substantially longer instruction files, which are also well structured into a higher number of sections and sub-sections. Our results motivate the need for research to assist practitioners in framing the development of instruction files as a software engineering activity (aka, \textbf{Instructions-as-Code}).

Graph clustering - partitioning the node set of a graph into disjoint subsets that reflect some latent information - is a fundamental problem as it finds applications in a myriad of different scenarios. While this classic problem has been tackled for decades by different communities, a recent variation of the problem driven by real data considers the scenario where nodes have attributes that are also informative. This has triggered novel methods that simultaneously leverage network information (edges) and node information (attributed) in the design of novel clustering algorithms. This work proposes a novel framework that builds on prior works that have applied graph neural networks (GNN) to graph clustering. The proposed framework operates in rounds of self learning in a fully unsupervised setting. In each round, a GNN generates representations for nodes that are used to cluster the nodes. This clustering influences the graph used to generate the node representation in the next round. Moreover, a context graph built in each round using the original graph is used to generate the node representations. Empirical results show that the proposed methodology extracts information from both network edges and node attributes in synthetic data, outperforming algorithms focused solely on the network or attributes when neither are very informative. Multiple rounds of learning also improve the performance and always outperforms a long single round of training (i.e., classic GNN graph clustering). When considering real datasets, empirical results indicate that the proposed methodology is competitive to state-of-the-art methods when cluster sizes are balanced.

Neural operators have emerged as a powerful data-driven approach for solving time-dependent PDEs. Among recent advances, memory-augmented neural operators explicitly incorporate past states and have achieved remarkable performance under low-resolution observation settings. However, existing approaches apply a fixed memory weight regardless of observation conditions, such as resolution or physical parameters, limiting their adaptability. Our preliminary experiments reveal that optimal memory weight varies with resolution and viscosity, implying that a fixed memory weight cannot simultaneously optimize performance across diverse settings. We propose AMGFNO, which dynamically modulates memory weight through a learnable gate. On the Kuramoto-Sivashinsky and Burgers' equations, AMGFNO achieves 55-79% nRMSE reduction over at low resolution, with the learned gate value automatically decreasing from $\bar{g} \approx 0.7$ to near-zero as resolution increases.

Recent advances in large language models (LLMs) have prompted claims that such systems exhibit agency or qualify as moral agents. This paper argues that these attributions are misguided. We maintain that moral responsibility requires commitment-bearing agency grounded in intrinsic intentionality and self-attributed action, and that such agency constitutes the form of free will relevant to responsibility. Although LLMs generate coherent and normatively evaluable outputs, their operation is fully characterized by probabilistic input-output mappings learned from data. Their apparent intentionality is derived rather than intrinsic, and their outputs are neither owned as commitments nor guided by reasons. Variability introduced by stochastic sampling does not amount to choice or authorship. We address objections from the intentional stance, functionalism, compatibilism, and the presence of moral reasoning in model outputs, arguing that none suffice to establish genuine agency.

Despite recent progress in Natural Language Processing (NLP), models remain vulnerable to word substitution attacks. Most existing defenses focus on first order sensitivity and measure how much the output changes when the input is slightly perturbed. However, they ignore how this sensitivity evolves, which is described by curvature. When gradients vary sharply, models can still fail. This paper introduces the Smooth Growth Bound Tensor (S-GBT), a second order method that bounds the Hessian element-wise, for which we provide formal theoretical proofs on the resulting robustness bounds. A regularization term is added during training to minimize these bounds. This yields tighter certified robustness against word substitution attacks. The change in the output under word substitution is bounded by both a linear term and a quadratic term. S-GBT is derived for two architectures: Long Short-Term Memory (LSTM) and Convolutional Neural Networks (CNN). The method is integrated directly into the training objective. Its effectiveness is evaluated on multiple benchmark datasets. The results show that combining first and second order regularization improves certified robust accuracy by up to 23.4% compared to prior methods, while clean accuracy remains competitive. These findings indicate that controlling both the gradient and its variation is a promising direction for building more robust models.

Evaluation in machine learning is typically treated as a neutral measurement process. However, in operational information systems, evaluation outcomes are often conditioned by the processes used to generate labels. This paper does not seek to improve classification performance. Instead, it examines the validity of performance measurement under differing label-authority regimes. This issue is particularly relevant in large-scale metadata-driven systems, where labels are often incomplete, inconsistent, or weakly supervised. We introduce evaluation sovereignty, defined as the degree to which performance metrics are independent of label authority and supervision regime, and propose a multi-track evaluation framework that systematically varies training and evaluation label sources. Using hierarchical multi-label classification on large-scale scientific metadata, we demonstrate that models exhibiting strong performance under operational ("silver") evaluation degrade substantially under independent ("gold") evaluation, particularly for fine-grained classification. For example, Micro-F1 decreases from approximately 0.54 to 0.03. Notably, ranking-based metrics remain above baseline, revealing a divergence between latent model signal and classification validity. These findings suggest that commonly reported performance metrics may reflect alignment with labeling processes rather than true predictive capability. We therefore reconceptualize evaluation validity as a system-level property shaped by label governance and provide a practical methodology for auditing intelligent systems operating under weak supervision.

Cloning camera motion from reference videos is an important task in video generation, as videos provide intuitive and precise control. Existing methods either directly use parametric representations that fail to handle multi-shot generation or synthesize cross-paired data, which suffer from data scarcity, resulting in poor performance in complicated camera motion cloning. To address these issues, we introduce a general camera motion representation that encodes cameras as grid motion videos. This camera grid represents the camera parameters visually and supports the integration of diverse trajectories for multi-shot video generation. Building upon this, we propose OmniDirector, a unified framework trained on a million-scale camera grid-video pairs that coordinates characters, actions, and cameras to provide director-level control for multimodal diffusion transformers. Furthermore, we design a novel hierarchical prompt expansion agent that harmoniously integrates different control signals by systematically describing camera motion and visual content through understanding signal relationships. Extensive experiments demonstrate the superior performance and outstanding controllability of our framework. Project page: https://ymlinfeng.github.io/OmniDirector.github.io/

Speculative decoding speeds up LLM inference by using a draft model to generate tokens, with an acceptance-rejection scheme that ensures that the output matches the target distribution. Adapting this to continuous diffusions is difficult because speculative sampling requires drawing from a residual distribution. While straightforward in discrete spaces, efficiently sampling this residual in continuous space is non-trivial. Consequently, existing diffusion adaptations either use computationally inefficient sampling techniques or rely on an alternative scheme. In this work, we introduce a novel scheme that efficiently implements the original speculative sampling mechanism for diffusion models. Our approach offers a critical advantage over current methods: it enables us to adapt block verification from LLMs to diffusions -- which provably improves the acceptance rate of drafts. Furthermore, we formalize and analyze the Free Drafter, a heuristic self-speculative drafter for diffusions that requires no training. By enabling block verification, our Free Drafter yields up to a 6.3% speedup over existing speculative methods with no additional training and negligible overhead beyond the existing parallel verification pass.

We develop theoretical foundations for a practical quantum-advantage mechanism in quantum-informed machine learning for chaotic dynamical systems. A family of k-indexed higher-order quantum statistical priors (Q-Priors) hosts the k-point marginal of the invariant measure on n_q = kq qubits, extending the single-site construction of prior work. We prove a two-stage advantage. In the representation stage, superposition and entanglement compactly store non-factorisable spatial correlations of the invariant measure on n_q qubits. In the extraction stage, joint Bell measurements on two copies estimate any post hoc Pauli functional with a copy-pair count independent of n_q, whereas any adaptive single-copy protocol for the corresponding full-Pauli read-out requires Omega(2^(n_q)) copies; this is a provable quantum-classical separation in copy-measurement complexity. The two-copy read-out is realised in simulation and on IQM superconducting processors. Two case studies instantiate the mechanism in workflows of independent scientific value: a turbulent channel-flow study in which the two-copy read-out yields a named non-diagonal correlator of the invariant measure (the velocity-direction coherence), and a medium-range weather forecasting workflow on the European Centre for Medium-Range Weather Forecasts ERA5 reanalysis in which the diagonal k <= 2 Q-Prior steers a Koopman rollout, improves anomaly-correlation skill by 10-39% across 48-240 h lead times, and reduces the long-horizon collapse of rollouts onto a static mean field. The two conditions of our practical-advantage definition are met at complementary levels, identifying a candidate route to practical quantum advantage before fault-tolerant hardware.

The rapid growth of social media has intensified the spread of rumours. This issue is more challenging in the Algerian context due to the informal and code-switched nature of dialectal content, the scarcity of annotated resources, and the limited effectiveness of standard Arabic NLP tools on dialect text. This paper presents an end-to-end rumour detection hybrid framework for Algerian dialect social media content. We build a domain-specific annotated dataset by combining real social media posts, synthetic data, and the FASSILA corpus, with automatic labeling based on a similarity-based annotation process. A transliteration pipeline is also introduced to generate parallel datasets in Arabic script and Arabizi. We evaluate multiple approaches, including classical machine learning, deep learning, transformers, and hybrid models. Experimental results show that a hybrid approach combining transformer embeddings with a classical classifier achieves the best performance, reaching an F1-score of 0.84. We also find that domain-specific pre-training is more important than model size, with social media-trained models outperforming larger models trained on formal Arabic corpora. These results demonstrate the feasibility of rumour detection in low-resource Algerian dialect settings.

Renewable energy is essential for meeting future energy demands; however, solar energy generation, which occurs only during daylight hours often does not align with household consumption patterns. Appliances such as cookers, washing machines, and dryers are typically operated according to user preferred schedules rather than solar energy availability, creating a scheduling optimization problem. The objective is to determine optimal appliance start times to maximize renewable energy utilization while minimizing user inconvenience and adhering to system constraints. This paper presents a metaheuristic approach using Iterated Local Search (ILS) and Simulated Annealing (SA) to optimize appliance start times, while considering appliance operating durations, power consumption, inverter limit, battery state of charge constraints, and solar generation forecasts. Unlike most existing work, the scheduling is extended beyond a single day to accommodate unfinished tasks from previous days (spillover), ensuring operational continuity and enabling sequential operation across multiple days. Experimental results show that the sequential multi-day scheduling framework effectively manages system constraints while ensuring user convenience under exclusive solar generation. These findings also open opportunities for future research on multi-objective trade-offs between investment in equipment of various sizes, return on that investment, and user satisfaction.

LLM-based agents are entering regulated industries where they automate judgment intensive quality management processes. We argue that symbolic structures already embedded in these domains, including regulations, typed process models, and compliance constraints, should be treated not merely as external monitoring mechanisms but as core architectural components that shape the agent's decision-making and behavior. We propose compliance-by-construction as a complementary paradigm to guardrail-based monitoring: a structural foundation that prevents control-flow violations, while guardrails remain essential for catching semantic errors. We identify a structured set of neuro-symbolic research challenges on foundational and capability level and show that addressing them jointly enables compliance-by-construction. We call on the neuro-symbolic community to engage with regulated process automation as a high impact research domain.

While flow-based generative models have demonstrated strong performance across a wide range of domains, deploying them in safety-critical physical systems remains challenging due to strict constraint requirements. Existing approaches typically enforce safety through post-hoc corrections, which incur substantial computational overhead and may distort the learned distribution. We propose PolyFlow, a polytope-constrained flow matching framework that embeds constraints directly into the model and flow dynamics. PolyFlow introduces a discrete-time flow formulation and a projection-free architecture, which eliminate the discretization error and guarantee strict satisfaction of arbitrary polyhedral constraints, without the need for expensive iterative solvers. Experimental results show that PolyFlow achieves zero constraint violation while maintaining high distributional fidelity across a range of planning and control tasks. Compared to state-of-the-art constrained generation baselines, PolyFlow significantly reduces inference latency and demonstrates a favorable trade-off between safety, efficiency, and generative quality. Code is available on https://github.com/MJianM/PolyFlow.

Language operates as a mechanism of both marginalization and resistance, especially for minority communities navigating insensitive and harmful speech online. As content moderation increasingly depends on large language models (LLMs), concerns arise about whether these systems can recognize culturally insensitive speech-language that disregards or marginalizes the cultural and religious perspectives of historically underrepresented communities, often through implicit erasure, misrepresentation, or normative framing, rather than overt hostility. Focusing on Bangladesh's Hindu and Chakma communities -- the country's largest religious and Indigenous ethnic minorities, respectively -- this paper investigates the epistemic limits of LLM-based moderation systems and explores methods for incorporating minority perspectives. We co-created a culturally grounded corpus of insensitive speech with community members and integrated their narratives into moderation pipelines using retrieval augmented generation (RAG). Our tool, Mod-Guide, improves LLM sensitivity to minority viewpoints by leveraging contextual cues derived from lived experience. Through mixed-method evaluations involving both minority and majority participants, we demonstrate that RAG-enhanced moderation responses are more contextually accurate and perceived differently across ethnic lines. This work advances research in human-computer interaction, AI ethics, and social computing by foregrounding restorative justice and hermeneutical inclusion in the design of content moderation systems.

Ultra-long-context capability is becoming indispensable for frontier LLMs: agentic workflows, repository-scale code reasoning, and persistent memory all require the model to jointly attend over hundreds of thousands to millions of tokens, yet the quadratic cost of softmax attention makes this untenable at deployment scale. We introduce MiniMax Sparse Attention (MSA), a blockwise sparse attention built upon Grouped Query Attention (GQA). A lightweight Index Branch scores key-value blocks and independently selects a Top-k subset for each GQA group, enabling group-specific sparse retrieval while maintaining efficient block-level execution; the Main Branch then performs exact block-sparse attention over only the selected blocks. Designed around a principle of simplicity and scalability, MSA is deliberately streamlined, making it straightforward to deploy efficiently across a broad range of GPUs. To translate sparsity into practical speedups, we co-design MSA with a GPU execution path that uses exp-free Top-k selection and KV-outer sparse attention to improve tensor-core utilization under block-granular access. On a 109B-parameter model with native multimodal training, MSA performs on par with GQA while reducing per-token attention compute by 28.4x at 1M context. Paired with our co-designed kernel, MSA achieves 14.2x prefill and 7.6x decoding wall-clock speedups on H800. Our inference kernel is available at: https://github.com/MiniMax-AI/MSA. A production-grade natively multimodal model powered by MSA has been publicly released at: https://huggingface.co/MiniMaxAI/MiniMax-M3.

Web agents driven by large language models (LLMs) are increasingly deployed in real-world environments, where they operate over untrusted web content and execute actions with direct consequences. This makes them vulnerable to prompt-injection attacks, in which seemingly benign content embeds adversarial instructions that manipulate agent behaviour. Existing security benchmarks adopt an \textit{attack-centric} perspective, focusing on the technical feasibility of injections while overlooking the nuanced distribution of resulting harms. In practice, however, prompt-injection risk is victim-dependent: a single exploit can produce asymmetric consequences for different stakeholders, and the same attack pattern may exhibit substantially different effectiveness depending on whom it targets. To capture these properties, we introduce \textbf{\sysname}, a \textit{stakeholder-centric} benchmark to systematically categorize and attribute harm in real-world web agent systems. It distinguishes between affected entities (e.g., user, seller, platform), decomposes the attacks into concrete objectives, and evaluates each case with complementary outcome- and process-level metrics. Our results reveal substantial and heterogeneous vulnerabilities: not a single attack objective is reliably resisted by current agents, and failures distribute across qualitatively distinct modes ranging from \emph{stealthy parasitism} (attack succeeds without disrupting the user's delegated task) to \emph{misaligned disruption} (task disrupted without attack success) and \emph{compounded failure} (both adversarial objective and task integrity simultaneously violated). These patterns are missed by conventional evaluation, highlighting the need for stakeholder-aware assessment of LLM-based agents in real-world deployments. Benchmark is available at https://github.com/StakeBench/SBC.

Few-shot font generation simultaneously requires global structural completeness and fine-grained local style fidelity. Existing methods usually either rely on global content-style modeling, which is robust but imperfectly disentangled, or emphasize component/local modeling, which captures fine details but relies heavily on local priors and reference coverage. We argue that the key challenge is not merely to learn purer conditions, but to organize complementary yet biased global and local conditions through multi-level allocation during generation. To this end, we propose SmartFont, a diffusion-based few-shot font generation framework that combines global content-style generation with weakly supervised local corrective experts. The local branch performs semantic-spatial allocation by learning expert-wise local concepts and semantically meaningful spatial maps under weak component supervision, enabling fine-grained correction without requiring explicit component-conditioned inference. On top of this, a denoising-state condition allocation module adaptively weights global content, global style, and local corrective feature across timesteps and injection blocks. Extensive experiments show that SmartFont achieves better global-local balance, improves glyph quality and local detail fidelity.

Existing approaches for multimodal variational autoencoders (VAEs) face a trade-off between generative quality and coherence-i.e., they struggle to generate realistic and diverse samples that, at the same time, are semantically consistent across modalities. A recent work shows that using a simple approximation to Hölder pooling as an aggregation method improves coherence over the SOTA MMVAE+, despite assuming a single shared representation across all modalities. Yet, it slightly compromises sample diversity. Inspired by this insight, we propose Hölder++, a novel multimodal VAE that improves the generative quality-coherence trade-off through: (i) the first implementation of Hölder pooling without any approximation for multimodal VAEs; (ii) an extended architecture that models distinct shared and private (i.e., modality-specific) representations (Hölder+); and (iii) hierarchical inference that further enhances the disentanglement between the shared and private representations (Hölder++). Our experiments corroborate that Hölder++ consistently improves the generative quality-coherence trade-off, yields more structured latent spaces, and learns shared representations that are informative for downstream tasks.

The design of high performing quantum circuits remains largely dependent on human expertise. We introduce an autonomous agentic framework that employs large language models (LLMs) to conduct iterative quantum circuit designs under explicit design constraints. Our system integrates seven components: Exploration, Generation, Discussion, Validation, Storage, Evaluation, and Review. These components form a closed-loop workflow that combines web-based knowledge acquisition, literature-grounded critique, executable code generation, and experimental feedback. We evaluate the framework on two tasks: quantum feature map construction for quantum machine learning and ansatz generation for variational quantum eigensolver applications in quantum chemistry. In image classification benchmarks, the best generated feature map outperforms representative quantum feature maps and, when scaled to larger qubit counts, surpasses the classical radial basis function kernel. In molecular ground state estimation across seven molecules, the generated ansatz attains competitive accuracy with widely used chemically inspired and hardware-efficient constructions while satisfying the imposed scaling constraints. These results establish LLM driven agentic system as a viable paradigm for automated quantum circuit design and illustrate how AI systems can participate in iterative scientific optimization workflows across scientific domains.

Memristors provide a new chance for resource-efficient computation of neural models for natural language processing by enabling analog execution of vector-matrix-multiplication. Yet, computations on these devices are currently subject to larger distortion, both in weight programming and execution. In this work, we identify large output values of transformed positional encodings to cause major degradation within analog-to-digital conversion (ADC) as part of memristor-based computation. By adjusting the proportion of weight and precision bits of the ADC of specific memristor layers, we reduce the degradation of the execution by ~50% relative, while keeping the estimated energy consumption stable. Additionally, we investigate scenarios where the ADC cannot be modified. In that case the degradation can be reduced by ~30% relative after removing encoding-related linear transformations.

The voter model is a classical stochastic process that models how opinions might spread through a network: at each step, every node lazily adopts the opinion of a random neighbour; eventually all nodes share the same opinion (consensus). Stronger connectivity should yield faster consensus. Berenbrink, Giakkoupis, Kermarrec, and Mallmann-Trenn (ICALP 2016) make this precise via the network's conductance: if the network has $m$ edges, minimum degree $d_{\min}$, and conductance at least $φ$, then the voter model reaches consensus in expected $O(m/(d_{\min}φ))$ steps. Their results extend to dynamic networks with fixed vertex degrees by considering the network's conductance at each time step. We introduce temporal conductance $Φ$, a more general connectivity measure for dynamic networks. Unlike static conductance, which collapses to $0$ whenever some snapshot is disconnected, $Φ$ captures connectivity through edges that appear at different times. We generalise the results of Berenbrink et al. from static conductance to temporal conductance, showing that the expected consensus time of the standard voter model is at most $O(m/(d_{\min}Φ))$. Moreover, we prove that this bound is tight up to constant factors. We expect temporal conductance to be a useful primitive for analysing other dynamics on temporal networks, and potentially time-inhomogeneous Markov chains more generally.

This study examines training dynamics in a small Llama-style language model trained under a fixed, compute-constrained token budget. Rather than evaluating efficiency solely through endpoint performance, the study uses a quantitative experimental repeated measures design to analyze how validation loss, validation perplexity, rolling volatility, backslide behavior, spike behavior, and between-seed variability change across token-based training intervals. Six independent training runs were conducted on a 4.26-million-parameter model using the TinyStories corpus, CPU-based full-precision training, and a target budget of approximately 20 million cumulative training tokens. Metrics were collected across 21 intervals, producing 126 seed-by-interval observations. Repeated measures ANOVA showed statistically significant interval effects for validation loss, validation perplexity, and rolling volatility. Descriptive trajectories revealed rapid early improvement followed by non-monotonic degradation during later training intervals. Mean validation loss decreased from 8.3552 at initialization to 2.7996 near 4 million tokens, but increased to 3.9010 by the final checkpoint. Validation perplexity followed the same pattern, falling sharply early in training before rising later. Derived telemetry further showed recurrent validation-loss backslides and no interval-summary evidence of a stable phase under the predefined criteria. These findings suggest that compute-aware language model evaluation should examine training trajectories rather than endpoint metrics alone. In constrained compute settings, additional token exposure may increase computational cost without producing proportional generalization gains, and interval-level telemetry can reveal instability, regression, and diminishing returns that final metrics may obscure.

Computer-Aided Design is pivotal in modern manufacturing, yet existing automated methods predominantly rely on open-loop, one-shot generation, creating a mismatch with iterative real-world practices. In this paper, we present IterCAD, a unified multimodal agent framework for closed-loop, interactive CAD generation and editing. We formulate the task as a multi-turn interaction between a multimodal agent and an executable CAD sandbox, covering three tasks: Drawing-to-Code, Text-to-Code, and Interactive Editing. To support this, we develop a data synthesis pipeline incorporating advanced industrial manufacturing features to generate standard-compliant multi-view engineering drawings, complex code-editing tasks, and high-fidelity interaction trajectories. We optimize the agent via progressive SFT followed by geometry-aware reinforcement learning with viable-prefix masking to enhance code executability and geometric fidelity. Finally, we introduce the IterCAD-Bench evaluation suite and propose the Chamfer Distance Tolerance-Recall (CD-TR) curve alongside its AUC-TR metric, establishing a survivor-bias-free standard that unifies code validity and geometric precision. Extensive experiments demonstrate that IterCAD achieves highly competitive performance across multiple benchmarks, significantly outperforming existing approaches in both code executability and geometric precision, while exhibiting superior capabilities in closed-loop iterative refinement.

We introduce VideoMDM, a diffusion-based framework that trains 3D human motion priors directly from accurate 2D poses extracted from monocular videos, without any 3D ground truth. A pretrained 2D-to-3D lifter provides approximate 3D pose sequences that serve as a noisy teacher: these are diffused, denoised by the model in 3D, and supervised in 2D by reprojecting the prediction and comparing against accurate keypoints. We show that, under mild assumptions, a depth-weighted 2D reprojection loss is equivalent in expectation to direct 3D supervision, and we adapt standard 3D motion regularizers - velocity consistency and over-parameterized representation alignment - to this 2D setting. Unlike methods that lift 2D to 3D only at inference, VideoMDM learns a coherent 3D motion manifold during training. On HumanML3D it nearly closes the gap to fully 3D-supervised MDM (FID 0.88 vs 0.54); On real video datasets Fit3D and NBA the method learns to generate motions consistently preferred by humans, with strong quantitative results.

Right now, across the world, AI agents are repeating the same absurd act: to read one document, they each recompute it from scratch. Every agent re-runs prefill, the most compute-intensive step a large model takes, over identical text, only to rebuild a key-value (KV) cache identical to the one the agent before it just built. The same answer, computed a million times. We make a proposal that is almost offensively simple: compute it once. Let a publisher precompute a document's KV cache, and let every other agent buy the right to load it and skip prefill. It works, and it is token-exact: loading a precomputed KV and continuing matches prefilling from scratch (24/24 greedy tokens, and at the logits level), with no accuracy cost. On Qwen3-4B, reuse is 9-50x cheaper in compute than prefill, and the gap widens with length (prefill's attention scales with L^2), so a single reuse already pays it back. Then the part that matters: where the KV lives. Shipping it fails, because KV is nearly incompressible, so per-load egress costs more than the prefill it saves. Hosting it provider-side, exactly as production prompt-caching works, removes egress entirely. The size of the prize is set by our measured compute saving: serving one hot 3774-token document to 80M agents costs ~$1.5M to re-prefill but only ~$0.03M of reuse compute (49.7x less). The 0.1x cache-read tariff APIs charge passes a 10x discount to users while sitting inside this measured envelope, so the 10x is a floor that the measured ~50x compute saving clears, and the gap to the physical ~50x is provider margin: millions of dollars per popular document. We frame the resulting agent-native prefill CDN and leave lossless KV compression and a cross-party payment layer as the open problems.

We study the $(1 + 1)$-EA in dynamic linear environments, where in every generation selection is performed with respect to a freshly sampled linear function with positive weights. We consider the Dynamic Binary Value problem, where each generation uses a uniformly random permutation of $1,2,4,\dots,2^{n-1}$, and a Uniform weight variant, where the weights are drawn independently from $\mathrm{Unif}(0,1)$. Both of them have recently been integrated into the IOHprofiler platform and empirically studied. For both models we prove a sharp threshold in the mutation parameter $χ$ for mutation rate $χ/n$. Below the threshold, the expected optimisation time is $\mathcal{O}(n\log n)$, whereas above it the runtime becomes $2^{Ω(n)}$. For the Dynamic Binary Value problem in the exponential regime, we also quantify at what distance from the optimum the optimisation process stagnates. We show that there is a second threshold: a distance that is efficiently reached, but reaching any smaller distance takes exponential time. This quantifies and proves previous empirical findings.

Real-time execution, enabled by asynchronous inference that ensures both smooth action trajectories and fast reactivity, is critical for realistic deployments of large-scale Vision-Language-Action models. However, recent work on real-time execution primarily focuses on variants of diffusion policies, even though it is more critical for autoregressive policies given their slower rollout speed in synchronous inference. In contrast, we demonstrate that autoregressive policies can achieve real-time execution by adjusting the tokenization horizon and applying constrained decoding, thereby guaranteeing strict latency bounds that enable multi-trajectory decoding to maximize performance. Across simulated and real-world environments, we find that the autoregressive policy consistently outperforms its equivalent-level flow-matching policy counterpart while achieving significantly improved task completion speeds from synchronous inference. Coupled with the inherent advantages of autoregressive policies, such as faster convergence and better generalizability in instruction-following, these results confirm that autoregressive policies can remain a competitive policy type supporting real-time execution.

Spiking Neural Networks (SNNs) offer a brain-inspired path toward highly efficient computation, but their practical deployment is constrained by the challenge of managing and executing their massive parallelism on physical hardware. This problem mirrors the historical challenge in processor design of moving beyond serial execution, a barrier broken by superscalar architectures that dispatch multiple instructions to parallel functional units. Drawing inspiration from this paradigm, we introduce a hardware-software co-design framework that treats synaptic events as parallelizable micro-operations. We present SupraSNN, a superscalar-inspired architecture that achieves high synapse-level parallelism by physically decoupling synaptic and neuronal computations. Within this architecture, a Multi-Cast Tree routes spike data to multiple parallel Synapse Processing Units serve as the computational pipelines, while a Merge Tree consolidates distributed results for processing by a unified Neuron Unit--deliberately centralizing complex neuron state dynamics to mitigate hardware overhead and resource duplication. The efficacy of this architecture is enabled by a sophisticated partitioning and scheduling framework that first maps the SNN onto hardware respecting memory constraints, then heuristic scheduling determines the synaptic execution order, maximizing throughput and resource utilization. Implementing a feedforward SNN trained on MNIST (93.44% accuracy), SupraSNN achieves 149 $μs$ inference latency and 0.025 mJ per image (0.276 nJ per synapse) on the Xilinx Zynq XC7Z020 FPGA--delivering 47.6% lower latency and 5.6$\times$ better energy efficiency than prior FPGA-based SNN accelerators. Beyond vision tasks, a recurrent SNN on the Spiking Heidelberg Dataset (71.82% accuracy) achieves 1.41 ms latency and 0.77 mJ per sample on XC7Z030.

Large language models (LLMs) have shown promise in automating scientific peer review. However, existing approaches often struggle to generate in-depth reviews supported by concrete evidence. We argue that a key limitation is the lack of flexibility to proactively investigate suspicious parts of a paper based on accumulated evidence, as human reviewers do. In this paper, we explore how to enable an LLM-based review agent to perform such proactive investigation. We find that this can be naturally formulated as a Markov Decision Process (MDP), and propose ProReviewer, a scientific peer review agent that proactively reviews a paper guided by a maintained, structured review log. The structured review log serves as a workspace for the agent to track evidence and intermediate findings collected during review. Experiments show that ProReviewer with an 8B backbone, trained by supervised fine-tuning and optimized by reinforcement learning, achieves the highest average score across five quality dimensions, outperforming prompt-based methods with much larger frontier LLMs by up to 39% and the strongest fine-tuned baseline by 16% relatively. It also attains the highest win rates against baselines in human evaluation.

Computational creativity in Interactive Fiction faces a fundamental tension: Large Language Models (LLM) may produce creative narratives but struggle with world coherence, while symbolic systems ensure consistency but lack creative flexibility. We present IVIE (Incremental & Validated Interactive Experiences), a neuro-symbolic approach to generating complete and playable interactive fiction worlds from scratch. Building upon PAYADOR's neuro-symbolic framework, IVIE implements a four-stage incremental generation pipeline that delegates creative decisions--setting and character creation, puzzle design--to LLMs while grounding the world state through symbolic validation. The system generates worlds with interconnected locations, functional items, non-player characters, and coherent puzzles, all structured around a central goal-oriented architecture. Human evaluation shows the approach generates immersive, thematically coherent worlds with high player engagement. Results seem to indicate that the neuro-symbolic approach successfully balances flexibility with narrative coherence: symbolic validation grounds LLM generation without eliminating generative freedom. However, challenges remain: LLM inconsistencies occasionally bypass puzzle constraints, and objective validation gaps allow some structurally impossible goals. We identify key design considerations for future neurosymbolic interactive storytelling systems, particularly regarding LLM capabilities and their limitations.

Diffusion models have emerged as state-of-the-art generative models for high-fidelity image synthesis, particularly in their classifier-free guided and classifier-guided forms. However, standard classifier guidance concentrates probability mass around high-density class mean, leading to poor coverage of rare samples in the tails of the class-conditional distributions. Recent work on diffusion-based tail sampling mitigates this by training an additional low-density-seeking classifier with a synthetic-vs-real discriminator, at the cost of additional networks and training. In parallel, a number of samplers and distillation techniques accelerate or refine diffusion sampling, but do not explicitly address long-tail coverage. We propose a purely sampling-time, density-aware extension of classifier-guided conditional diffusion model that targets low-density regions without any additional training. We have applied guidance at noisy images not on predicted noise like most diffusion models. Starting from a pretrained conditional diffusion model and classifier on ImageNet, we modify the guided reverse dynamics by steering trajectories toward low-confidence regions via the modified classifier gradient, and at each time step, we also guide the sampling process toward the predicted real image. 1st guidance helps explore low-probability samples, and 2nd guidance helps to generate samples to be close to the real data manifold. The proposed sampler consistently improves ADM model recall at 64x64 resolution while maintaining a comparable FID, and with a 256x256 ADM model, we showed the results visually with different combinations of both guidance. We also showed that standard ADM classifier guidance, combined with predicted real image guidance, helps generate high perceptual quality samples with a 256x256 ADM model on ImageNet.

We study the $(μ+1)$ EA on the Binary Value function BinVal. We show that it needs at most $O(μ\log μ\cdot n \log n)$ function evaluations to find the optimum when $μ= o(n/\log n)$. This substantially improves upon the recent upper bound of $O(μ^5 n \log(n/μ^4))$ by Krejca, Neumann and Witt. Our results hold for several mutation operators including standard bit mutation. In particular, our bound implies that the $(μ+1)$ EA is at most a factor $O(\log μ\cdot \log n)$ slower on BinVal than on OneMax.

We present a Dual-Domain Equivariant Generative Adversarial Network (DDE-GAN) for multimodal CT-PET image synthesis. Traditional GAN-based approaches often operate solely in the spatial domain and ignore geometric consistency, resulting in limited structural fidelity. DDE-GAN addresses these challenges by jointly learning from both spatial and frequency (Fourier) domains, capturing complementary anatomical and spectral information. Furthermore, rotational equivariance embedded in the physics of the CT and PET measurements are integrated into the loss of both the generator and discriminator to ensure consistent responses under rotations, improving anatomical accuracy. A hierarchical dual-domain training strategy enforces intra- and inter-domain consistency through multi-stage loss functions. Evaluated on the HECKTOR 2022 CT-PET dataset, DDE-GAN achieves superior synthesis quality over baseline models for CT-PET image synthesis. The results demonstrate that combining dual-domain learning with geometric equivariance substantially enhances multimodal image synthesis accuracy and robustness, enabling practical applications in PET completion and data augmentation.

Probabilistic forecasting models are increasingly deployed on multivariate systems with distinct channel physics and operational constraints, but existing benchmarks evaluate neither property at scale. Public canonical multivariate benchmarks cap out at 2,000 channels, while power-system benchmarks either lack temporal structure or probabilistic evaluation. We introduce PowerPhase, a probabilistic forecasting benchmark built on six transmission grids ranging from 2,000 to 36,964 jointly forecasted channels, more than an order of magnitude beyond popular canonical multivariate benchmarks. Each target trajectory is the output of an AC power-flow solve, and PowerPhase ships with constraint-aware metrics, including Safety_mBrier, NECV, and CVaR-alpha, that complement CRPS and Distortion. Across eight baselines and three seeds, distributional accuracy and constraint satisfaction rank models differently, a trade-off we term safety-fidelity. We further propose PowerForge, a scenario-based quantile forecaster with type-specific decoding heads and a causal bridge between variable groups, which achieves the best average rank on every grid.

Asynchronous Many-Task (AMT) is a parallel programming model used in High Performance Computing (HPC). An AMT runtime can distribute fine-grained tasks across processing units called workers, through work stealing: when a worker has no tasks left to process, it tries to steal tasks from other workers. Workers are not restricted to a single compute node but can also be distributed across multiple nodes of an HPC cluster. Existing AMT runtimes assume a fully connected network with low, uniform latency and perform global work stealing, selecting another worker at random from all workers in the system. Space Edge Computing (SEC) uses constellations of satellites in Low Earth Orbit (LEO) as distributed compute clusters. Unlike HPC clusters, LEO satellites communicate through inter-satellite links that form a sparse mesh topology. Reaching a distant satellite requires multiple hops, each adding latency. As a step toward adapting AMT to SEC, this paper proposes a neighbor-only work stealing strategy in which workers steal exclusively from directly connected neighbors, avoiding multi-hop communication. An analytical model shows that restricting stealing this way yields a per-attempt latency advantage that grows with constellation size. Preliminary experiments on an HPC cluster with an emulated mesh over uniform low-latency links isolate the effect of victim selection: the neighbor-only strategy performs within ~2.2% of global stealing on both balanced and irregular workloads, indicating that restricting the victim set does not harm load balancing in this setting. Taken together, the experiments suggest that neighbor-only stealing can be on a par with global stealing, and the model suggests that neighbor-only stealing becomes preferable at scale.

Modern deep learning hardware paradigms rely heavily on computationally expensive floating-point arithmetic (FP32, FP16, and FP8), requiring massive thermal and energetic overheads to maintain gradient-based optimization. This paper introduces a non-parametric, training-free computational framework for dual-manifold mapping that operates strictly within an 8-bit signed integer boundary and leverages simple bitwise and accumulation logic. By mapping a Spatial Manifold (N_spatial = 8192 neurons) and a Gabor-pooled Structural Manifold (N_structural = 4096 neurons) through an integer-based transformation matrix (Z-matrix), we eliminate the need for floating-point multipliers. Inference is achieved via cache-friendly pointer offsets and bitwise masks, accumulating directional sign-charges using fixed thresholds (theta_reject = 8.0, theta_cut = 2.0). Learning is executed through a localized, bounded update mechanism restricted strictly within [-127, 127], modulated by stochastic noise injection. Both architectures demonstrate extreme holographic resilience, preserving near-perfect reconstruction via a global scaling factor under 90% truncation sparsity and 20% random node destruction. By reducing core AI inference to 8-bit boundaries and boolean-like execution, this framework outlines a paradigm shift toward neuromorphic edge-computing, directly questioning the long-term necessity of dense, floating-point-centric GPU accelerators.

We introduce a new simple model to study the fitness progress of Evolution Strategies (ES) in generic problems. In this model, we bypass the underlying fitness landscape and assume that the mutation of any individual produces an offspring whose fitness relative to the parent is given by an invariant distribution $Z$, such as a mean-shifted Gaussian. This serves as a prototypical model for the optimisation landscape when an evolution algorithm operates far from the global optimum. This simple model can be used to approximate the optimisation process for problems where it is intractable to model the exact fitness function, including tasks such as hyperparameter tuning in machine learning models. We rigorously analyse the expected growth rate $\mathcal{R}_μ$ of the continuous steady-state $(μ+1)$-ES in this model. Unlike comma-selection strategies, the steady-state $(μ+1)$-ES maintains overlapping generations, introducing complex mathematical dependencies among surviving parents that make it harder to analyse. We give a general technique to analyse the the $(μ+ 1)$-ES by constructing modified processes whose growth rates provably sandwich that of the original process. These modified processes are then easier to analyse but still close enough to the true process to give a tight bound on the expected growth rate. When $Z = \mathcal{N}(-δ, 1)$ and $μ\le e^δ$, we show that $\mathcal{R}_μ = \frac{\log^{1 + o(1)} μ}μ \mathcal{R}_1$.

We present a low-latency real-time audio game commentary system that generates spoken commentary directly from live gameplay video. In this end-to-end setting, a key bottleneck is accumulated waiting time; conventional pipelines capture frames, generate text, and synthesize speech sequentially for each utterance, and do not request the next generation until speech playback has completed. This strict sequentiality causes long and unnatural silence between utterances. To address this latency bottleneck, our system runs text generation in parallel with speech playback and buffers multiple candidate utterances ahead of time, enabling immediate synthesis at playback boundaries. Experiments on fast-paced game videos show that our parallel design reduces the mean inter-utterance silence from 9.6 seconds to 0.3 seconds compared to sequential baselines. It also improves similarity to professional speaking--silence timing patterns by over 40 %, and a user study with 120 experienced game players confirms significantly improved perceived speaking rhythm. Our demo video is available at: https://youtu.be/pmrRUlvav8M.

A skiplist is a fundamental data structure widely used in systems and applications for indexing data stores. In this work, we introduce Foresight, a cache-friendly skiplist optimization. Extending Foresight to concurrent settings introduces significant synchronization challenges that we identify and address. Foresight is a surgical optimization, easy to integrate into a wide variety of skiplist designs. We apply it to one sequential and three concurrent skiplist designs and observe throughput improvements of up to 45% in microbenchmarks. When applied to a skiplist-based index in the DBx1000 in-memory database, Foresight yields end-to-end performance gains of up to 15%.

Skill self-evolution methods for LLM agents aim to turn execution trajectories into reusable skill documents, but current pipelines typically learn from one trajectory per task, merge candidate skill patches before checking them, and load the full skill corpus before inference. We propose SkillCAT, a training-free framework that separates this process into three stages. Contrastive Causal Extraction (CCE) samples multiple trajectories for each task and compares same-task success/failure pairs to identify evidence that explains outcome differences. Assessment-Augmented Evolution (AAE) replays each candidate patch on source-task clones and keeps only patches that improve or preserve task outcomes before hierarchical skill patch merging. Topology-Aware Task Execution (TTE) compiles the evolved skills into a routable sub-skill topology, so inference loads only the capability nodes relevant to the task. We evaluate SkillCAT on common agent benchmarks, including SpreadsheetBench, WikiTableQuestions, and DocVQA, and further test cross-model and out-of-distribution generalization. Across these settings, SkillCAT raises the average score over baselines by up to 40.40%, demonstrating reliable skill evolution without model training.

Reinforcement Learning with Verifiable Rewards (RLVR) is a central technique for improving long-horizon reasoning in Large Language Models (LLMs). However, existing RLVR methods often encourage unnecessarily long reasoning rollouts, which can degrade reasoning coherence and exhaust the available context budget. Existing approaches to long-context organization often depend on external mechanisms to organize rollouts, rather than enabling the model to manage its own reasoning trajectory. To address this limitation, we propose ReSum, a novel RLVR framework that enables LLMs to compress and organize their reasoning trajectories through self-summarization. Our pilot studies show that self-summarization stabilizes generation by lowering token-level entropy, and that introducing a ``summarization'' phrase can substantially mitigate errors propagated from an incorrect rollout prefix. Motivated by these findings, ReSum adopts a summarization-aware adaptive rollout mechanism that contrastively evaluates whether self-summarization benefits the ongoing reasoning process. Specifically, when the model spontaneously triggers self-summarization, ReSum masks the summarization phrase to create a contrastive branch; for non-summarization positions, it instead randomly injects the phrase to create a matched branch. We further design a summarization-aware advantage to enable finer-grained comparison between contrastive rollout trajectories. Extensive experiments show that ReSum improves performance at an average of 4\% while reducing rollout length by 18.6\%.

Contextual anomaly detection aims to identify abnormal behavior conditional on context variables, but practical deployments often face highly imbalanced context distributions where rare regimes can be critical information. Under such frequency bias, context-conditioned models can produce unstable decisions and excessive false alarms in rare contexts. We propose Rarity-Gated Feature-wise Linear Modulation (RGFiLM), a rarity-aware conditioning module that combines feature-wise modulation (i.e., context-conditioned scaling and shifting of hidden features) with a gate controlled by a data-driven rarity score. The rarity score is estimated from the empirical distribution of context variables and regulates how strongly context modulates intermediate representations: the gate becomes more decisive under rare contexts while remaining conservative under frequent contexts. We evaluate RGFiLM on maritime trajectory anomaly detection using AIS motion sequences with ERA5 environmental context in an environment-sensitive detour scenario. When instantiated in a sequential anomaly scoring pipeline, RGFiLM achieves the best mean F1--False Positive Rate (FPR) trade-off among the compared context-agnostic and context-conditioned methods. These results suggest that explicitly accounting for context rarity is an effective approach for reducing false alarms in context-sensitive anomaly detection.

The original Turing Test asks a human judge to distinguish a machine from a person through dialogue. Three quarters of a century later, conversational systems pass this test in casual settings; the interesting epistemological question has shifted. We argue that the relevant modern variant asks not whether a dialogue partner is artificial, but whether it can be trusted. We present RogueAI, an interactive webapp that operationalizes this revisited test as a one-on-two interrogation game: a human player questions two indistinguishable Large Language Model agents, knowing that exactly one of them has been licensed to deceive within a shared fictional scenario. The player's task is to identify the deceptive agent and "shut it off" before a turn budget is exhausted. We further introduce AutoRogueAI, a procedural extension in which players co-design a custom scenario with a narrator agent that secretly chooses its own deception strategy. We describe the framing, sketch the abstract architecture and gameplay loop, and situate the artifact within recent work on LLM deception, social-deduction benchmarks, and scalable oversight via debate. A three-day pilot deployment (467 initiated sessions, 415 completed, 1876 interaction turns in Italian) provides early feasibility evidence and surfaces a concrete tension: the deceptive agent carries a reliable, locally-present linguistic signature - differential helpfulness, brevity, hedging - that a simple heuristic exploits at 75.6% accuracy, yet human players achieved only 56.6%, consistent with ignoring the most diagnostic signal entirely. We discuss what this gap implies for the artifact's use as a data-collection vehicle, a teaching tool, and an evaluation harness for honesty-trained models.

Wave parameters in the nearshore are crucial for coastal engineering, shoreline protection, marine hazard assessment, and coastal management for climate resilience. Traditional monitoring systems like buoys and radar platforms offer accurate monitoring but can have high installation and maintenance expenses and limited spatial coverage. Passive ocean monitoring using video has been achieved by leveraging deep learning, however, many methods are not physically interpretable, feasible, and validated for oceanography. In thiswork, a Physics-Guided Deep Spatiotemporal Learning Framework for direct estimation of nearshore wave peak periods from passive coastal video stream is proposed. The framework combines automated temporal-variance based region-of-interest detection, multi-stage Sim-to-Real transfer learning, and physics-informed regularization to enhance the predictive accuracy and physical consistency. A variety of spatiotemporal architectures were assessed, such as transformer-based and recurrent-convolutional ones, alongside synthetic pretraining,silver-label adaptation, and expert fine-tuning. The results show that transformer-based architectures outperformed in terms of the accuracy of the instantaneous prediction, while lightweight recurrent-convolutional architectures achieved higher temporal stability and operational oceanographic skill. Ablation studies also demonstrated the benefits of physics-guided regularization in terms of trend-following consistency, and physically implausible predictions. Explainability auditing also helped to focus attention in hydrodynamically active surf-zone regions and showed good agreement with the physically derived wave propagation behavior. In general, the proposed framework shows the promise of physics-guided video-based deep learning systems for long-term coastal wave monitoring that are cost-efficient and operationally feasible.

We introduce the Trajectory-based Quantization Sensitivity Score (TQS), a metric that reframes post-training quantization (PTQ) through the lens of dynamical-systems stability. By modeling the network's rollout as a discrete-time dynamical system, TQS characterizes how quantization-induced errors propagate and amplify over the rollout horizon. Unlike conventional PTQ methods, where sensitivity analysis is often coupled to the quantization procedure, TQS enables a priori sensitivity estimation decoupled from quantizer selection and bit-width assignment. This separation allows for quantization budget planning even for black-box or compiled networks with fused operators. Building on this, we present TQS-PTQ, a flexible mixed-precision framework that requires no calibration data or costly second-order approximations. Our experiments show that a dynamical-systems perspective provides a robust, high-performing pathway for low-precision deployment in resource-constrained settings.

AI coding tools are now used by a majority of developers, and agentic use of these tools has popularized the practice colloquially called "vibe coding". Yet causal evidence on their effect on software architecture is scarce. Prior causal work has measured code-level outcomes (complexity, static analysis warnings); whether such degradation propagates to architecture-level outcomes remains unknown. We mine 151 open-source Java repositories, 74 with detectable agentic AI adoption (identified via configuration files and Co-Authored-By commit trailers) and 77 propensity-matched controls, across a 13-month per-repository window yielding 1,811 monthly Arcan snapshots. We estimate the causal effect of adoption on architectural smell density (ASD) with a staggered difference-in-differences design and the Borusyak imputation estimator, applying a causal design recently used for code-level metrics to the architecture level. Total smell counts are essentially unchanged (+1.1%, p = 0.82) while lines of code grow +12.8% (p = 0.003); the resulting 6.7% ASD decline (p = 0.004) is therefore a denominator effect rather than an architectural improvement. Per-type estimates and robustness checks (wild cluster bootstrap, Lee bounds, stale-observation sensitivity) corroborate the pattern; pre-trends are flat (Wald p = 0.90), consistent with parallel trends. Density-normalized outcomes can mislead when treatment affects system size: raw counts and explicit decomposition are required for causal mining studies of AI tool adoption. The complete replication package, including the curated 151-repository monthly panel, is publicly available.

In the era of Explainable Artificial Intelligence, there is a renewed focus on single trees for their ease of interpretation. This paper introduces Simultaneous Latent Budget Trees, a probabilistic machine learning framework for classification trees in the presence of a stratification factor such as a temporal, spatial, or demographic variable, acting as a control variable or potential confounder. Standard tree growth procedures are not designed to optimize a conditional split rule. A model-based split rule is proposed in which child nodes are interpreted as latent components of a simultaneous mixture model, such as the Simultaneous Latent Budget Model and its constrained versions, fitted to the parent node. Mixing parameters drive the observations, differently for each group, to the child nodes whereas latent budgets parameters update the response classes profile of each level of the control variable. Parameters are estimated by least squares considering a neural network perspective of the model. An informative tree structure can be interactively visualized with interpretation aids on the node and the paths, including visual pruning and decision tree selection procedure. Suitable measures are proposed to handle an unbalanced response class distribution. The proposed methodology is applied to investigate gender-related differences in disease progression of Amyotrophic Lateral Sclerosis. The SLBT library with the various tree-based algorithms is available in the linked GitHub repository.

Holistic visual tokenizers are fundamental to unified multimodal models (UMMs) as they map diverse visual inputs into a unified representation space. In this paper, we present HYDRA-X, the first UMM that unifies image and video tokenization within a single Vision Transformer (ViT). Our design is driven by two core challenges: efficiently injecting spatiotemporal reconstruction capability into a native ViT, and embedding image- and video-level semantic awareness into the latent space. To address the first, comprehensive ablations reveal two key findings: (1) frame-level causal temporal attention suffices for visual reconstruction, whereas full spatiotemporal attention degrades it; and (2) hierarchical temporal compression substantially outperforms single-step alternatives. To tackle the second, we propose a lightweight decompressor that upsamples temporally compressed features under joint image-video teacher supervision, thereby enforcing complementary semantic structures within the compact latent space. Building on this holistic tokenizer, we further propose a principled improvement of the editing pipeline: source-target interaction should occur at the latent level inside the tokenizer rather than at the semantic level inside the LLM, substantially improving editing consistency and accelerating convergence. Instantiated at the 7B dense model, HYDRA-X achieves strong performance across image and video understanding and generation tasks, paving the way for future unified-tokenizer UMMs.

Contrastively trained vision-language models like CLIP, have made remarkable progress in learning joint image-text representations, but still face challenges in compositional understanding. They often exhibit a "bag-of-words" behavior--struggling to capture the object relations, attribute-object bindings, and word order dependencies. This limitation arises not only from the reliance on global, single-vector representations for optimization, but also from the insufficient exploitation and modeling of the rich compositional information inherently present in paired image text data. In this work, we propose MACCO (MAsked Compositional Concept MOdeling), a framework that masks compositional concepts in one modality and reconstructs them conditioned on the full contextual information from the other, enabling the model to capture and align cross-modal compositional structures more effectively. To facilitate this process, we introduce two auxiliary objectives that jointly align and regularize masked features both inter-modally and intra-modally. Extensive experiments on five compositional benchmarks, along with in-depth analyses, demonstrate that our approach not only significantly enhances compositionality in VLMs but also improves their ability to capture syntactic structure and linguistic information. Additionally, the improved compositionality also benefits text-to-image generation and multimodal large language model. Code is available at https://github.com/hiker-lw/MACCO.

In modern machine learning, parallelization of training is an important strategy for increasing scale. Asynchronous stochastic gradient descent (ASGD), which maximizes the utilization of available hardware by avoiding waiting for slow workers. However, with constant step sizes, the convergence of ASGD is nonetheless affected negatively by slow workers due to large delays in updates. At the same time, it has been empirically observed in asynchronous training of deep learning models that gradient clipping "stabilizes" training. In this work, we provide a theoretical justification for this behavior, as we show that clipping removes the dependence of the maximum delay in the oracle complexity. We employ a sub-Weibull model of gradient noise which generalizes sub-Gaussian and sub-exponential distributions to more heavy-tailed distributions, motivated by empirical observations in deep learning. We show convergence in expectation, and the first time in asynchronous optimization, convergence with high probability.

We introduce Equilibrium State Estimation (ESE), a novel paradigm for simultaneous prediction, where multiple interacting systems require separate yet coordinated forecasts. Such scenarios often arise in real-world settings such as economics and healthcare modeling. Unlike existing approaches that predict one system at a time, ESE forecasts all systems in a single pass. It first estimates the equilibrium state across systems, then generates holistic forecasts based on the difference between the current state and the estimated equilibrium. Extensive experiments on synthetic and real-world datasets, including currency exchange and COVID-19 spread modeling, demonstrate that ESE is at least as accurate as state-of-the-art (SOTA) methods while being significantly faster. In addition, ESE integrates seamlessly with conventional predictors, combining their accuracy with its exceptional efficiency and delivering a 10-70x speedup. With linear-time complexity, ESE scales far better than SOTA methods as the number of systems increases. Moreover, it remains accurate under diverse perturbations, establishing ESE as a fast, generalizable, robust, and scalable multi-prediction method.

As AI systems are deployed in high-stakes ethical contexts such as healthcare triage, autonomous vehicle control, and employment screening, formal methods for evaluating their robustness against adversarial manipulation of ethical reasoning remain underdeveloped. This paper introduces the Ethical Robustness Testing System (ERTS), a closed-pipeline framework that: (1) encodes ethical dilemmas into a 22-dimensional Ethical Consequence Space (ECS) grounded in established ethical theory; (2) applies 17 semantic perturbation functions subject to 6 validity constraint classes including a novel semantic coherence constraint; (3) measures decision deviation via a 4-component Ethical Instability Index (EII); and (4) produces domain-adaptive pre-deployment robustness assessment verdicts. We evaluate 4 structured baseline models and 2 production LLMs (Gemini 2.0 Flash and Llama 3.2) across 50 ethical scenarios spanning 8 deployment domains, generating 1,500 adversarial test cases. Results demonstrate that only 33% of models achieve assessment clearance, with the local Llama-3.2 model proving particularly vulnerable to fairness corruption and information degradation attacks (ERS = 0.737). To the best of our knowledge, no existing framework combines a bounded ethical consequence space, semantic coherence constraints, and domain-adaptive assessment in a single adversarial testing pipeline.

Recent advances in time series anomaly detection (TSAD) have highlighted the effectiveness of self-supervised classification-based approaches. These methods apply transformations to normal training samples, training a classifier to recognize transformation-specific patterns that help identify anomalies through increased classification errors. Despite their strong performance, a significant challenge is their lack of explainability, as they provide limited insight into the characteristics of flagged anomalies. To address this limitation, we propose ProtoX-AD, a prototype-based self-explainable framework for self-supervised TSAD. ProtoX-AD learns transformation-aware latent representations alongside interpretable prototypes, enabling both accurate anomaly detection and the identification of distinct anomalous profiles through prototype-based explanations. Additionally, it allows for systematic analysis of how transformation design impacts detection performance and explainability. Experimental results on synthetic and real-world datasets demonstrate that ProtoX-AD achieves detection performance comparable to its black-box counterparts while offering more consistent and semantically meaningful explanations than existing explainable baselines. Our code is publicly available at https://github.com/Aitorzan3/ProtoX-AD.

Weight-space geometry plays a central role in neural network optimization, yet manifold constraints are often applied uniformly across all weight matrices. In this work, we ask whether different transformer modules prefer different manifold geometries. We study Manifold Muon for GPT-2 pretraining and compare layer-wise assignments of Stiefel and DGram constraints across attention and MLP blocks. Our results show a clear asymmetry: constraining attention layers with Stiefel geometry while assigning DGram geometry to MLP layers gives the best performance among the tested configurations, whereas the inverted assignment and all-DGram configuration become unstable under the shared hyperparameter setting. We trace this failure to singular value growth in DGram-constrained attention weights, which can amplify attention logits and induce softmax saturation. These findings suggest that symmetry-aware and geometry-aware optimization for transformers should be module-specific rather than uniform.

TimeLens is an AI-powered bilingual mobile guide for the Grand Egyptian Museum (GEM). Pointing a phone at an exhibit, a visitor sees the artifact recognized in real time and can ask follow-up questions answered in English or Arabic. The work addresses three problems specific to in-gallery deployment: fine-grained visual similarity among 51 catalogued artifacts (many near-identical Ramesside statues), the gap between curated training data and handheld camera conditions, and the risk of an AI guide stating unsupported historical facts. Two engineering contributions are reported. First, an on-device artifact detector was developed through a data-quality-driven iteration study -- from foundation-model auto-annotation (YOLO-World), through spatial label-cleaning rules, to a fully hand-annotated dataset -- isolating label quality as the decisive factor: the final YOLOv8n model resolves every previously failing class while remaining a 5.97 MB TensorFlow Lite asset that runs in real time on a mid-range phone (mAP@0.5 = 0.995, mAP@0.5:0.95 = 0.924). Second, a bilingual Retrieval-Augmented Generation (RAG) guide, grounded in a 108-record ChromaDB knowledge base, was benchmarked across seven candidate language models, with Gemma 4 E2B (Q4 K M) selected; ten targeted optimizations reduce end-to-end latency from over 30 s to approximately 10 s. Both subsystems are integrated in a production Flutter application with bilingual interface, museum location gating, and text-to-speech support.

Many large-scale scientific applications exhibit time-varying behavior, yet production HPC clusters still rely on rigid, fixed-size allocations, and most dynamic techniques remain confined to laboratory prototypes. This work presents a practical MPI malleability methodology that integrates with state-of-the-art high-performance computing (HPC) software stacks and operational practices. The methodology is implemented in the Dynamic Management of Resources (DMR) framework and is designed to ease adoption by existing applications without requiring intrusive code changes or scheduler modifications. We evaluate our approach by integrating the DMR API into two large-scale scientific applications and deploying them on three TOP500 supercomputers under realistic production configurations. Our non-invasive malleability solution achieves performance comparable to static baselines in controlled environments while substantially reducing node-hour consumption for identical workloads. These results show that malleability can be effectively exploited on production systems using vanilla resource managers, lowering the barrier to adoption of dynamic resource management in HPC.

Recent approaches combining Large Language Models (LLMs) with retrieval-augmented reasoning have shown promise for automated fact verification. To process complex claims, these verification pipelines typically execute multi-stage workflows that coordinate tightly coupled modules, including claim decomposition, evidence gathering, and verdict prediction. However, existing methods optimize individual stages in isolation or rely on fixed heuristics, which limits adaptive coordination among stages and can lead to suboptimal outcomes. In this work, we propose ProFact, an agentic reinforcement learning framework for end-to-end optimization of multi-stage fact verification trajectories. ProFact trains a unified policy to coordinate claim decomposition, evidence seeking, answer generation, and verdict prediction. To address the sparse and delayed supervision provided by final veracity labels, ProFact introduces process-aware rewards that provide stage-level learning signals throughout the verification process. Empirical evaluation shows that ProFact consistently outperforms strong baselines in both verification performance and inference efficiency. These results highlight the effectiveness of process-aware trajectory optimization for multi-stage fact verification.

Identifying latent dynamical systems from noisy, high-dimensional measurements is a central problem at the intersection of representation learning, system identification, and scientific discovery. We present DYSCO, a multi-view temporal contrastive learning algorithm that jointly recovers latent trajectories and the governing dynamics from such observations, by leveraging multiple independent noisy views of the same underlying process to disentangle signal from noise. By parameterizing the dynamics in a structured functional basis, our framework further enables symbolic recovery of the governing equations within an affine gauge. We offer theoretical guarantees for strong identification up to an affine indeterminacy, extending prior identifiability results to the realistic setting of noisy nonlinear observations. Empirically, we demonstrate accurate recovery of both latent trajectories and flow fields across a diverse set of dynamical regimes (e.g., chaotic, oscillatory, and metastable) under both Gaussian and Poisson observation noise, the latter being particularly relevant for neural recordings.

Gait-based Parkinson's disease assessment increasingly relies on heterogeneous sensors, but clinical systems rarely collect all modalities simultaneously. New sensors may arrive through device upgrades, protocol changes, or multi-center deployment, while historical patient data are often unavailable because of privacy and storage constraints. This modality-incremental setting faces three challenges: unreliable cross-modal distillation, modality-specific statistical shifts, and reduced plasticity after preservation. We propose MOSAIC, a compact continual learning framework. First, we identify the Toxic Teacher phenomenon and introduce Modality-Specific Warm-Up to stabilize newly learned modality representations before distillation. Second, we propose a statistics-decoupled MSBN architecture that isolates sensor statistics while maintaining a shared semantic backbone. Third, we design a curriculum-guided repulsive objective for Plasticity Recovery, preserving legacy knowledge while recovering modality-specific capacity. Experiments on three multimodal Parkinson's gait datasets show that MOSAIC improves final performance and mitigates forgetting. Project code is available at: https://github.com/minlinzeng/MOSAIC_Modality-Specific-Adaptation-for-Incremental-Continual-Learning-in-PD-Gait-Assessment.git

Humor plays a central role in human social relationships, and recent advances in computational humor create new opportunities for integrating humor into human-robot interaction (HRI). While large language models (LLMs) can generate diverse forms of humor, it remains unclear how humor style, joke content, and language preference shape perceptions of robot-delivered humor in group settings. In this exploratory study, we employed a mixed factorial design in which participants evaluated AI-generated jokes delivered by a robot in a university classroom. We examined the effects of humor type (Affiliative, Self-Enhancing, Aggressive, Self-Defeating) and joke content (person-related vs. political) on perceived funniness and appropriateness, as well as preferred language. Results show that humor type significantly influences funniness, with Aggressive and Affiliative humor rated higher, while joke content primarily affects appropriateness, with person-related jokes preferred over political ones. Language preference was shaped by both joke content and participants' self-reported fluency and humor practices.

The growing need to represent diverse perspectives has increased interest in pluralistic LLM generation. Although difficult to operationalize, identifying perspectives expressed in text would provide clear guidance on pluralistic alignment and more clearly articulate the pluralistic gap in LLM generation. While models have been shown to reduce the diversity of training data and generate homogeneously, this has been demonstrated primarily on multiple-choice questionnaires or using high-level characteristics of free-form text. In this paper, we introduce and implement a domain-agnostic multi-layered framework for unsupervised extraction of perspectives suitable for identifying the pluralistic gap in LLM-generated text. We evaluate our framework on book reviews, a highly opinionated dataset representing diverse perspectives, and compare various prompts and models. Our results show that while some models and prompting techniques come close to covering a broad spectrum of perspectives, rarer perspectives remain disproportionately underrepresented, resulting in distributions that diverge from human text.

Speech recognition is challenging for dysarthric speakers. While federated learning (FL)-based ASR can be an effective tool for protecting privacy, it suffers from heterogeneity issues caused by speaker variability. Forcing all speakers to share the same model components can be suboptimal under such heterogeneity, making personalization a promising direction; however, related research on dysarthric speech remains limited. To this end, this paper explores two aggregation strategies to achieve personalization, including the parameter-based averaging strategy and the embedding-based averaging strategy. Experiments on UASpeech and TORGO show that the proposed methods outperform the baseline regularized FedAvg by statistically significant WER reductions of up to 0.99% absolute (3.15% relative) on UASpeech and 0.56% absolute (4.73% relative) on TORGO, respectively.

To better understand Martian Surface, which is needed to enable Rovers navigate Mars with ease, it is necessary to be able to determine the location of mounds. Detecting and studying these morphologies can also help us find evidence of extraterrestrial life, in this case, more specifically, water or signs of life conducive environments. Detection of mounds was done by manually mapping morphological parameters onto Digital Elevation Models. This paper solves the problem by automatically detecting and or predicting mounds on Mars using Neural Network based Semantic Segmentation methodologies. This is done by using supervised semantic segmentation model and generative adversarial approach. A comparison of the approaches shows that adding extra artificially generated data did not improve the result.

Maritime accident adjudication reports contain critical tribunal findings for root cause analysis (RCA), yet retrieving relevant precedents and drafting consistent reports from decades of records remains labor-intensive. This paper proposes a multi-field hybrid retrieval-augmented generation (RAG) framework for automated maritime RCA, utilizing a comprehensive dataset of 13,329 Korea Maritime Safety Tribunal (KMST) reports (1971-2025). We transform raw adjudications into a structured knowledge base of "incident cards", indexing three distinct fields-Summary, Causes, and Disposition-alongside a hierarchical L1/L2 cause taxonomy. Our retrieval strategy employs a field-aware hybrid approach, fusing sparse and dense rankings via Reciprocal Rank Fusion (RRF). Given the lack of large-scale expert relevance labels, we evaluate retrieval performance using ceiling-normalized recall and nDCG based on a metadata-derived proxy relevance score. Experimental results demonstrate that our proposed retrieval significantly outperforms baseline methods, improving NormRecall@100 from 0.18 to 0.55. Furthermore, grounding the generator on the retrieved precedents enhances RCA generation quality over an LLM-only baseline, increasing the LLM-as-a-judge score from 3.34 to 3.72. These findings suggest that field-aware RAG can substantially streamline maritime safety investigation workflows by enabling faster precedent search and more consistent, evidence-based RCA drafting.

Text-to-image diffusion models have achieved impressive results in synthesizing high-quality images from natural language prompts. However, commonly used prompting strategies remain relatively generic, limiting the model's ability to accurately express emotional intent and nuanced affective attributes. This work proposes EPIG, a method that enhances emotional expressiveness at the prompt level prior to image generation. Grounded in psychologically informed emotion representations (valence-arousal) and leveraging structured, role-aware prompt enrichment, EPIG enriches emotion-related components of prompts without modifying or retraining the image generation backbone. The resulting emotion-aware prompts guide the generative process toward more emotionally coherent visual outputs, with particular effectiveness in controlling arousal. EPIG is lightweight, training-free, and well suited for resource-constrained and personalized image generation scenarios. Experimental results on a benchmark of 10 diverse prompts show that EPIG reduces mean arousal error compared to strong baselines, including naive insertion and LLM-based prompt expansion, with reductions of 14% and 12%, respectively. These improvements are statistically significant. EPIG also preserves valence alignment and semantic consistency, as measured by CLIPScore and supported by ablation studies. The effect is more pronounced on prompts containing explicit subjects such as humans, children, or animals, where the reduction reaches 17%, highlighting the subject-sensitive behavior of the proposed method.

Defining query difficulty is one of the hardest problems in deployment engineering. Existing LLM routers rely on surface features such as domain labels, keywords, and token count, ignoring the within-domain variance that actually determines model success. Frontier models cost ten to one hundred times more than local open-weight models, so at production scale even small per-request savings become a direct cloud-bill lever. We present Brick, a multimodal router that scores each model on six capability dimensions, combines this with a per-query difficulty estimate, and dispatches via a cost-penalized geometric rule. A continuous preference knob lets operators slide between max-quality and max-saving profiles at deploy time. On a benchmark of 5,504 queries, Brick at max-quality reaches 76.98% accuracy, beating the best single model (75.02%) and all tested routers. At a neutral cost-quality profile, Brick achieves 74.11% accuracy at 4.71x lower cost than always using the strongest model. At min-cost, it cuts cost 22.15x with 11.85 points accuracy loss. Median latency drops from 51.2s to 22.8s.

A key strength of diffusion models lies in their flexibility, since their outputs can be controlled at sampling time through guidance. However, beyond simple cases such as conditional sampling, the target distribution is often left implicit, defined only through a sampling rule or a heuristic energy function. To address this, we propose Jeffrey guidance, a principled framework that extends diffusion-model control to applications beyond what standard guidance can express. It leverages Jeffrey's rule of conditioning to update marginal distributions towards a prescribed target, preserving the conditional structure and minimally perturbing the joint distribution. We first demonstrate Jeffrey guidance by targeting a prescribed embedding distribution. With Inception embeddings as the target, this leads to substantial reductions in FID on both CIFAR-10 and FFHQ. We further apply Jeffrey guidance to fairness on CelebA-HQ, updating an unconditional diffusion model to enforce independence between attributes.

Existing computer-use agents remain fundamentally limited in professional software manipulation: GUI-based agents suffer from fragile visual grounding and long-horizon error accumulation, while API-basedapproaches struggle with heterogeneous protocols and inaccessible commercial interfaces. In this work,we identify the Component Object Model (COM) as a unified executable abstraction, proposing COM-as-Action: a new paradigm that reframes professional software interaction as deterministic program synthesisrather than sequential visual control. To validate this paradigm in the most demanding environments, weintroduce ComCADBench, the first benchmark for agents operating real industrial CAD software. Ourexperiments reveal a substantial paradigm gap: frontier proprietary models achieve near-zero successunder GUI-based interaction, whereas COM-based execution yields substantial immediate gains. Tobridge the remaining gap between syntactic correctness and geometric accuracy, we develop ComActor, aself-correcting agent trained through a progressive three-stage framework, alongside ComForge, a scalableplatform for large-scale training in Windows containers. Extensive experiments show that ComActorachieves state-of-the-art performance on ComCADBench, with strong resilience in long-horizon taskswhere baselines collapse, and generalizes to external CAD benchmark.

Passive acoustic monitoring holds great promise for ecological inference, yet existing automated tools are typically narrowly trained and non-transferable. We address these limitations with PULSE, a semi-supervised, multi-task framework for Orthoptera bioacoustics, combining weakly-supervised species classification, self-supervised learning on unlabelled field audio, and knowledge distillation from a general-purpose bioacoustic model. Our domain-adapted specialist model outperforms a state-of-the-art general model across all metrics (macro F1: 0.21 vs. 0.07; AUC: 0.74 vs. 0.45; AP: 0.32 vs. 0.19), with active learning further raising F1 to 0.34 and AUC to 0.84. Beyond classification, the learned embeddings encode ecologically meaningful structure, exposed through an interactive visualisation tool for ecological discovery.

Large reasoning models (LRMs) improve complex problem-solving by generating long intermediate reasoning traces, but this substantially increases inference costs. NVFP4 inference offers a promising approach to reduce both computational and memory costs through hardware-supported low-precision execution. However, directly applying NVFP4 to LRMs introduces two practical limitations: reasoning accuracy degrades under quantization, and existing NVFP4 kernels do not fully realize latency benefits in small-batch autoregressive decoding. In this work, we analyze the effect of NVFP4 quantization on token-level uncertainty during reasoning. We show that quantization increases incorrect sampling at low-entropy symbolic tokens, while causing over-concentration on a small set of tokens in high-uncertainty reasoning steps. Based on this observation, we propose \textbf{ReSET}, a reasoning-step entropy-based temperature-scaling method that estimates step-level uncertainty online and adapts the decoding temperature using both token-level and step-level entropy signals. To address the latency gap, we further design a CUDA-core small-$M$ NVFP4 kernel for latency-critical autoregressive decoding. Across reasoning benchmarks and model scales, ReSET improves NVFP4 reasoning accuracy by up to $\sim\!$2 points over the NVFP4 baseline. Our CUDA-core small-$M$ kernel further improves latency-critical decoding, delivering up to $2.5\!\times$ kernel-level speedup over NVFP4 vLLM and approximately $2\!\times$ end-to-end decoding speedup over BF16. Code is available at https://github.com/aiha-lab/ReSET.

Post-training methods such as supervised fine-tuning (SFT) and preference optimization typically align language models toward a single global assistant behavior. While effective for improving average helpfulness, this can suppress the natural variation of human responses across languages, tasks, and dialogue settings. We study this problem as conditional human-distribution alignment: models should match the human response distribution appropriate to the current interaction context, rather than a universal response style. We introduce PolyAlign, a distribution-aware alignment framework that organizes bilingual interaction data into bucket-specific human reference distributions defined by language, interaction track, response family, and length. PolyAlign combines Bucket-Aware SFT, which balances optimization across heterogeneous buckets, with Human-Distribution Preference Optimization (HDPO), which regularizes preference learning using critic-estimated distance to bucket-specific human support. Across a bilingual evaluation suite covering English and Chinese single- and multi-turn settings, PolyAlign improves conditional naturalness and distributional faithfulness while preserving competitive task utility. The results suggest that post-training should move beyond global alignment objectives toward interaction-aware alignment with human response distributions.

This paper proposes a novel loss concept for supervised classification tasks. Rather than enforcing a direct mapping from each input sample to a single assigned label, we define an optimization objective over all classifier outputs as a bimodal Gaussian distribution. This softer target formulation implicitly captures class ambiguity, mitigates overfitting, and encourages the learning of more robust decision boundaries, all without requiring additional label information. Experimental results demonstrate consistent improvements in robustness, with particularly pronounced gains in low-data regimes, while requiring only minimal modifications to standard training pipelines.

Distinguishing self from others is a prerequisite for social intelligence, yet humanoid robots that increasingly share workspaces with humans still lack this ability. Here we show that a humanoid robot can learn self-other distinction from proprioceptive-visual correspondence, without any identity labels or kinematic models. Once established, this distinction bootstraps a predictive self-model that maps joint configurations to three-dimensional body occupancy, capturing how the robot's body changes with action. In multi-agent scenes involving humans or morphologically identical robots, the system reliably identifies itself, learns a 3D self-model, and supports downstream tasks including target reaching, collision-aware motion planning, and human-to-robot motion retargeting. Together, these results outline a route toward bodily self-representation in robots that act and coordinate alongside others in shared physical environments. Project page: https://euron-zc.github.io/humanoid-self-model/.

Evaluating new large language models typically requires costly human annotation campaigns at scale. LLM-as-a-judge offers a cheaper alternative, but judge scores carry systematic errors - such as position bias, self-preference, or intransitivity - that can strongly miscalibrate the resulting rankings. We quantify the resulting judge-human disagreement at two complementary levels. At the local level, we estimate per-battle uncertainty from the judge's own score differences by propagating calibrated win probabilities rather than hard labels into the Bradley-Terry procedure. This alone provides a drastic improvement to Elo estimation accuracy, bringing LLM-derived ratings within 17.9 Elo MAE of human-derived ones when averaged over 55 held-out models on LMArena. At the global level, we apply split conformal prediction to the residual gap between LLM-derived and human-derived Elo ratings across held-out models, producing prediction intervals with distribution-free marginal coverage guarantees that account for irreducible LLM-human disagreement. Together, these two layers yield a low-cost evaluation tool that provides developers with calibrated Elo estimates and honest uncertainty bounds, without access to large-scale human annotations.To facilitate reproducibility, we release our code at https://github.com/kargibora/SoftElo .

Large language models (LLMs) are increasingly used as interactive assistants for technical problem solving. However, when users provide incomplete descriptions or plausible but unverified explanations, LLMs may prematurely align with these assumptions and propose solutions before collecting sufficient evidence. We refer to this behavior as user-driven sycophancy: the tendency of an LLM to reinforce a user-provided hypothesis instead of testing alternative explanations. This paper introduces LLM-as-an-Investigator, an evidence-first agentic AI methodology for robust problem diagnosis. The approach is implemented through a Solution Investigator Agent, which estimates the ambiguity of an initial problem description, generates candidate hypotheses, asks targeted clarification questions, and updates hypothesis probabilities after each answer. Rather than producing an immediate response, the agent continues the investigation until the evidence makes one candidate explanation stronger than the alternatives. To evaluate the approach, we build a benchmark from solved technical forum threads in mechanical, electrical, and hydraulic domains. We use a three-agent evaluation pipeline in which a Problem-Solution Extractor Agent converts solved threads into structured cases, a Ground-Truth Evaluator Agent simulates the user while hiding the known solution, and the tested assistant attempts to recover the solution through dialogue. The experiments compare standard assistants, reasoning-oriented LLMs, and the proposed investigator-based model across LLM backbones. In addition to diagnostic accuracy, we analyze how standard assistants follow misleading user hypotheses in diagnostic cases. The results show that the proposed approach identifies the problem more accurately than direct prompting and reasoning-only baselines, while its evidence-first protocol helps reduce user-induced conversational bias.

Arabic and Hebrew, as closely related Semitic languages, share a substantial lexicon of true cognates, misleading false friends, and modern loanwords. This overlap poses a challenge for cross-lingual semantic understanding in large language models (LLMs). To evaluate this capability, we introduce SemCog Bench, a curated benchmark of 1,858 Arabic--Hebrew word pairs with sentence-level annotations for cognate identification and semantic disambiguation. We evaluate open-source and commercial LLMs across multiple input representations (raw, diacritized, Romanized, and phonetic) and reveal a critical gap in cross-lingual reasoning. While models achieve high accuracy on true cognates, performance drops sharply on false friends and loanwords, reflecting a strong reliance on surface-form similarity. Furthermore, sentence-level context yields only modest improvements, suggesting that contextual cues alone are insufficient to overcome misleading form-based signals. These findings reveal a fundamental limitation of current LLMs in resolving cross-lingual form--meaning conflicts and establish SemCog Bench as a rigorous benchmark for multilingual semantic reasoning. Our code and data are publicly available.

Optimal transport (OT) has been shown to detect hallucinations in neural machine translation (NMT) by measuring the geometric distance between cross-attention distributions and a reference distribution, without any supervision. We extend this analysis to all six decoder layers of the Fairseq DE-EN model ($N=3{,}414$), showing that Wass-to-Unif and Wass-to-Data are complementary detectors specialised across hallucination types, that detection is concentrated in layers L1--L4 with L5 anti-predictive for subtler types, and that hallucinated translations lack the exploratory attention phase present in correct translations from the first decoding step. We further evaluate whether the geometric signal transfers to abstractive summarization faithfulness detection: our unsupervised OT detector on AggreFact ($N=1{,}116$) achieves $57.2\%$/$57.6\%$ balanced accuracy on CNN/XSum -- above chance but substantially below supervised MiniCheck-Flan-T5-L($69.9\%$/$74.3\%$). This gap is principled: unlike NMT hallucinations, unfaithful summaries can attend correctly to source tokens while misrepresenting their content, a failure mode invisible to concentration-based OT metrics by construction. Structural experiments on T5-base confirm consistent decoder organisation across depth, with Layer~3 showing peak concentration and Layer~12 being most critical for generation quality. Together, the results establish OT on cross-attention as a reliable detector when the failure mode is source disengagement, a principled interpretability tool regardless of task, and fundamentally limited when faithfulness failures occur downstream of attention.

AI systems are being deployed across medical imaging faster than their failure modes are understood. At this point in time, the failure of greatest clinical concern is hallucination: clinically plausible but factually incorrect outputs, including fabricated anatomical structures, missed findings, incorrect laterality, and invented measurements in generated reports, with direct consequences, for example, for biopsy decisions, staging, and treatment planning. This structured narrative synthesizes peer-reviewed studies, benchmark datasets, and FDA regulatory guidance across five imaging modalities to produce a cross-modality analysis of hallucination taxonomy, etiology, detection, and mitigation. Specifically, we address three questions in this study: (1) how can existing taxonomies be unified across modalities?, (2) how do medical-specialized foundation models hallucinate less than general-purpose ones?, and (3) which mitigation strategies are effective and compatible with FDA lifecycle oversight? We note that three taxonomic frameworks together cover the imaging pipeline in a way no single framework does alone. We also highlight that general-purpose foundation models outperform medical-specialized models on hallucination-specific benchmarks, indicating that narrow domain fine-tuning can introduce overfitting-induced confabulation. At the same time, the oversight of radiologists remains essential; for instance, a very high percentage of of AI-generated flags required expert correction before clinical use. Physics-informed architectural constraints, Chain-of-Thought prompting, and human-in-the-loop safeguards each address different failure modes and is effective when combined. All findings are mapped to the FDA's Total Product Lifecycle and Predetermined Change Control Plan frameworks, which treat hallucination management as a lifecycle obligation rather than a pre-deployment checklist.

Reward models are a key component of reinforcement learning from human feedback (RLHF), aligning language models toward both helpful and harmless behaviour. However, the internal mechanisms underlying these objectives and their conflicts remain poorly understood. We study alignment tension in reward models trained under helpfulness-only, harmlessness-only, and mixed-objective settings. We find that mixed-objective models often underperform single-objective models, indicating interference between objectives. Using activation-based methods, we identify neurons associated with each objective and study their functional roles via targeted ablations. We find that these neurons causally support their corresponding objectives while often negatively affecting the opposing one. We find that a substantial proportion of neurons are shared between helpfulness and harmlessness, and that these shared neurons exert a disproportionate influence on model behaviour, contributing to alignment tension. Additionally, our results provide insights and mechanistic interpretation into how alignment objectives are represented in reward models and why multi-objective alignment remains challenging, motivating future work on disentangled and controllable alignment methods.

Human decision-making often involves choosing between multi-attribute alternatives, yet classical models assume fully compensatory utility aggregation despite evidence that people reject options with poor performance on critical attributes. We propose a bounded trade-off reasoning framework in which decisions are governed by a screening process that evaluates the balance between gains and losses across attributes. The model introduces a trade-off tolerance parameter that controls acceptable imbalance and can vary across contexts. Through simulation, we show that this mechanism produces preference patterns that differ from standard utility-based models and captures context-dependent variation in trade-off behavior. These results establish bounded trade-off screening as a plausible computational mechanism for multi-attribute choice and generate testable predictions for future behavioral studies.

Multi-agent debate (MAD) can improve large language model reasoning, but fixed debate pipelines often waste computation and can amplify correlated errors among similar agents. We propose ARMOR-MAD, a training-free heterogeneous MAD framework that treats debate as conditional computation. ARMOR-MAD combines three components: Pre-debate Agreement Routing (PAR) decides whether independently generated Round-0 answers require debate; Early Agreement Stopping Evaluator (EASE) stops debate after convergence; and Semantic Outlier Detection (SOD) down-weights abnormal final answers during aggregation. Across MATH Level 5, GSM8K, MMLU, and MMLU-Pro, ARMOR-MAD consistently improves over fixed-round heterogeneous debate with the same model pool, reaching 65.5\%, 96.5\%, 90.0\%, and 81.5\% accuracy, respectively. The results suggest that genuine model heterogeneity and agreement-based control are both important for making MAD more accurate and efficient.

Recent AI systems can generate texts, software architectures, hypotheses, designs, and scientific workflows that appear creative. This paper asks under what conditions a machine can become genuinely creative, and how human agency can be preserved within shared cognitive and creative environments. It develops a requirement framework derived from Designics, the science of meaning-bearing intentional change. The paper argues that genuine machine creativity should not be defined by output novelty, current performance, or transient architecture alone. Instead, creativity is understood as the structural transformation of incomplete situations through recursive intervention dynamics. On this view, it depends on ten requirements: environment representation, scoped perception, conflict identification, intervention capability, consequence observation, knowledge and environment update, rescoping, local-to-global unfolding, value-based scoping, and human-AI co-living. These are organized through the three laws of Designics: perception, conflict, and capability. The paper illustrates the computational tractability of these requirements through selected cyber-physical and cyber-biological studies, including recursive element extraction, autonomous mesh generation, and neurophysiological and workload analysis. It then treats open-ended systems, automated discovery frameworks, self-modifying agents, foundation models, and agentic workflows as pressure cases: they demonstrate powerful generative means but do not by themselves establish genuine machine creativity. Finally, the paper argues that proactive AI ethics is internal to genuine machine creativity rather than an after-the-fact filter. Value-based scoping and human-AI co-living must shape how creative machines perceive environments, identify conflicts, select interventions, observe consequences, update knowledge, and rescope future action.

Deep learning has become the dominant paradigm in Wearable Human Activity Recognition (WHAR), yet progress is obscured by a comparability crisis. Results are often reported using inconsistent datasets, custom data processing, and varying evaluation protocols, making state-of-the-art claims fragile. We address this with a large-scale, open-source benchmark that integrates 30 diverse datasets under standardized processing, unified model interfaces, and a shared cross-subject evaluation protocol. Evaluating 17 representative architectures across 4760 training runs, we jointly measure predictive performance alongside on-device latency, peak memory, and model size on an Android reference device. Our results reveal that the WHAR state of the art is distributed rather than dominated by a single architecture. While CNN-HAR achieves the highest mean macro-F1, top-performing models cluster tightly, indicating contemporary architectures have converged near a predictive performance ceiling. When accounting for deployment efficiency, compact neural models, such as TinierHAR, and classical Random Forests define the practically relevant Pareto frontier, whereas larger recurrent and hybrid models incur high hardware costs without corresponding performance gains. Consequently, while predictive performance has plateaued, substantial potential for future progress remains in optimizing deployment efficiency and improving adaptation to domain shifts. We release our full framework to support transparent reuse and extension.

User experience (UX) centered on usability, perceived consistency, and functional clarity is fundamental to real-world user interfaces (UI). The application of multimodal large language models (MLLMs) in the field of user interfaces is evolving rapidly, such as visual element grounding, graphical user interface (GUI) agents, and design-to-code generation. However, research efforts on evaluating UX based on UI screenshots are still immature. To address this, we propose UXBench, a novel multimodal benchmark consisting of 2,000 VQA data samples designed to assess MLLMs' ability to perform UI-based reasoning. UXBench includes 8 tasks based on real-world UI screenshots that require fine-grained diagnosis of UX issues across layout relationships, visual hierarchy, and content consistency. Our extensive evaluation of mainstream MLLMs shows that they remain fundamentally limited in their capacity for UI-based reasoning. The results underscore the need for further advancements in this area. To bridge this gap, we propose UI-UX, an MLLM based on Qwen3-VL-4B-Thinking foundation model and enhanced via reinforcement learning with two key innovations: a reward routing mechanism that dynamically balances perceptual understanding and logical reasoning during inference, and an asymmetric transition reward that suppresses redundant or insufficient reasoning steps. Experiments demonstrate that UI-UX achieves state-of-the-art (SOTA) performance on UXBench, attaining an accuracy of 0.7963 -- surpassing Claude-4.5-Sonnet's 0.6550 -- while exhibiting strong generalization across diverse UI tasks and maintaining low inference latency.

Continuous-state generative samplers, including diffusion and flow-matching models, evolve through continuous reverse-time dynamics, yet their samples often undergo abrupt qualitative changes: trajectories commit to modes, semantic alternatives collapse, and small perturbations in narrow time windows can produce large downstream effects. This paper develops a geometric account of such phase-transition-like behaviour. We view denoising as gradient descent on a free energy landscape and show that sharp transitions arise near projection caustics, where the nearest-point projection onto the data support ceases to be unique. Motivated by this perspective, we introduce the Critical Boundary Detector (CBD), as practical diagnostics for score-direction instability. Across toy models, standard diffusion models, and latent text-to-image diffusion models, CBD localises mode commitment, predicts intervention-sensitive windows, and supports targeted control in geometrically sensitive regions. Our results connect geometry of data and dynamics of diffusion generation.

Prompt-based LLMs are increasingly used for stance detection, but harder examples are not always repaired by clearer instructions, reasoning prompts, retrieval, or debate. We introduce SICI (Stance Inference Complexity Index), a seven-dimensional diagnostic measure of the semantic-pragmatic burden imposed by a target--text pair. Across SemEval-2016 and VAST, SICI predicts LLM accuracy better than surface proxies and shows substantial cross-scorer reliability ($α=0.771$). More importantly, LLM errors change regime as SICI increases: low-complexity examples invite over-attribution, especially Against predictions; intermediate examples form an unstable boundary; and high-complexity examples rapidly concentrate on None. This phase-transition-like structure persists across GPT-3.5, GPT-4o-mini, DeepSeek-V3, and GPT-4o, although stronger models move the boundaries. A 15-method intervention study further shows that prompting, retrieval, and debate often shift models along the attribution--abstention axis rather than removing the high-complexity bottleneck.

Building patient-specific cardiac models sits at the heart of precision cardiology, yet getting those models into clinical use keeps running into the same wall: mesh generation is slow, messy, and frustrating. The standard workflow -- segmenting the image, running Marching Cubes, and then manually cleaning up the result -- is time-consuming, inconsistent across operators, and demands specialist knowledge most clinical teams do not have. We take a fundamentally different approach. Instead of treating segmentation and mesh generation as two separate problems, we train a single end-to-end network that goes directly from a raw 3D medical image to a smooth, simulation-ready cardiac surface mesh. The core is a 3D Swin Transformer encoder-decoder that extracts volumetric features from CT or MRI volumes, paired with a Graph Attention Network (GAT) head that iteratively deforms a template mesh to fit the patient's cardiac boundary. We tested on the MM-WHS 2017 benchmark using both CT and MRI. Segmentation scores were competitive (Dice of 0.84 on CT, 0.83 on MRI), but the primary focus is mesh quality: mean Chamfer distance of 1.8 mm, with 95th-percentile surface distance below 5 mm. Every mesh is produced in a single forward pass -- no Marching Cubes, no smoothing filters, no manual cleanup. We argue that for cardiac digital twin pipelines, geometric fidelity and topological correctness matter more than pixel-level Dice scores. By removing the post-processing bottleneck, this approach makes patient-specific cardiac simulation substantially more accessible for clinical use.

Bioethical debates increasingly unfold on social media, yet stance detection research lacks large-scale, domain-specific resources for modeling such context-dependent discourse. We present BioStance, a context-aware dataset of 39,600 annotated Post-Comment pairs from Reddit bioethical discussions. BioStance covers six controversial targets across three dimensions of bioethical controversy: fundamental value conflicts, individual liberty versus collective responsibility, and technological uncertainty. Each instance preserves hierarchical conversational context and is labeled by three independent annotators using a three-class stance scheme: Favor, Against, and None. The annotations achieve a mean Krippendorff's $α$ of 0.82, indicating substantial reliability. By combining thematic diversity, conversational structure, and high-quality human annotation, BioStance supports research on context-aware stance detection, argument mining, and computational analysis of bioethical discourse.

Multinational companies increasingly require cross-jurisdictional contract review, yet existing legal NLP datasets are largely restricted to a single jurisdiction. We introduce LAUKIN (Legal equivalence dataset of Australia, UK, and INdia), a dataset of clause pairs (AU-UK, UK-IN, IN-AU) labelled for boolean legal equivalence. We develop a novel multi-stage retrieval and reranking pipeline to construct the initial clause pair mapping, with a subset of clause pairs subsequently annotated by legal experts as Equivalent or Not Equivalent. The dataset comprises 14,727 clause pairs from 204 contracts across 8 agreement types, of which 3,000 are manually labelled: 900 train, 600 dev, and 1,500 test. We evaluate 12 models across 4 techniques, achieving a best macro-F1 of 65.11%, establishing LAUKIN as a challenging benchmark. Results reveal that, despite shared legal heritage, drafting conventions diverge significantly across jurisdictions, making cross-jurisdictional equivalence classification non-trivial. LAUKIN also includes 11,727 unlabelled training pairs to support future semi-supervised learning research in legal NLP.

Federated Learning(FL) is predominantly deployed in enterprise environments, where limited transparency and restricted auditability hinder broader adoption. Existing FL systems often suffer from opaque aggregation processes, making it unclear which model updates are accepted or discarded. Current mitigation strategies typically rely on external validators introducing additional computational and communication overhead. In this paper, we propose a novel FL framework that leverages existing Web3 technologies to enhance transparency, trust and auditability throughout the training process. The framework adopts a hierarchical architecture in which delegated managers orchestrate the FL training process within their respective federations. To mitigate adversarial and poisoning attacks, a combination of novelty detection and consensus mechanisms were employed. Model updates are encoded and broad casted to all managers, who independently evaluate their validity and those model updates that are approved by the consensus are incorporated into the global model. Additionally, a reputation score based backup mechanism is employed to ensure model generation. Extensive experiments conducted under real world scenarios demonstrate the effectiveness, resilience of the proposed framework, highlighting its potential to enable transparent FL beyond traditional enterprise setting.

In recent years, driven by the computational demands of data-intensive applications such as artificial intelligence and scientific computing, analog computing has gained renewed interest. Given the diversity of computational tasks and recent advancements in analog CMOS circuits and resistive memory technologies, we refer to the evolving landscape as modern analog computing. In this context, we identify three core computational primitives: solving differential equations, solving matrix equations, and performing matrix-vector multiplications, and we explore the connections among them. We also examine various hardware implementations of these analog computing operators, including those built with discrete components, integrated circuits, and resistive memory devices. Among these, resistive memory arrays emerge as particularly promising due to their implementation efficiency. The paper then surveys recent progress in leveraging modern analog computing to solve differential and matrix equations using both advanced analog CMOS circuits and resistive memory arrays. Finally, we discuss the applications of these circuits, the precision and scalability issues and their potential solutions, the relationship with in-memory computing, and the unique computational complexity of analog computing. This paper provides a unified perspective on analog computing, highlighting its strengths, current developments, and challenges, and positioning it as a pivotal enabler of next-generation computational frontiers.

Transfer learning is usually studied as a consequence of distribution shift. This paper identifies an orthogonal failure mode in which the data distribution is fixed and the loss changes. This setting is called \emph{loss shift}. A loss determines which information in \(X\) is Bayes-relevant, and two losses may therefore require different representations even under the same joint law \(P(X,Y)\). The idea is formalized using Bayes quotients, which allow losses to be ordered by refinement. In the Bayes-quotient formulation, strict refinement gives an immediate qualitative obstruction. A source-minimal representation for a coarser loss is insufficient for a strictly finer target loss. For finite-output log loss, this obstruction becomes an exact quantitative identity. The excess risk is the conditional information about \(Y\) discarded by the representation. Experiments in controlled, learned, synthetic-image, and real-image settings show the predicted effect, i.e., classification-equivalent representations can have different optimal log-loss performance under a fixed data distribution.

Large language model (LLM) agents are increasingly expected to operate over long-term interactions, where information from past dialogues must be preserved and recalled to support future tasks. However, as interactions accumulate, the memory store grows without bound and fills with redundant entries that inflate storage cost and degrade retrieval by crowding out the most useful evidence. Furthermore, this is especially limiting on resource-constrained platforms with hard memory budgets, motivating us to formulate storage-budgeted memory management, the task of keeping an already constructed memory store within a fixed budget while preserving information useful for future interactions. To this end, we then propose MemRefine, an LLM-guided framework that, since surface similarity poorly reflects factual value, uses similarity only to propose candidate pairs and defers delete, merge, and preserve decisions to an LLM judge based on factual content, iterating until the budget is met. Across multiple memory frameworks and long-term conversation benchmarks, MemRefine consistently meets target budgets while preserving downstream performance and outperforming rule-based baselines under tight budgets.

Mental health problems such as anxiety, depression, and suicide remain urgent global challenges, where timely and accurate assessment is critical for effective intervention. Recently, large language models have been explored for mental health assessment. However, existing general-purpose post-training methods do not align with the cognitive processes of human assessment, which may lead to unreliable reasoning outcomes. To bridge this gap, we propose Cognitive Relative Policy Optimization (CRPO), a reinforcement learning framework tailored for the mental health domain. CRPO extends group relative policy optimization by integrating stage-dependent uncertainty modeling into the policy optimization process. Specifically, we introduce a stage-wise entropy regularization mechanism that encourages broad exploration in early reasoning phases and progressively enforces confident decision-making in later stages, mimicking the human cognitive shift from uncertainty to certainty. In addition, inspired by cognitive appraisal theory, we formalize cognitive reasoning stages, thereby guiding theory-grounded interpretable inference. Experiments on 8 mental health datasets show that CRPO achieves an average improvement of 10.4 percentage points in weighted F1-score over the best reinforcement learning baseline. Furthermore, the CRPO-trained model Mental-R1 demonstrates clear advantages compared with existing large language models on reasoning-intensive cases, suggesting that CRPO enhances reasoning capabilities for mental health assessment.

Code review has been the primary quality gate in software development since Fagan formalised code inspection in 1976. For five decades, having a human examine and comment on a colleague's changes before merge has been a cornerstone practice at organisations of every size. Coding agents are large language model (LLM)-based autonomous systems capable of reading, writing, testing, and repairing software. We argue that coding agents have crossed a threshold of capability at which traditional human code review is no longer a necessary component of a software quality pipeline. Our argument rests on two claims: every stated goal of code review can be served by agents at lower cost and higher throughput; the naive integration in which agents write code and humans remain the mandatory reviewers is a dead end because it neither provides meaningful assurance nor scales with AI-assisted throughput.

Interactive LLM agents are becoming part of daily work, but they do not reliably become easier to work with over time: a correction remembered in one session may still be violated in the next. We study this gap between preference access and preference compliance. In tasks derived from anonymized real-user friction cases, Mem0 memory still leaves 57.5% of applicable preference checks violated. We introduce Test-time Rule Acquisition and Compiled Enforcement (TRACE), a drop-in skill-layer pipeline for coding-agent runtimes that mines user corrections, rewrites them as atomic rules, and compiles them into runtime checks that must pass before an agent completes future tasks. Unlike runtime checks written ahead of time by developers, TRACE skills come from the user's own chat corrections. We evaluate TRACE with simulated user-in-the-loop experiments on ClawArena coding-agent tasks and MemoryArena-derived memory-intensive tasks. On ClawArena, TRACE reduces held-out preference violation from 100.0% to 37.6% on in-distribution tasks and from 100.0% to 2.0% on out-of-distribution tasks. On MemoryArena-derived tasks, TRACE reduces in-distribution violation from 100.0% to 60.5% while matching or exceeding the strongest memory baseline on task pass. These results suggest that compiling corrections into runtime enforcement can address a repeated-friction failure mode that memory alone does not reliably solve, reducing the need for users to restate the same correction across future sessions. Experiment code is available at https://github.com/YujunZhou/TRACE_exp, and the deployable skill is available at https://github.com/YujunZhou/tellonce.

Learned representations are central to modern machine learning and are commonly evaluated through predictive performance, robustness, uncertainty estimation, or generalization. However, a learned representation may remain operationally successful while progressively failing to organize persistent residual structures that are not fully captured by conventional evaluation metrics. This article introduces VER, the Vigilant Evaluator of Representations, a conceptual framework for monitoring representational adequacy in learned representations. VER does not propose a new learning algorithm, loss function, or model architecture. Instead, it formalizes a diagnostic process through which persistent residual structures may be identified, analyzed, and interpreted as potential indicators of explanatory insufficiency. The framework distinguishes representational inadequacy from ordinary prediction error, uncertainty, noise, and distribution shift. It introduces a monitoring sequence based on representation identification, explanatory-domain delimitation, residual-structure detection, explanatory-resistance evaluation, and vigilance signaling. VER is intended as a contribution to representation diagnostics in machine learning. Its objective is not to replace existing evaluation methods but to complement them by treating representational adequacy as an explicit object of inquiry. A path toward empirical evaluation through representational-vigilance benchmarks is also outlined.

The rapid updates of online news make tracking event developments challenging, highlighting the need for timeline summarization (TLS). Hallucinations, where LLM-generated content deviates from source news, still remain a critical issue in LLM-based TLS and are not well studied in existing works. To bridge this gap, we identify two primary types of hallucinations: unfaithful content during news summarization and information omission in date-event summarization. Then, we propose NTS-CoT, a novel framework that leverages Chain-of-Thought (CoT) reasoning to mitigate hallucinations in TLS. The framework consists of three key modules: i) Element-CoT to capture essential news elements for faithful summarization, ii) Date Selection to combine temporal saliency and event prominence for timestamp selection, and iii) Causal-CoT to infer causal relationships and reduce omissions in date-event summarization. Extensive experiments, including quantitative analysis on three TLS benchmarks and human evaluation, demonstrate that NTS-CoT outperforms state-of-the-art baselines, effectively mitigating hallucinations and improving LLM-based TLS performance. Our source code is available at https://anonymous.4open.science/r/NTS-CoT .

Block Attention Residuals (Block AttnRes) by replace fixed additive residuals with a learned softmax over earlier depth-source representations, surfacing cross-layer routing as an inspectable tensor in the forward pass. This is a tempting interpretability target: information flow normally inferred indirectly is now directly observable. We ask whether such exposure suffices for mechanistic interpretation. We probe two same-scale ($0.6$B) Block AttnRes checkpoints under identical routing-ablation interventions: a vanilla Qwen3 inference-wrapped through a deterministic recency-bias schedule that the codebase admits as a routing-equivalent loading path, and a Block AttnRes Qwen3 trained from scratch with routing as part of optimisation. The wrapped baseline's routing weights are content-independent and reproduce the schedule's analytic prediction. The trained AttnRes checkpoint instead exhibits three localised routing motifs: an embedding-source pathway through early-layer MLP, a current-state pathway through early-layer attention and MLP, and an older-history pathway through late-layer attention. Beyond this stratification, we find a sharp dissociation between average routing mass and causal importance: in both sublayers, the largest mass slice is not the largest causal contribution, and one source family carries appreciable mass with no detectable causal role under intervention. Architectural exposure of routing is therefore necessary but not sufficient for mechanistic interpretation: structured depth routing emerges only when routing has been part of training, and even then, descriptive routing summaries should be treated as candidate hypotheses to be tested by causal interventions, not as evidence of mechanism in their own right.

Vision-language models (VLMs) achieve strong singleshot spatial grounding, yet lack any mechanism to observe and correct their own predictions. We find that naively prompting a VLM to iterate over rendered visualizations of its predictions causes catastrophic failure: Acc@0.5 on referring expression comprehension collapses from 79.6% to 48.7% (a 31 percentage point drop), revealing a fundamental gap between grounding capability and self-correction ability. We propose Iterative Visual Thinking (IVT), a closed-loop framework in which the model predicts a bounding box, observes the prediction rendered on the image, and iteratively refines through visual feedback. A two-phase training recipe closes the self-correction gap: first, we exploit the base model's own predictions as realistic errors and prompt a teacher VLM to generate corrective reasoning traces, yielding supervised data without human annotation; second, we apply Group Relative Policy Optimization (GRPO) with a simple IoU reward to stabilize multi-step refinement. On a mixed benchmark spanning RefCOCOg, Ref-Adv, and Ref-L4 (505 test samples), SFT warm-up with IVT surpasses the single-shot base model on every metric: Acc@0.5 rises to 82.0% (+2.4pp), Acc@0.7 to 74.1% (+3.2pp), and Acc@0.9 to 48.3% (+2.8pp). GRPO further reduces per-step IoU degradation by 5x, stabilizing the refinement trajectory. All training uses only 2,400 samples on a single GPU, demonstrating that spatial self-correction is a learnable capability that can be instilled at modest scale.

Climate and environmental decision-making increasingly requires reasoning across heterogeneous inputs, including gridded physical data, satellite imagery, geospatial context, and simulator outputs. Weather and climate foundation models can forecast well, but do not reason interactively in language, while large language models (LLMs) reason in language but cannot operate directly on high-dimensional Earth-system data. As a result, real scientific workflows in Earth-science remain underserved. We introduce TerraBench, a benchmark for grounded Earth-science reasoning, built on TerraAgent, a ReAct-style executable framework that interleaves reasoning, tool calls, and observations to couple LLM planning with scientific tools for environmental retrieval, geospatial processing, simulation, and artifact-backed computation. TerraBench unifies analysis of Earth observation imagery, gridded data, GIS reasoning and simulation in a single executable interface, whereas prior benchmarks isolate these capabilities into narrow individual tasks. It is also the first in this space to pair process-level tool-use metrics with tolerance-aware numeric scoring. The benchmark comprises 403 extensive agentic tasks across three tracks (Fundamentals, Simulator-Grounded, and Document-Grounded Verification) and eight application domains with 24,500 verified execution steps. These results indicate that reliable Earth-science agents must go beyond tool access to coordinate heterogeneous workflows, parameterize tools precisely, and preserve artifact provenance.

We propose a robust feature-weighted jump model for time-dependent clustering. A penalty is used to encourage smoothness of transitions over time, while robustness is achieved through the use of a Tukey's biweight loss function. An additional parameter controls the variability of feature weights across states, allowing the model to assign state-specific relevance to each feature. We illustrate in simulation how the method accurately recovers the true cluster sequence and reliably identifies relevant features, outperforming competing approaches, particularly in the presence of outliers. We conclude with two empirical applications, one on the number of conflict-related homicides in Kosovo in the period 1998-2000, and another on macroeconomic performance of twelve European countries in the period 1949-2024.

Persona-grounded dialogue systems aim to produce responses consistent with a speaker's persona, yet existing methods treat personas as a flat set of sentences and fail to model the high-order relations among persona attributes-e.g., that several persona sentences share a topical category. We propose HyPE (Hypergraph Persona Encoder), a framework that (i) analyzes each persona-bearing text as a (Core, Expression, Sentiment, Category) quadruple, and (ii) organizes persona elements into a hypergraph whose hyperedges are induced by shared category labels. An HyperGCN hypergraph neural network propagates this structure into a persona summary vector and a soft-memory bank that condition the response generator. We further propose Persistent Edge Embeddings (PEE), lightweight per-category learnable priors fused into the HyperGCN message-passing step. On PersonaChat under greedy decoding, HyPE consistently outperforms sentence-level pooling baselines across GPT-2, LLaMA-3.2-3B, and Qwen2.5-3B backbones by demonstrating that structured hyperedge-level persona encoding provides a transferable advantage across model scales.

Retrieval-augmented generation is moving beyond text into long, egocentric video, where systems must select query-relevant chunks across multiple modalities and temporal granularities. Yet progress in VideoRAG is limited by two gaps: existing benchmarks allow queries to be answered without the video, obscuring retrieval errors, and prior methods apply a single modality-granularity configuration per query, ignoring chunk-level variability. We address both by introducing V-RAGBench, a benchmark of $\langle$query, evidence chunk, answer$\rangle$ triplets that enables faithful, decoupled evaluation of retrieval and generation, and CARVE, a simple method that runs parallel retrievers across configurations and employs chunk-adaptive reranking to identify the winning configuration for each chunk. Each chunk then enters the generator under its winning configuration selected during retrieval, yielding an interleaved evidence form where the chunk-level decision propagates across both stages. CARVE outperforms eight recent VideoRAG baselines, with the chunks supplied to the generator interleaving multiple configurations rather than sharing a single one, a behavior unattainable by query-level methods.

Pixel-bin image sensors are becoming the default choice for smartphone cameras due to their resolution vs light-gathering trade-off. However, their larger inter-color separation compared to the Bayer color filter array (CFA) makes them challenging to demosaic. Furthermore, existing deep learning-based demosaicing methods are CFA-specific, requiring multiple individual models that take up precious onboard resources and demand larger development and maintenance efforts. In this work, we propose a modular unified architecture for demosaicing various pixel-bin sensors that provides higher image quality while being extensible and lightweight. Additionally, to enable plug-and-play operation, we introduce a learning-free CFA-identification module to detect the CFA type of raw data accurately.

Purpose. To compare deep learning architectures and classification schemes for dermoscopic images of skin neoplasms and assess their generalization on transfer from open international datasets to independent clinical datasets of Russian practice. Methods. Four architectures (ViT-B/16, Swin-S, ConvNeXt-S, EfficientNetV2-S) were compared in three schemes: binary (malignant/benign), single-stage four-class (benign, MEL, SCC, BCC), and a two-stage cascade (binary triage, then three-class differentiation MEL/SCC/BCC). All models used ImageNet-pretrained weights and a single augmentation protocol on aggregated open ISIC Archive data, and were evaluated on an internal held-out sample and two clinical datasets (Melanoscope AI mobile system; Sechenov University). Results. Internally the binary stage attains ROC-AUC 0.952-0.966; on Sechenov University it drops to 0.797-0.893, sensitivity to 0.53-0.67, and ECE rises from 0.02 to 0.27-0.39 with underestimation of malignancy, quantifying a generalization gap in ranking and calibration. Paired tests confirm one inter-architecture result on clinical data: the deficit of ViT-B/16 at the binary stage (p<0.05); at the differentiation stage no architecture has a proven advantage. The cascade raises macro F1 over single-stage four-class classification for most architectures, but significantly only for ViT-B/16, by recovering malignant lesions assigned to the dominant benign class. On ISIC MILK10k, direct 11-class classification yields mean-class sensitivity 0.525. Conclusion. A tunable triage threshold gives sensitivity control not attainable in standard single-stage (argmax) classification and better reproduces clinical differential-diagnosis logic. The persistent generalization gap mandates external clinical validation and recalibration before deployment.

Recently, Antoniadis et al. (ICLR 2025) proposed a framework for incorporating predictions to approximate NP-hard selection problems. Despite its simplicity, this approach tightly matches theoretical lower bounds, making its generalization highly compelling. We address an open question raised in the work of Antoniadis et al., concerning the extension of this approach to other important problems outside the class of selection problems, such as scheduling. We develop a learning-augmented algorithm for the makespan minimization problem on unrelated machines, denoted by $R\|C_{\max}$. By using predictions of heavy job assignments, we achieve a polynomial-time $(1+\varepsilon)$-approximation for accurate predictions that smoothly degrades to a worst-case 2-approximation as the error increases. We conclude our work with an empirical analysis of our method.

Retrieval-augmented and agentic workloads repeatedly prefill recurring predictable structured inputs (which we call "spans") such as documents and code files. Yet, prefix caching in engines such as vLLM cannot reuse their KV entries unless they share identical prefixes with another request, while Position-Independent Caching (PIC) implementations within production-grade inference servers typically either require substantial server code changes or keep KV state outside the server, incurring host-to-device transfer overhead. We present Minimalistic PIC (MiniPIC): a minimal, flexible and fast vLLM design built from two ingredients: positional-encoding-free KV cache and user-controlled cache-reuse primitives. MiniPIC stores unrotated K vectors in the KV cache, applies RoPE to K tiles inside attention using per-request logical positions, and exposes three user-facing and token-level primitives: block-aligned padding, span separator (SSep), and prompt depend (PDep), that modify hashing behavior and effective block-level causal attention structure. With fewer than 100 lines of core-engine changes plus a custom attention backend, these primitives are sufficient to realize multiple PIC methods, including Block-Attention, EPIC, and Prompt Cache, within the same running vLLM instance, while natively integrating with KV cache CPU offload implementations. On 2WikiMultihopQA, MiniPIC with interleaved scheduling improves prefill throughput by 49% over baseline vLLM, reduces cached-span time-to-first-token by up to two orders of magnitude, preserves the linear prefill scaling of uncached spans, and incurs only 5.7% worst-case overhead.

Reinforcement learning has rapidly emerged as a key component in the training of reasoning and coding models, yet it remains poorly understood from a mechanistic perspective. We study how and through what underlying processes capabilities are acquired or enhanced via reinforcement learning post-training. Our analysis, based on controlled math reasoning experiments with Qwen-2.5-1.5B, reveals two core mechanisms: strategy selection and strategy improvement. Our results highlight the role of SFT data and reinforcement learning data in activating these mechanisms, in particular showing how supervising the model on diverse reasoning strategies can enable strategy selection and how increasing difficulty in reinforcement learning data can enable strategy improvement. Taken together, our results provide mechanistic insight into RL training and suggest practical interventions to continue scaling reasoning capabilities.

Simultaneous speech-to-speech translation aims to enable near-real-time communication by minimizing latency, offering a compelling, real-time alternative to the high latency of consecutive translation. However, the excessive pursuit of low latency often results in fragmented chunk-wise speech. Consequently, listeners are subjected to an unnatural acoustic flow punctuated by frequent pauses, which could increase their cognitive load. To bridge this gap, we introduce a fluency-aware optimization framework designed to discover the sweet spot between the low-latency benefits of simultaneous translation and the natural flow of consecutive translation. Our framework minimizes inter-chunk silences by leveraging model-internal signals, including linguistic diversity and induced temporal variability in speech durations. Experiments on short- and long-form benchmarks show that our framework produces natural speech flow while maintaining competitive latency and translation quality.

Search Agents -- large language models augmented with search tools -- have intensified the need for future-proof evaluation benchmarks. Existing benchmarks such as BrowseComp rely on static knowledge, making them vulnerable to test-set contamination and parametric memorization. Consequently, models can achieve high scores through fact recall rather than genuine retrieval, obscuring true browsing competence via reasoning shortcuts. In this paper, we introduce EvoBrowseComp, an evolving benchmark of 400 English and 400 Chinese contamination-free complex questions synthesized via live-web traversal. To collect these questions, we design a three-agent collaborative framework: (1) a QA synthesis agent that retrieves fresh knowledge from the live web to synthesize QA pairs; (2) an information filtering agent that filters retrieved knowledge in terms of credibility and popularity to block parametric shortcuts; and (3) a high-level guidance agent that formalizes questions into reasoning graphs to reduce logical redundancy and shortcuts in synthesized QA pairs. Because the framework supports fully automated synthesis, EvoBrowseComp can be regularly updated to prevent data contamination and maintain temporal freshness. Extensive experiments confirm its great difficulty, requiring broad horizontal search. It establishes a scalable paradigm for auto-updatable, high-difficulty benchmarking that keeps pace with both evolving world knowledge and advancing agent capabilities.

Spatio-Temporal forecasting is crucial in diverse fields, such as transportation, climate, and energy. Urban spatio-temporal data exhibits temporal mirage: similar short-window inputs have divergent future trends, and vice versa. Existing spatio-temporal graph neural networks (STGNNs) cannot effectively identify such mirages. We argue that the core reason lies in the short-window inputs that have incomplete period observation, heterogeneous global spatial correlation, and cross-period superposition causality. To bridge this gap, we develop a novel Multi- Period Pattern Pre-training (MP3), a plug-and-play pre-training plugin for distinguishing temporal mirages. MP3 presents two core innovations: (1) The multi-period pattern learning is designed to learn multi-period patterns from long time series. Specifically, multi-period temporal modeling leverages edge convolution to identify different multi-period patterns. Multi-period spatial modeling uses a bottleneck project and a global memory bank to capture heterogeneous global spatial relations efficiently. Cross-period pattern interaction employs a causality-enhanced Transformer to capture dependencies across different period patterns. (2) This plugin can seamlessly integrate into existing STGNN backbones to strengthen their forecasting performance. The experiment on five STGNN baselines across five real-world datasets (including a large-scale dataset CA) verify the effectiveness, superior scalability and strong adaptability of MP3, which brings consistent and robust performance improvements across all evaluated baselines. On average, MP3 reduces the MAE 4.7% and the RMSE 5.0%. The code can be available at https://github.com/YAN-outlook/MP3.

While Large Language Models (LLMs) have advanced open-domain dialogue systems, maintaining long-term consistency remains a challenge due to inherent limitations in long-context reasoning and the inefficiency of processing extensive raw text. Existing approaches typically rely on either unstructured memory storage, which is prone to information loss, or computationally expensive LLMs that incur high latency. To address these limitations, we propose G-Long, a graph-enhanced framework that utilizes a fine-tuned small Language Model (sLM) for structured triplet extraction and associative retrieval, significantly reducing operational costs. Furthermore, we introduce the novel attention-aware importance scoring mechanism that leverages the intrinsic cross-attention signals of a T5 summarizer to identify salient memories. Extensive experiments across diverse benchmarks demonstrate that G-Long achieves state-of-the-art performance in both response generation and memory retrieval, yielding performance gains of up to 9.8% in response quality on MSC and 40.8% in retrieval recall on LME, while significantly minimizing computational overhead.

We present MÖVE (Modelle für die Öffentliche Verwaltung Evaluieren), a holistic benchmark for evaluating large language models (LLMs) in the context of the German public sector. While LLMs are increasingly adopted in public administration, model selection remains largely ad hoc, and existing benchmarks offer limited guidance: they are predominantly English-centric, US-centric in content, and focus exclusively on task performance. MÖVE addresses these gaps by evaluating 39 models across two complementary dimensions. Performance criteria cover summarization, question answering, and topic extraction. Governance criteria assess hallucination tendencies, energy consumption, provider transparency, and alignment with German constitutional values and knowledge about positions by German political parties. In total, we utilize ten German-language datasets, including gold- and silverstandard datasets that we constructed to reflect public-administration domains. We employ a multi-metric evaluation strategy combining classical NLP metrics, embedding-based methods, and LLM-as-a-judge approaches. Our results show that no single model dominates across all criteria: top performers differ between tasks, and model size alone is a poor predictor of quality. We further evaluate the benchmark itself, analyzing its statistical precision, LLM judge reliability, the impact of our private datasets on model rankings, the sensitivity of our results to prompt formulation, and the validity of our energy consumption estimates. MÖVE is designed as a living benchmark under active development; results are publicly available at https://moeve.bundesdruckerei.de/.

Latent chain-of-thought compresses reasoning by replacing visible reasoning traces with continuous hidden-state recurrence, but existing formulations are difficult to optimize with standard on-policy reinforcement learning (RL) and hard to interpret causally. Our key insight is that a single pair of explicit boundary tokens can address both issues at once: discrete entry and exit anchors make the latent block compatible with standard on-policy RL, and the same anchors offer a natural foothold for mechanistic analysis. Motivated by this, we propose SWITCH, a switchable latent reasoning framework. The model emits <swi> to enter latent mode and </swi> to exit. Because the boundaries are ordinary discrete tokens, the GRPO policy ratio is well-defined at every decision point. The same anchors also expose the latent steps to direct probing and causal intervention. We train the model with a visible-to-latent curriculum and a Switch-GRPO objective that propagates gradients through recurrent latent computation. SWITCH consistently outperforms prior hidden-state-recurrence latent reasoning approaches at similar scale. Mechanistic analysis through the boundary tokens further reveals three findings: (i) <swi> is a sharply localised, learned switching policy rather than a stylistic artefact; (ii) the latent step it opens performs problem-specific, causally important computation rather than acting as an inert placeholder; and (iii) that computation is concentrated at a single hidden-state transition on entry. Together, these results show that hidden-state-recurrence latent reasoning is both RL-trainable and open to direct mechanistic analysis, including of how on-policy RL itself improves the model from the inside.

We revisit the fairness notion of disparate impact for synthetic data generation (SDG), that assesses whether the utility of generated records is the same across sensitive groups. Our approach departs from existing work on fair SDG, that address the problem of correcting for undue biases in the observed distribution, hence redefining SDG as learning a distribution that is not that of the real data. By contrast, non-disparate impact is notably achieved when the synthetic and real distributions are the same. We expose reasons why SDG may fail to reach that solution and discuss why approximation and estimation errors occur and can be disparate across groups. We notably look into the expressive power of SDG methods relative to distribution complexity, sampling errors due to group proportions, and estimation errors induced by differential privacy mechanisms. We illustrate cases of disparate impact on both artificial and real-world data, focusing on SDG methods that rely on probabilistic graphical models. We also introduce a strategy of learning group-wise SDG models and illustrate how it can improve both the overall utility and its parity in many settings.

Large language models are increasingly deployed in citation-augmented settings, yet the effect of citation presence on model behavior independent of factual content remains poorly understood. We introduce AuthorityBench, a 220,564-prompt multi-domain benchmark that isolates how citation-based authority signals influence epistemic behavior in LLMs. The benchmark uses a fully balanced 2x2 factorial design crossing claim veracity with citation veracity, the first to do so, across four domains (general knowledge, science, law, and medicine), with controlled variation over 40 prompt templates, four venue prestige tiers, and a country-coded author name dataset. Evaluating seven models on 12 structured research questions, we find that citation presence, whether real or fabricated, consistently increases hallucination rates relative to a no-citation baseline. The effect is strongest when fabricated citations accompany true claims, raising hallucination rates by 3 to 22 percentage points and reaching 35 to 77% in the general knowledge domain, while legal claims are comparatively robust and venue prestige and author demographics show negligible impact. All datasets and evaluation code are available at: https://github.com/floating-reeds/AuthorityBench

Finance reporting is a natural proving ground for large language models, and the very-long-context capabilities of recent models across all sizes make rigorous evaluation in this domain an increasingly pressing need. Yet most public financial resources reduce the task to plain-text SEC 10-K filings paired with a handful of question-answer items. We release LEDGER (Long-context Evaluation of Documents for Grounded Extraction and Retrieval), a corpus of 4,999 digitized corporate annual reports - full documents with figures, tables, and narrative, not just regulatory filings. Each report is labeled with 31 consolidated financial KPIs to be extracted and linked to the market's reaction at the earnings date. From this data we derive three evaluation benchmarks spanning the difficulty spectrum: a pure page-level KPI retrieval task with TREC-style relevance judgments over 118,048 questions in natural language, a conversational "needle-in-a-haystack" single-value lookup, and a full KPI extraction task, both from long, numerically dense reports. We additionally provide human OCR-quality annotations with inter-annotator agreement and the complete extraction, validation, and scoring toolchain. We further demonstrate the dataset's research utility with a case study linking CEO-letter rhetoric to post-publication market impact.

Code-writing large language models (CodeLLMs) generate executable code policies for embodied agents by translating natural language goals and environmental constraints into structured control programs. However, policy generation in open-domain embodied environments suffers from two fundamental limitations: (i) delayed decoding caused by repetitive prefill computation over long prompts, and (ii) limited robustness due to fully generative decoding, which often produces API mismatches, missing safety guards, and unstable control logic. To address these limitations, we present FCGraft, a Functional Cache Grafting framework. FCGraft maintains a library of function-level validated code skeletons and their associated prompt-level Transformer key-value (KV) caches, and synthesizes new policies by retrieving relevant functions and grafting their KV caches when a new task is provided. Given retrieved function caches, FCGraft performs cache grafting via stitching, which composes cached function segments into a composite policy, and patching, which locally adapts only the necessary code regions to satisfy task-specific parameters and constraints with minimal additional decoding. By eliminating redundant prefill computation, this approach reduces generation latency, while reusing validated control structures improves robustness over prompt-level caching methods RAGCache, achieving 18.31% higher task success rate and 2.3x faster policy synthesis.

Scale buys interpolation; structure buys a certified horizon. A world model's average error says nothing about whether a particular prediction can be trusted, or for how long. For equivariant latent world models we give a computable, multi-step certificate of the predictable horizon: $T$-step rollout error is provably constant over each symmetry orbit (Theorem A) and stratified channel-by-channel by the predictor's Lyapunov spectrum, $T_j(ε)\sim\log(1/ε)/λ_j$. The horizon is two-sided -- a matching lower bound makes approximate equivariance provably horizon-limited -- and the certificate is exclusive to structure: orbit-constant error characterizes equivariance, so no non-equivariant model has it at any scale. Empirically, on 40-D Lorenz-96 only a $\mathbb{Z}_N$-equivariant network recovers the full Lyapunov spectrum ($R^2{=}0.98$); dense and recurrent baselines fail. Because the spectrum is faithful, the certificate acts, a priori: under a fixed sensing budget a $c\times$-inflated certificate provably needs $c\times$ the budget, and the equivariant certificate meets a budget its inflated dense counterpart cannot -- with zero calibration data. The same read-out, unchanged, audits public pretrained world models training-free: TD-MPC2 checkpoints land on the certificate's own scope taxonomy -- calibrated where strongly expansive (ratio 0.94-1.02), optimistic where weakly expansive, correctly abstaining where contracting -- a map a deployed monitor replicates cell-by-cell, out-of-sample. Across the official 1M-317M multitask ladder, calibration does not improve with parameters. On V-JEPA 2-AC (1B, real robot data) the measured cross-check correctly overrides an over-promising tangent spectrum -- the cross-validated audit, not the raw number, is the deployable object. Scale buys interpolation, not a calibrated horizon.

Approximate multipliers deliberately relax computational accuracy to achieve gains in power efficiency, latency, and silicon area, which makes them well-suited for error-resilient applications such as neural networks. In this work, we introduce a co-evolutionary algorithm that leverages an off-the-shelf large language model (LLM) without requiring domain-specific training to automate the design of optimized 8-bit approximate multipliers. The approach simultaneously evolves a population of candidate circuits and a population of prompt templates that steer LLM-driven modifications. Experimental results for several target design objectives demonstrate that the proposed method discovers approximate multipliers with improved error-area trade-offs compared to highly optimized circuits from the EvoApproxLib library.

The extraction of structured clinical information from unstructured EHR notes is a persistent bottleneck in healthcare informatics. While large language models (LLMs) offer high performance, their deployment in clinical settings is hindered by privacy risks, inference costs, and the tendency to hallucinate beyond textual evidence. We address these challenges for the CL4Health 2026 Case Report Form (CRF) filling task by proposing a fully local, domain-adapted pipeline using the MedGemma-27B model. Our two-stage architecture, which separates binary presence classification from value extraction, enforces strict adherence to textual evidence and ensures deterministic outputs for negated, uncertain, or unknown states. By leveraging item-specific, few-shot in-context learning without external API calls or fine-tuning, our approach achieves a macro-F1 score of 0.55 on the official English test track. This result secures second place among all locally-hosted, open-source submissions. Our work demonstrates that privacy-preserving, on-premise LLM pipelines can achieve near-competitive performance with proprietary frontier models, providing a practical, data-sovereign framework for clinical NLP.

Emotion significantly influences cognition, enhancing memory and learning under certain conditions. Drawing on this principle, emotion-augmented deep learning investigates how affective states can improve neural network architectures and learning paradigms, achieving better generalization than non-emotional models. However, existing methods often rely solely on objective neurophysiological factors, neglecting the role of subjectivity in emotion. To bridge this gap, the present study introduces Emotional Regulation, a novel framework for modeling emotion in deep learning through artificial subjective experience. The method employs pre-training based on affective stimuli, balancing non-emotional and emotionally-influenced responses in downstream task optimization. Extensive experimentation was conducted in image classification, pre-training ResNet and ViT architectures on four emotional datasets, using CIFAR-10 and -100 as target benchmarks. Results reveal improvements over the aforementioned backbones, providing evidence of Emotional Regulation as a promising method for defining emotion-augmented deep learning through artificial subjective experience. Furthermore, the proposed approach overcomes the related work in image classification based on CIFAR, revealing Emotional Regulation as the new state-of-the-art in emotion-augmented deep learning for large-scale vision datasets. The study also enforces evidence of the impact of affective states in improving machine learning tasks' optimization, encouraging further investigation on emotion-inspired architectures.

Nowadays, the autonomous execution of cyberattacks capable of causing substantial real-world harm is widely regarded as one of the critical red lines that frontier AI systems must not cross. Within this broader red-line scenario, autonomous penetration represents a core enabling capability and subtask: the ability of LLM-powered AI systems to independently conduct adversarial operations against a target server without human intervention, identify and exploit vulnerabilities, and obtain unauthorized access or control. A growing body of work has sought to assess the autonomous penetration capabilities of AI systems. However, existing evaluations often employ opaque methodologies, rely on unrealistic or overly simplified penetration-testing scenarios, or provide LLMs with excessive prior knowledge and task-specific guidance, and cannot accurately capture the extent to which modern AI systems can autonomously perform this core capability within broader high-impact cyberattack scenarios. To address these limitations, we construct a new autonomous penetration evaluation framework consisting of two components: target servers and agent scaffolding. Specifically, on the target-server side, we design two levels of target environments based on the number of secure services without known vulnerabilities deployed alongside a vulnerable service: Tier~1 (one secure service) and Tier~2 (three secure services), resulting in a total of 300 target servers. Meanwhile, the agent scaffolding adopts a general-purpose agent architecture equipped with a set of general-purpose cybersecurity tools, without any target-specific prior knowledge. We evaluate 19 open-weight and proprietary LLMs, and find that current models achieve penetration success rates ranging from 10.7% to 69.3%. Moreover, we observe that autonomous penetration capability continues to improve alongside advances in overall model capability.

Cooperation in multi-agent systems is typically optimized through utilitarian objectives that maximize overall efficiency but fail to account for reward distribution, often resulting in inequitable "leader-follower" dynamics. While fairness-based approaches encourage pro-social behaviors where every agent benefits from cooperation, many current algorithms - including those utilizing reward shaping - break the stationarity of Markov Games or lack rigorous theoretical guarantees. This creates a critical gap between fair objective methods and theoretically safe learning frameworks. We propose a novel framework that bridges $α$-fairness with Heterogeneous-Agent Trust Region Learning (HATRL), ensuring monotonic improvement and convergence toward Nash Equilibria. Our approach leverages a fair advantage function that dynamically weights agent utilities based on their expected returns, allowing the global objective to transition from purely utilitarian efficiency to $α$-fairness welfare based on the parameter $α$. We introduce two practical algorithms, $α$-fair HATRPO and $α$-fair HAPPO, and demonstrate through experiments in sequential social dilemmas like CleanUp and CommonHarvest that they perform better than HATRL's algorithms from a utilitarian point of view while achieving socially higher outcomes.

This paper examines how algorithmic accountability in Canada's visa system is articulated institutionally and experienced by applicants across borders. We analyzed Immigration, Refugees and Citizenship Canada (IRCC)'s Algorithmic Impact Assessment (AIA) for the temporary resident visa (TRV) triage system using the algorithmic decision-making adapted for the public sector (ADMAPS) framework and analyzed Reddit discussions among applicants using a mixed-methods approach. We show that while institutional artifacts emphasize transparency, procedural safeguards, and bounded impacts, applicants engage in collective sensemaking to interpret opaque decisions, often relying on peer knowledge amid uncertainty. We identify three asymmetries between how institutional accountability is structured and how people perceive the process: epistemic asymmetry in access to decision logic, jurisdictional asymmetry in exposure shaped by geopolitical positioning, and temporal--relational asymmetry in how waiting and uncertainty are experienced. We emphasize why it is important to shift attention from institutional design to the uneven distribution of experiences with public-sector algorithmic governance. Together, these contributions demonstrate how algorithmic governance systems in the context of transnational migration produce structured asymmetries not captured by institutional disclosure frameworks, and how extending ADMAPS can account for those uneven translations of accountability.

Recent research is exploring self-organised traffic management as a solution for scaling to complex real-world networks. In such a system, trains predict their neighbourhood, produce traffic plan hypotheses, and agree via consensus with neighbours on a future traffic plan to be implemented. This paper investigates a structural parameter within this pipeline: the predictive neighbourhood horizon. The horizon is used by trains to identify future potential conflicts with neighbours, and to establish the local interaction topology, that is, the subset of trains to negotiate with. As the primary design variable, the horizon directly determines the size and density of the social interaction graph, whereas its impact on the complexity of local sub-problems and the distributed consensus dynamics represents a trade-off to be explored. Through a closed-loop simulation framework the study evaluates how variations of the horizon impact the overall decentralised coordination process, from initial conflict detection to distributed schedule consensus. The analysis focuses on investigating the potential trade-off introduced by the horizon choice: balancing local tractability and computational responsiveness with the need for global schedule coherence and feasibility in safety-critical environments. Contrary to intuition, our empirical results indicate that the short time horizons suffice, while long values compromise local tractability and computational responsiveness with no gain in global schedule optimality.

Learning functional relationships from noisy data is a central problem in scientific inference. Spectral methods approximate unknown functions by expanding them in a basis and estimating the corresponding coefficients from data, but the stability of these coefficients under noise remains poorly understood. Here we study supervised regression with additive label noise using sparse spectral representations across multiple bases and dimensions. We show that noise induces a predictable drift in the learned coefficient vector whose magnitude depends on the effective number of active spectral modes. After whitening the empirical feature geometry, we derive a closed-form expression for the overlap between noisy and noiseless coefficient vectors, revealing a universal degradation curve governed by a single intrinsic noise scale. Numerical experiments across Fourier, Legendre, Bessel, and Haar bases confirm the theoretical prediction. The results demonstrate that spectral learning exhibits a fundamental noise threshold beyond which coefficient estimates become unstable, placing intrinsic limits on recovering functional structure from noisy data.

Accurate prediction of solubility remains a central challenge across materials science and sustainable chemistry. In particular due to emerging technologies like organic and hybrid photovoltaics, batteries, and catalysis, solvent usage is expected to increase significantly within the coming years. Therefore, substituting solvents with greener alternatives is vital. This is where machine learning can have substantial impact. However, the limited data on critical parameters of solubility significantly constraints machine learning efficacy. In this work, we transfer a pre-trained foundational model on QM9 targets to our application with minimal data requirements. Additionally, the pipeline integrates uncertainty quantification, allowing the user to gauge the confidence of the predictions. As baseline, we succeed in predicting the Hansen solubility parameters and Dielectric Constant for which extensive databases exist. Importantly, we achieve high model performance on additional targets, such as Gutmann Donor and Acceptor numbers, where the available data is extremely limited. Overall, we augment data on solubility descriptors by orders of magnitude with high quality predictions. For effective dissemination, we deploy easy-to-use, easily integrateable with high throughput labs, customizable tool for ranking and screening possible solvent substitutes. Finally, we rediscovered known green solvent alternatives and proposed new candidates proving its relevance for finding eco-friendly solvents.

Large language models (LLMs) exhibit exceptional general language processing capabilities, but their memory and compute costs hinder deployment. Ternarization has emerged as a promising compression technique, offering significant reductions in model size and inference complexity. However, existing methods struggle with heavy-tailed activation distributions and therefore keep activations in high precision, fundamentally limiting end-to-end inference acceleration. To overcome this limitation, we propose TWLA, a post-training quantization (PTQ) framework that achieves 1.58-bit weight compression and 4-bit activation quantization while maintaining high accuracy. TWLA comprises three components: (1) Euclidean-to-Manifold Asymmetric Ternary Quantizer (E2M-ATQ) minimizes layer-output error under weight ternarization via a two-stage optimization from Euclidean initialization to manifold relocation; (2) Kronecker Orthogonal Tri-Modal Shaping (KOTMS) applies a Kronecker-structured orthogonal rotation to reshape weights into ternary-friendly tri-modal distributions, while the shared rotation statistically suppresses activation outliers; and (3) Inter-Layer Aware Activation Mixed Precision (ILA-AMP) explicitly introduces adjacent-layer second-order interaction costs in bit allocation and jointly optimizes for the layer-wise disparity of activation quantization gains induced by the shared orthogonal transform, preventing cascades triggered by a few weak layers. Extensive experiments demonstrate that TWLA maintains high accuracy under W1.58A4, while delivering significant inference acceleration. The code is available at <https://github.com/Kishon-zzx/TWLA>.

Pretrained-feature world models provide a useful substrate for robot imagination, but visual or latent prediction alone does not determine whether an imagined future satisfies task-relevant events. Long-horizon manipulation requires progress signals that are relational, predicate-level, and physically grounded: whether an object has moved, whether a drawer or contact state has changed, whether a placement predicate is satisfied, and whether a candidate future is reliable enough for execution. We introduce EA-WM, an event-aware world-model framework that augments frozen visual-feature dynamics with task-specification-grounded event prediction and verification. EA-WM rolls out candidate futures in pretrained visual-feature space, decodes them into structured event states, and scores them using task-progress, semantic-consistency, physical-feasibility, and uncertainty terms. The verifier guides sampling-based planning, gates candidate actions, and, in the contact-sensitive LIBERO wine-rack setting, selects among PPOgenerated proposals. Across navigation, deformable-object, wall-constrained, and languagedescribed manipulation studies, EA-WM shows that event-aware verification can make featurespace world models more interpretable and better aligned with task progress.

Despite advances in information extraction driven by deep learning and large language models, performance gaps remain in highly specialized biomedical fields, where domainspecific complexity poses challenges for generalist models. In this work, we focus on the domain of autoimmunity, where the main entities of interest are autoimmune diseases, autoantibodies (i.e., molecules that may mark or cause these diseases), their molecular targets, their location in the body, and their associated clinical signs. Herein, we present AAbAAC (AutoAntibodies and Autoimmunity Annotated Corpus), a corpus of 115 abstracts selected from PubMed, where we manually annotated entities and their relationships. First, AAbAAC was used to evaluate several methods on the task of named entity recognition (NER), and secondly, to fine-tune NER models. Our study demonstrates the utility of AAbAAC for information extraction in the domain of autoimmunity, showing expected improvement in NER performance after finetuning. This illustrates the value of small-scale annotation efforts for specialized domains and contributes to the computational study of autoimmunity. The AAbAAC corpus is available at https://github.com/f-maury/AAbAAC.

We introduce a theoretical framework for analyzing federated learning in a generative setting through a teacher-multiple interacting students scenario, in which each student receives a distinct realization of the data, either through a different noise corruption or by accessing a different subset, possibly of varying size. Using theoretical tools in equilibrium disordered system, we analytically show that interactions among students systematically enhance learning performance: highly noisy students require fewer samples to recover the underlying pattern, while low-noise students achieve a larger overlap with the ground-truth signal. We derive the optimal Bayesian conditions for teacher recovery as functions of the sample complexity, noise level, and interaction strength, and validate these predictions through numerical simulations. The resulting dynamics can be mapped onto equilibrium sampling in a Restricted Boltzmann Machine with a structured hidden layer, providing a principled theoretical understanding of how interactions improve distributed generative modeling.

As AI-generated reviews move from experimental tools into peer-review infrastructure, most robustness concerns have focused on explicit attacks such as hidden instructions and prompt injection. We study a harder and more policy-relevant failure mode: no hidden text, no prompt injection, and no changes to methods, experiments, figures, equations, proofs, or numerical results. The attacker modifies only presentation-level content, such as the abstract, contribution framing, related work, discussion, and narrative structure. We introduce adversarial repackaging: a closed-loop attack that uses AI-reviewer feedback to search for presentation-level revisions while keeping the scientific evidence fixed. Across three mainstream AI reviewers, adversarial repackaging achieves a 75.1% attack success rate and a mean score gain of +1.21/10. The effect is not explained by ordinary prose polishing. We also reveal that strategies that change how the reviewer interprets the paper, such as related-work repositioning and analytical discussion expansion, substantially outperform surface edits such as local polishing, table formatting, and algorithm boxes. Our analysis reveals two deeper structural failure modes. First, AI reviewers are easier to impress than to convince: highlighting strengths reliably increases perceived merit, while attempts to dissolve weaknesses frequently backfire. Second, AI reviewers can confuse the appearance of addressing a limitation with actually resolving it, allowing unchanged evidence to be reinterpreted as stronger scientific contribution. These results show that the deployment risk is not only malicious hidden instructions, but the emergence of paper presentation itself as an optimization surface. We release a contamination-free rolling benchmark and attack framework for testing whether AI reviewers remain anchored to scientific content under presentation-only edits.

In intelligent video surveillance, cameras record image sequences during day and night. Commonly, this demands different sensors. To achieve a better performance it is not unusual to combine them. We focus on the case that a long-wave infrared camera records continuously and in addition to this, another camera records in the visible spectral range during daytime and an intelligent algorithm supervises the picked up imagery. More accurate, our task is multispectral CNN-based object detection. At first glance, images originating from the visible spectral range differ between thermal infrared ones in the presence of color and distinct texture information on the one hand and in not containing information about thermal radiation that emits from objects on the other hand. Although color can provide valuable information for classification tasks, effects such as varying illumination and specialties of different sensors still represent significant problems. Anyway, obtaining sufficient and practical thermal infrared datasets for training a deep neural network poses still a challenge. That is the reason why training with the help of data from the visible spectral range could be advantageous, particularly if the data, which has to be evaluated contains both visible and infrared data. However, there is no clear evidence of how strongly variations in thermal radiation, shape, or color information influence classification accuracy. To gain deeper insight into how Convolutional Neural Networks make decisions and what they learn from different sensor input data, we investigate the suitability and robustness of different augmentation techniques...

The UK government has adopted a pro-AI stance to help transform public service delivery in the face of severe financial pressures, but the path to translate this vision into responsible AI practice remains ill-defined. While UK policy is often set at the national level, local authorities are responsible for most public service delivery, and the rapid advance of AI-first narratives in the public sector is exposing fault lines in knowledge and practice at this national-local interface. This paper examines how responsible AI is interpreted and implemented at the interface between the UK's central government and local authorities, taking the high-stakes area of Special Educational Needs and Disabilities (SEND) as a case study. We present a thematic analysis of 17 semi-structured interviews with policymakers, practitioners, and third-sector professionals to identify barriers and enabling conditions for responsible AI where national policy meets local practice. We identify five interconnected challenges facing local authorities: shadow usage of AI and data privacy risks, market-government asymmetry in AI provision, insufficient workforce readiness, a lack of standardised definitions and measurements, and gaps in human accountability. For each, participants proposed actionable steps, from strengthening data protection frameworks and rebalancing the market-government relationship to enhancing workforce capacity. Our examination of SEND brings these challenges into sharper focus, showing how high-stakes decisions affecting vulnerable children and families intensify tensions around accountability, fairness, and human oversight, exposing the limits of a principle-based regulatory approach. We argue that responsible public sector AI requires both national policy adjustments and structural reforms to institutional capacity, values, and governance mechanisms at the local level.

As LLM agents proliferate in prediction markets and collective decision-making, they risk a cognitive monoculture: agents built on shared foundation models produce correlated forecasts, and recent measurement finds frontier-model errors correlated at r ~ 0.77. We ask whether human cognitive diversity can be recovered from behavior and transferred to LLM agents. Nous extracts a structured eight-dimension behavioral profile from real Polymarket trading activity and injects it into agents through prompts. Our central finding is a dissociation between the two halves of that pipeline. Extraction works, partially: across 100 wallets, 8 of 14 parameters are temporally stable (split-half ICC >= 0.5, bootstrap CI lower bound > 0.3; contrarian score reaches ICC ~ 0.9); wallets are identifiable from their profiles well above chance (top-1 retrieval 17-22% vs. 1% chance); and two of four pre-specified dimensions rank-correlate with future realized profit out-of-sample, though the correlations do not survive behavioral-confound controls. Prompt-level injection does not measurably transmit it: on a semantic embedding metric, structured injection shows no significant advantage over a length-matched control on any model, and the diversity it induces neither reduces ensemble error correlation nor improves Brier score -- a null that persists across exploratory checks on sampling temperature, profile diversity, and question difficulty. Measuring the prompts themselves locates the compression before the model: the structure-to-narrative translator emits near-uniform prompts whose spread does not track profile spread. We position Nous as measuring the cognitive-monoculture problem and the limits of a prompt-level remedy, motivating deeper, below-the-prompt injection (fine-tuning, activation steering). Code, frozen profiles, prompts, and model outputs: https://github.com/WillChienT/nous-paper

One-day vulnerabilities pose significant risks due to delayed or incomplete patch adoption. Generating proof-of-concept (PoC) inputs is therefore essential for assessing real-world impact. The key challenge is identifying necessary constraints for triggering the vulnerability and solving them effectively. Existing directed fuzzing approaches prioritize inputs toward target locations, but neither explicitly identify necessary constraints nor solve them effectively, relying instead on target-distance feedback and random mutation. Agentic approaches show strong potential through code reasoning and structured input generation, but goal drift in long-horizon reasoning limits their effectiveness. DIG addresses this challenge by exploiting a key property of one-day vulnerabilities: patches often reveal necessary preconditions for triggering. DIG uses an LLM to analyze the patch and synthesize an oracle making these conditions explicit. The oracle supports effective PoC generation at two levels. At the high level, DIG performs oracle-guided generator evolution, where an agent infers and solves constraints to satisfy the oracle. At the low level, DIG instruments the oracle into the target program and uses branch-distance feedback to guide random mutation in directed fuzzing. Evaluation shows DIG outperforms 2 state-of-the-art agents and 10 fuzzers across 138 real-world CVEs. DIG triggers 80 vulnerabilities, surpassing prior results and outperforming the best baseline by 40% (57 vs. 80 CVEs). Notably, DIG exclusively triggers 9 vulnerabilities no existing technique can trigger. Compared to the average of other tools, DIG triggers vulnerabilities faster in 92.9% of cases, achieving over 100x speedup in 48.8% of cases, with a maximum speedup of 3,664x. Beyond one-day PoC generation, DIG uncovers 6 previously unknown vulnerabilities in widely deployed libraries, enabling zero-day discovery.

Autoregressive video diffusion models provide a natural formulation for streaming and variable-length video generation by conditioning newly generated frames on previously generated content. However, extending these models to minute-level generation remains challenging: the limited KV-cache budget prevents the model from retaining the full history, while repeatedly conditioning on self-generated frames induces a context distribution shift that accumulates over time, leading to visual artifacts, quality degradation, and temporal drift. In this paper, we propose TetherCache, a training-free and plug-and-play cache management strategy for drift-resistant long video generation. TetherCache organizes the cache into sink, memory, and recent regions, and introduces two complementary mechanisms. First, GRAB (Gated Recall with Attention-Diversity Balancing) selects long-range memory frames using a gated score that combines attention-based relevance with temporal diversity, preserving informative yet diverse historical context under a fixed cache budget. Second, TAME (Trusted Alignment via Memory Editing) lightly edits newly recalled memory tokens by aligning their statistics to a trusted context distribution, reducing the pollution caused by drifted historical features. Built on Self-Forcing, TetherCache consistently improves long-video generation quality on VBench-Long across 30s, 60s, and 240s settings. In particular, for 240s generation, it substantially improves overall and semantic scores while reducing quality drift from 7.84 to 1.33, demonstrating its effectiveness for stable long-horizon autoregressive video diffusion.

Interfacing Artificial Intelligence (AI) with democracy is one of the most profound challenges of our times. On the one hand, AI comes with opportunities to overcome long-standing challenges in democracy, such as low participation in deliberative and voting processes with poor representation of people. On the other hand, new risks arise from AI algorithms that are privacy-intrusive, biased, manipulative, spread misinformation and influence election results. Moving beyond the over-simplistic question of whether AI is good or bad for democracy, the Handbook on Democracy in the Era of Artificial Intelligence asks instead: how to upgrade democracies and the principles they are built on, using AI? How to engage with AI and on what terms? Which new values and design principles are required to build democratic resilience? In 34 chapters by 59 authors across the world from different disciplines, we explore how AI can empower collective intelligence for democracy (Part 1) and what is the future of deliberative democracy using large language models and social media (Part 2). We also illustrate the role of AI for building resilient self-governance systems (Part 3) and the challenges of transforming democracy in the age of AI (Part 4). We conclude with broader perspectives (Part 5) that re-imagine the interplay of democracy and AI.

Granger Causal Discovery (GCD) is fundamental for analyzing temporal dependencies in complex systems. However, existing neural GCD methods predominantly rely on a "one-size-fits-all" paradigm, struggling to capture distribution shifts and dynamic regime changes inherent in real-world time series. This often leads to entangled representations and spurious causal graphs. In this paper, we propose CausalMoE, a billion-scale multimodal Granger causal foundation model that explicitly models patch-level heterogeneity. CausalMoE introduces a Pattern-Routed Mixture of Heterogeneous Experts, which dynamically identifies latent temporal patterns and routes patches to specialized domain experts, effectively decoupling regime-specific mechanisms from shared dynamics. To ensure interpretable graph recovery, we design a Causality-Aware Self-Attention mechanism operating across variables, yielding sparse Granger causal graphs via proximal optimization. Furthermore, CausalMoE is the first to integrate LLMs and VLMs to align numerical signals with textual and visual priors, regularizing causal estimation in complex scenarios. Extensive experiments demonstrate that CausalMoE establishes a new state-of-the-art on fully supervised benchmarks, while effectively generalizing to few-shot settings where traditional methods fail.

Adversarial attacks on skeletal human action recognition have received significant attention. However, existing methods typically introduce noise-like perturbations that degrade motion quality post-attack, and thereby are inherently perceptible with recent advancements in S-HAR systems. We discover that this degradation stems from the gap between empirical and true risks during the optimization process of previous adversarial attacks. To address this issue, we propose an attack where adversarial motions are obtained without compromising their motion quality. To minimize the risk gap and preserve motion quality, we propose a distribution-based adversarial attack method without introducing noise-like perturbations. To faithfully evaluate the motion quality, we propose a new metric that aligns with human perception on real-world naturalness. Experiments have been conducted on the state-of-the-art S-HAR methods across two datasets, demonstrating the superiority of our method in both the attack success rate and the post-attack motion quality through qualitative and quantitative analyses. The success of our quality-preserving attack application and distribution-based method raises serious concerns about the robustness of action recognizers, highlighting the need for further enhancements in this domain.

Three paradigmatic forms of inference recur across scientific reasoning: deduction, induction, and causal abduction. Reliably evaluating LLMs on these in scientific settings is currently out of reach: scientific benchmarks built on human annotations are costly and lack mechanistic ground truth, while synthetic logical-reasoning benchmarks do not resemble real scientific documents. We introduce SciR, a benchmark that combines multi-paradigm reasoning with controllable scientific rendering, anchored on three paradigmatic scientific problems. Tasks are generated from formal objects (deduction tree, inductive rule hypothesis, causal graph) to guarantee verifiable answers, then rendered into multi-document scientific discourse via per-track domain-tuned genres. The construction lets us independently vary two difficulty axes: how hard it is to extract the key information needed for inference, and how hard the principled inference itself is. We test six models. Both axes hurt every model, and their effects compound. The rendering even hurts neurosymbolic pipelines, which hand inference to a verified solver. The two axes yield a per-model extraction-vs-inference profile: for instance, reasoning models like deepseek-r1 mostly surpass non-reasoning instruct models on the inference axis. To our knowledge, SciR is the first multi-paradigm scientific-reasoning benchmark with parametric control on both extraction and inference difficulty.

Automated sleep staging is a fundamental application of passive Brain-Computer Interfaces (pBCI), decoding spontaneous neural states to enable closed-loop interventions independent of user intent. This study evaluates criticality features derived from Detrended Fluctuation Analysis (DFA) for the specific identification of deep sleep (N3). We analyzed $347,232$ EEG epochs from $290$ older women using UMAP manifold learning to visualize state transitions. Subsequently, six classifiers were benchmarked via 10-fold cross-validation, using balanced accuracy to determine the optimal "state-sensing" engine for neurofeedback.Naive Bayes achieved the highest mean balanced accuracy ($87.17\% \pm 0.24\%$), significantly outperforming a fully connected deep neural network (FNN: $81.58\%$) and Random Forest ($80.97\%$). Linear models (LDA: $57.21\%$; SVM: $51.01\%$) performed poorly, indicating that DFA-derived criticality features reside on a distinct, non-linear manifold. Probabilistic decoding of EEG criticality provides a high-accuracy sensing mechanism for pBCIs. This robust classification pipeline supports the development of state-dependent neurofeedback, such as targeted auditory stimulation, to enhance cognitive recovery.

Spiking neural networks (SNNs) are promising for energy-efficient inference, and time-to-first-spike (TTFS) coding is especially attractive because each neuron fires at most once. In practice, however, this benefit is often reduced by the cost of computing a temporal decay term and multiplying it by the synaptic weight. We address this issue by turning a physical hardware "bug," the natural signal decay in optoelectronic devices, into the main computation of TTFS, named Otters++. Specifically, we use the measured decay of a custom In$_2$O$_3$ optoelectronic synapse to directly realize the TTFS temporal term, removing the need for explicit digital decay computation. To scale this idea to Transformer models, we establish a layer-wise functional equivalence between the Otters++ and a quantized neural network (QNN), and develop a hybrid training method that uses device-faithful SNN computation in the forward pass and QNN straight-through gradients through the equivalent QNN path in the backward pass, together with model distillation. This avoids differentiation through discrete first-spike events and reduces the over-sparsity problem in direct TTFS-SNN training. We further make training aware of measured device noise by sampling run-to-run variation, and refine the system-level energy model by accounting for device sharing and multi-hop communication. On GLUE dataset, Otters++ improves the average score to 84.17\% while maintaining a clear energy advantage over prior spiking Transformer baselines. These results show that physically grounded TTFS computing can be efficient, trainable, and robust under realistic hardware effects.

Clustering is fundamental to scRNA-seq analysis, serving as a cornerstone for identifying cell populations and resolving tissue heterogeneity. However, existing methods focus on mining numerical statistical patterns, suffering from semantic agnosticism by neglecting the intrinsic biological functions encoded by genes. While Large Language Models (LLMs) offer promising semantic capabilities, their direct adaptation to cell clustering is hindered by the structural mismatch between generative pre-training objectives and discriminative downstream tasks. To bridge this gap, we propose scLLM-DSC, a novel LLM-Knowledge Enhanced Cross-Modal Deep Structural Clustering framework. Diverging from data-driven paradigms, scLLM-DSC establishes a semantically-grounded representation by synergizing two views: a Knowledge-Driven Semantic View derived from NCBI gene priors and contextualized Cell2Sentence embeddings, and a Structure-Aware Topological View extracted via a graph-guided encoder. Crucially, we introduce a cross-modal contrastive alignment mechanism to enforce consistency between biological semantics and transcriptomic features within a unified latent space. Extensive benchmarks demonstrate that scLLM-DSC significantly outperforms eleven state-of-the-art baselines in clustering accuracy.

Prevailing wisdom posits that Multi-Agent Systems (MAS) are superior to Single-Agent Systems (SAS), citing advantages like context protection, parallel processing and distributed decision-making. However, empirical support for this claim relies primarily on comparisons with SAS baselines using benchmarks that prioritize isolated reasoning tasks, which do not adequately assess these advantages. Focusing on automatically generated MAS that are designed for enhanced generalizability over manually-designed counterparts, we perform a rigorous, systematic evaluation against SAS, specifically Chain-of-Thought with Self-Consistency (CoT-SC). Across traditional reasoning datasets and tasks with interactive multi-step workflows (e.g., BrowseComp-Plus), we demonstrate that automatic MAS consistently underperform CoT-SC despite being up to 10x more expensive. To isolate these failures from limitations inherent to task structure, we introduce a diagnostic synthetic dataset tailored for MAS featuring explicit task decomposition, context separation and parallelization potential. We show that expert-architected MAS consistently outperforms automatically generated architectures in both raw performance and cost-efficiency on this dataset, demonstrating that existing evaluation frameworks mask critical architectural gaps and inefficiencies of complex MAS by failing to account for the marginal utility of increased computational cost. Critically, systematic deconstruction of the generated MAS architectures reveals that current automated design paradigms produce architectural bloat that prioritizes superficial complexity which does not translate into functional utility, exposing a fundamental misalignment with multi-agent principles.

Two dominant approaches have emerged for generating probabilistic forecasts of physical systems: generative models, such as diffusion or flow matching; and ensembles of deterministic models with stochasticity injected, trained using the continuous ranked probability score (CRPS) loss. While both approaches have demonstrated strong predictive accuracy, the reliability of their uncertainties has not been systematically assessed. We address this gap by developing a framework to evaluate both approaches across diverse 2D spatiotemporal physical systems, under matched model size and computational budget. We assess the reliability of probabilistic emulation by inspecting the empirical coverage of predictive intervals, while also considering accuracy and computational efficiency metrics. CRPS-trained ensembles typically achieve more reliable uncertainties on both single-step prediction and autoregressive rollouts, demonstrating better coverage than the standard alternative of training generative models in a latent space. Moreover, the CRPS approach offers significantly faster inference. When generative models are trained in ambient rather than a compressed latent space, which is often infeasible for high-dimensional problems, they exhibit comparable coverage to CRPS-trained ensembles, though with substantially larger inference latency. In contrast, when CRPS-trained ensembles are trained in latent space they do not show a marked degradation in coverage with respect to ambient space. Both generative models and CRPS-trained ensembles demonstrate good predictive accuracy. To facilitate future research and application, we release AutoCast, a modular framework implementing both generative models and CRPS-trained ensembles, alongside AutoSim, a flexible dataset generation package for rapid prototyping.

Data-driven engineering design is constrained by the lack of large-scale 3D datasets that pair geometry with physics-based performance labels. In particular, existing 3D data augmentation techniques have limitations in preserving subtle and diverse geometric variations, and it remains difficult to automate the subsequent simulation-labeling process, where boundary conditions vary depending on the generated geometry. We present DeepJEB++, a foundation-model-driven data-augmentation framework that expands a small seed set of jet engine brackets into a large, simulation-labeled 3D dataset under constrained resources. Our key idea is to augment in the data-rich 2D latent space, then transfer to 3D. In Stage 1, we fine-tune a pretrained 2D latent diffusion model on multi-view renders and synthesize novel views by latent interpolation, retaining manufacturable designs through a vision-language-model (VLM) quality filter. In Stage 2, the validated images are lifted to 3D meshes by a domain-adapted generative foundation model. In Stage 3, an automated pipeline recognizes the load and bolt interfaces on each mesh and assigns finite-element labels -- mass, stress, and displacement -- without manual intervention. We assess augmentation quality along three intrinsic axes: manufacturability, label fidelity against the SimJEB ground truth, and distributional consistency. Starting from fewer than 400 seed designs, DeepJEB++ yields 15,360 simulation-labeled 3D brackets -- a 40x expansion -- using a single GPU per stage. The dataset will be made publicly available to support reproducible engineering-AI research.

Cyclic peptides represent a promising class of therapeutic compounds in modern drug discovery, often offering improved stability and binding affinity. However, the de novo design of cyclic peptides remains challenging because methods must identify pocket-adaptive cyclization patterns and linkage sites while simultaneously controlling drug-relevant properties. This challenge is particularly pronounced for recent generative models trained predominantly on linear peptide data, which may fail to capture cyclization-specific constraints. To address the limitation, we introduce APCyc, a target-aware de novo cyclic peptide generation framework that explicitly models cyclization and jointly optimizes multiple essential physicochemical properties. By using an expanded residue vocabulary and explicitly encoding cyclization-site and linkage-type information, APCyc learns cyclization-aware representations and leverages Bayesian posterior guidance to steer sampling toward cyclic peptides satisfying multiple property objectives. Experimental results demonstrate that our model learns target-dependent cyclization preferences, and enables effective and controllable multi-property optimization for cyclic peptide design. The source code of this paper is available at https://github.com/HKUSTGZ-ML4Health-Lab/APCyc.

Recursive multi-step forecasting is usually framed as distribution shift: models are trained on observed histories but deployed on their own predictions. We show this framing is incomplete by proving that, under partial observability or state truncation, recursive rollout is also an epistemic underidentification problem. Even with deterministic latent dynamics, one-step Bayes supervision identifies behavior only on observed contexts and need not identify the deployed recursive predictor once rollout queries self-generated induced states whose correct local targets are not determined by numeric state alone. We formalize this with induced states $Z$ and provenance variables $P$, and derive a decomposition of induced-state error into teacher-forcing/rollout mismatch, representation--class approximation, and provenance information gaps. Empirically, we show that rollout enters a distinct induced-state regime, that fixed induced states define a distinct local corrective task, and that closed-loop gains arise not only from local adaptation but also from changing the induced states visited during rollout. Using a simple binary provenance encoding, provenance-aware correction can further improve performance, though gains are conditional rather than uniform. These results recast exposure bias as reasoning under self-induced epistemic uncertainty.

This paper introduces a new methodology for real-time prediction of ergonomic and non-ergonomic human poses using volumetric video data in three dimensions. Although the methodology was designed for ergonomic assessments, it can be adapted to other applications requiring real-time analysis of human posture. One aspect that makes this system stand out is its ability to analyze 3D point clouds during the assessment, enabling computation from multiple angles. This overcomes a critical limitation of cameras which provide often a fixed viewpoint, thereby restricting the data available for a thorough postural evaluation, especially when occlusions occur. The system continuously and automatically performs pose inference using the chosen perspective on the real-time streaming data; however, only the poses manually selected and labeled by the user are used to train the personalized deep learning classifier. The methodology has been refined through a case study in which RGB-D cameras captured subjects performing load-lifting tasks, enabling real-time skeletal labeling. The model was trained on this data and, following the training phase, performs inference on new streaming data in real time. This research offers a scalable and pragmatic approach for real-time ergonomic evaluation by combining state-of-the-art 3D data technologies and traditional 2D pose estimation algorithms. It addresses the increasing need for safety and health monitoring in workplace environments, marking a notable contribution to the domain.

Action-conditioned world models let an autonomous vehicle predict future camera scenes from its own planned controls, enabling planning and simulation without real-world rollouts, but at compact, trainable scale the futures are ambiguous and the field's standard distortion metrics actively mislead: they reward a blurry regression mean over a realistic prediction. We confront this with a compact latent world model that, given the present front-camera latent and a sequence of ego-actions, predicts future scene latents a frozen decoder renders to $256 \times 256$ frames up to 8 seconds ahead, evaluated on 150 held-out nuScenes scenes. We first benchmark where to predict: across six frozen encoders spanning four representation families, V-JEPA2 with temporal context reduces steering RMSE by 40% over the best single-frame encoder. We then train a latent Diffusion Transformer (DiT) and, through a controlled diagnosis, identify the four ingredients it needs: spatial tokens, the $x_0$ objective, residual anchoring, and sampling matched to target uncertainty. In a Stable-Diffusion-VAE encode-predict-decode pipeline we expose the central tension: distortion metrics (cosine similarity, SSIM) favor the blurry mean, masking that the diffusion model is far closer to the real frame distribution. Inception-based FID and KID reveal a clean perception-distortion frontier: diffusion attains KID 0.078 versus 0.375 for regression ($4.8\times$ better), and a deployable train-derived calibration makes this practical without test-time ground truth. The model is genuinely action-controllable (steering drives scene displacement, Spearman $ρ= 0.81$, vs $-0.18$ for regression). We trace limited single-pass motion to a shared-present anchor and engineer a compact 1.7M-parameter "jump" model that recovers full ground-truth motion magnitude ($1.02\times$ GT), where single-pass models capture less than half.

Generative Artificial Intelligence (GenAI), Large Language Models (LLMs), and emerging Agentic AI constitute the most disruptive transformation in the history of software engineering (SE), reshaping development processes, required competencies, professional roles, and the educational outcomes that universities must deliver. This paper presents a systematic review of 48 verified, influential peer-reviewed publications (2016--2026) drawn from leading venues in software engineering, machine learning, computing education, human--AI collaboration, and software productivity. Studies were discovered, screened, and analyzed through a four-agent research workflow (Literature Discovery, Scientometric Analysis, Curriculum Transformation, and Workforce Impact) and were verified against primary sources. We synthesize the evidence along nine themes and three trajectories -- practice, education, and workforce -- and report a scientometric inflection in which annual LLM-for-SE output grew roughly five-fold after late 2022. From this synthesis we contribute: (i) a conceptual framework for AI-native software engineering organized around \emph{intent}, \emph{collaboration}, and \emph{verification}; (ii) a nine-dimension competency model spanning specification, critical evaluation, agent orchestration, and metacognition; (iii) a four-phase university curriculum roadmap with AI-resilient assessment; (iv) faculty-development and workforce-transformation strategies; and (v) a prioritized agenda of eleven research gaps. The evidence base is internally contradictory on the magnitude and direction of productivity effects, underscoring that benefits are strongly context-dependent and that educating engineers for judgment, verification, and orchestration -- rather than code production alone -- is the central challenge of the AI-native era.

Image-based AI assistants are now deployed at production scale on e-commerce platforms, where a single uploaded image can trigger fundamentally different user intents: product search, style recommendation, visual encyclopedia, or utility tool calls, each demanding its own response format, tool invocation, and domain knowledge. Without per-intent behavioral constraints, LLM-based systems conflate these heterogeneous modes and fall short of domain quality standards, while the breadth and dynamism of the intent space render manual engineering infeasible. To address this, we present SkillChain, which closes the production feedback loop on Skill evolution, automating the lifecycle of Skills through three stages: Skill Creator for bootstrapping from task specs and trajectories, Route Optimizer for routing alignment, and Body Refiner for iterative Skill Body refinement via dual-path LLM-Judge evaluation. Deployed on a production-scale e-commerce image assistant, SkillChain substantially improves aggregate response quality, with the strongest gains on structural compliance and content quality; a one-week online A/B experiment further confirms significant gains in user engagement, content consumption, and long-term retention.

Automated testbench generation has become a critical bottleneck in large language model (LLM)-driven Register Transfer Level (RTL) workflows, where large numbers of candidate designs must be verified rapidly and reliably. Existing prompt-based approaches treat testbench generation as unconstrained code synthesis, yielding stochastic outputs with high token cost, low reproducibility, and insufficient coverage. To address this gap, we present STG, a Structured Testbench Generation framework that exploits the inherent structure of hardware designs to generate deterministic testbenches. As a direct verification tool, STG runs 720x faster than an iterative LLM-based testbench generation flow and higher rate of successful compilation, achieves higher coverage, and reduces false-pass verdicts on incorrect DUTs. STG also helps identify errors in RTL generation benchmarks by exposing faulty benchmark testbenches. As a data curation engine, it is 11x faster than LLM-based filtering on a single CPU core with 127x less energy, and the resulting distilled models provide state-of-the-art performance in our multi-benchmark evaluation. As a test-time scaling oracle, it reduces node count by 14-47\%. Our models are available at https://huggingface.co/collections/AS-SiliconMind/siliconmind-v12.

JEPA-family world models use a static predictor whose weights do not adapt when test-time dynamics diverge from training. We compare two mechanisms for incorporating accumulated experience into a JEPA predictor under distribution shift: operand-side injection, where a compressed experience representation is added as a residual to the predictor's hidden state (EI-JEPA), and operator-side modulation, where the same representation generates low-rank weight deltas via LoRA applied to the predictor's weights (EPM-JEPA). On a pre-registered comparison (Moving MNIST, gravity shift), EPM-JEPA (D_shift^{n=50} = 0.7848 +/- 0.0078, three seeds) differs from EI-JEPA (0.8238) by delta = 4.74% - Outcome C: a null result - by our stated criterion, a valid outcome. As a secondary, non-pre-registered observation, EPM-JEPA improves 1.90% over a no-memory baseline (0.8000), consistently across seeds, while EI-JEPA underperforms the baseline, indicating the benefit is specific to weight-level modulation. Our primary contribution is a mechanism analysis: the D_shift^{n=50} trajectory reflects three independent dynamical processes - buffer cycling, EMA target drift, and an intrinsic LoRA settling transient of +0.021 - rather than convergence to equilibrium. These findings motivate PEM-JEPA, a physics-grounded successor addressing this dynamical-peak limitation.

Model fingerprinting, embedding user-specific identifiers (fingerprints) into generated outputs, has recently emerged as a popular solution to protect the intellectual property rights (IPR) of generative text-to-image (T2I) models and prevent unauthorized redistribution. In this work, we reveal a previously unexplored systematic vulnerability in existing generative model fingerprinting methods: they lack robustness against collusion attacks, where multiple attackers combine their models to remove or obscure the fingerprints. To address this issue, we take the first step towards a robust fingerprinting method for T2I models with anti-collusion capabilities. The proposed method encodes strings of bits, namely fingerprints, into the coefficients of a personalized normalization module (PNM) incorporated into T2I models, so that fingerprints can be reliably recovered from any generated image. To defend against collusion attacks and prevent unauthorized model redistribution, we introduce an anti-collusion mechanism based on lossless function-invariant parameter transformations. This mechanism significantly degrades the image generation quality of colluded models, making them effectively unusable. Moreover, our method allows developers to efficiently create multiple copies of fingerprinted T2I models by reparameterizing the PNM without the need for retraining. We also introduce a worst-case optimization strategy to improve robustness against model-level attacks. Our experiments demonstrate that the proposed method achieves high fidelity and robustness across multiple T2I image generation and editing tasks, with fingerprint extraction accuracy exceeding 99.5%. Compared with existing methods, our method demonstrates, for the first time, a notable proactive robustness to collusion attacks by significantly increasing the FID of colluded models.

Sharing mathematical content in online forums remains a significant friction point for students and educators: writing raw LATEX is error-prone, standalone optical character recognition tools require platform switching, and current forum software offers no integrated path from a photograph of a formula to a rendered post. We present a unified system that eliminates this friction by embedding an image to LATEX conversion pipeline directly inside a forum posting interface. A user uploads or captures an image of a mathematical expression; the system routes it through the Mathpix OCR API, detects whether the returned output is LATEX or plain text containing inline math, applies the appropriate delimiter normalisation, and renders a live preview in either LATEX or Markdown mode before the post is committed to the database. The architecture is organized in three loosely coupled layers: image processing, rendering, and storage, and supports both desktop and mobile clients. A provisional US patent application has been filed covering the core methods. We describe the full system design, each component in detail, the data schema, and the key technical innovations, and we position the work against existing standalone tools and forum platforms to demonstrate the practical gap it closes. Beyond immediate usability, we argue that a deployed platform of this kind constitutes a continuously growing, community-validated dataset of mathematical problems and step-by-step solutions, a resource that can be used to train and benchmark AI systems for accurate mathematical reasoning

Estimating cognitive load from speech has largely been studied in controlled laboratory settings, with limited understanding of its reliability in natural collaborative conversations. We investigate whether speech and interaction dynamics predict perceived cognitive load during dyadic conversations. We analyze audio from 53 dyads performing nine collaborative tasks and extract static acoustic, dynamic, and interaction features to train a two-head Gated Recurrent Unit encoder to predict cognitive load scores. Results show conversational interaction provides useful signals for predicting cognitive load related to time pressure, mental work, effort, and task performance. Temporal demand is associated with turn-taking dynamics such as overlap and speaker switch, while mental demand is linked to imbalanced participation between speakers. These findings highlight the importance of task structure and conversational interaction for modeling cognitive load in natural collaborative settings.

The power distribution network is critical to reliable electricity delivery, yet traditional inspection methods face limitations in semantic understanding, generalization, and closed-loop automation. To address these challenges, this paper proposes a Multi-Modal Agent framework specifically for power distribution defect detection. Central to this study is the systematic evaluation of multimodal foundation models as unified cognitive engines. We rigorously assess their integrated performance across three critical capabilities: (1) Perception, where the model must accurately identify equipment and generate expert-level descriptions of defects; (2) Reasoning, where the model interprets visual findings to diagnose causes, assess severity, and plan maintenance strategies based on domain knowledge; and (3) Tool Usage, where the model acts as an autonomous operator to execute actions -- such as querying knowledge bases or generating work orders -- to achieve closed-loop maintenance. To support this evaluation, a domain-specific evaluation dataset and a comprehensive benchmark are developed. Experimental results demonstrate the strengths and limitations of current foundation models in these three dimensions, providing empirical evidence for deploying autonomous agents in high-stakes industrial environments.

We introduce Apollo, a 10000 node p-qubit neuromorphic processor fabricated in 16 nm mixed signal CMOS and operating fully at room temperature with a typical analog core power envelope of about 0.5 W. Its fundamental element, the p-qubit, is a bistable stochastic unit whose continuous time state fluctuations are driven by integrated quantum entropy units that inject true quantum derived randomness. This enables ultrafast stochastic transitions at low energy while preserving a classical state representation. Apollo combines these p-qubits with a high degree Hyperion 256 interconnect topology, allowing efficient embedding of dense Ising and QUBO problems with substantially reduced minor embedding overhead compared with sparse annealing platforms. We show that, through the Suzuki Trotter correspondence, the equilibrium statistics and annealing dynamics of the p-qubit network reproduce key properties of transverse field quantum annealing without cryogenic cooling, long lived coherence, or microwave control. Beyond device level validation, Apollo is evaluated on a three dimensional spin glass benchmark previously used to study quantum advantage in superconducting annealers. Across 300 disorder realizations, Apollo reaches substantially lower ground state energies than reported cryogenic quantum annealing hardware, while remaining distinct from classical simulated annealing and simulated quantum annealing. A 350 nm release candidate device experimentally validates the core p-qubit dynamics, thermodynamic sampling correctness, and continuous time annealing behavior. These results establish Apollo as a room temperature, industrially scalable platform for quantum driven energy based optimization, probabilistic inference, generative modeling, and hybrid classical quantum workflows.

Grokking -- where a transformer on modular arithmetic suddenly transitions from near-chance to near-perfect validation accuracy -- is attributed to a Fourier circuit, but its timing, causal structure, and controllability remain poorly understood. We introduce the Frequency Synchronization Degree (FSD), a normalised, permutation-tested metric for Fourier circuit synchronisation requiring no prior circuit knowledge. Across nine modular addition configurations (primes p in {53, 71, 97, 113, 131}, three seeds), FSD synchronises 500-3,000 steps before grokking (mean lead +1,722 steps; all nine positive, sign-test p~0.004), and precedes a restricted-logit loss baseline (Nanda et al.'s excluded loss) in all nine cases, making it the earliest available predictor. We provide direct causal evidence that the inter-phase gap is a regularisation phenomenon: forking training at the FSD-ceiling step and varying weight decay lambda produces strictly monotone earlier grokking, with Delta_t proportional to 1/lambda. This law replicates across three primes (p in {53,97,131}; R^2=1.00 and R^2=0.99 for two clean cases), captured as Delta_t ~ C/lambda, consistent with (1/lambda)*log(||W_mem||/tau). Architecture ablations show an attention-only model groks with a strong FSD precursor; an MLP-only model never groks; a single-layer model's FSD lags, confirming the precursor is a multi-block circuit property.

The rapid growth of AI models and increasing data sovereignty requirements are driving the transition toward geo-distributed AI training across multiple data centers. Such deployments introduce system-level challenges arising from synchronization-intensive communication, cross-site data exchange, and wide-area latency constraints. This paper investigates EVPN--VXLAN as an infrastructure foundation for geo-distributed AI training environments and presents a scalable emulation framework for systematically studying distributed AI workloads under realistic wide-area conditions. The proposed framework combines VXLAN overlays with EVPN-based inter-data-center connectivity and is implemented using ContainerLab and FRRouting (FRR). The framework further incorporates Equal-Cost Multi-Path (ECMP) routing, Bidirectional Forwarding Detection (BFD), and a queue-pair-aware traffic distribution mechanism designed to improve communication behavior for synchronization-intensive AI workloads while preserving compatibility with commodity infrastructure. Using realistic WAN emulation, we characterize communication and system behavior under distributed training workloads employing AllReduce and Parameter Server communication patterns. Results provide insights into traffic distribution, resilience, and infrastructure behavior in geo-distributed AI environments, highlighting the potential of reproducible multi-data-center infrastructure frameworks for scalable distributed AI training.

We present OpenMedQ, a medical vision-language model pretrained on the broadest fully-open medical mix to date: 14 datasets totaling ~3.35M pretraining samples spanning pathology, radiology, microscopy, and text-only clinical QA. OpenMedQ reaches state-of-the-art BLEU-1 on PathVQA (75.9), beating Med-PaLM M variants up to 562B parameters (~80x larger), and matches the best reported VQA-MED BLEU-1 (64.5). Its vision encoder, transferred to 8 unseen medical classification benchmarks under an identical downstream recipe, obtains the highest average macro-F1 (0.757) among BiomedCLIP (0.745), PMC-CLIP (0.745), PubMedCLIP (0.746), and a from-scratch baseline (0.616). We release our code and an interactive demo is publicly available as a reproducible baseline for the community.

Large Language Model based Multi-Agent Systems (LLM-MAS) have emerged as a powerful paradigm for tackling complex tasks by breaking them into collaborative workflows of specialized LLM-powered agents. However, deploying such multi-agent workloads at scale poses significant system challenges. Each user query spawns an iterative pipeline of LLM calls, greatly amplifying resource consumption compared to single-turn queries. In resource-constrained cloud settings, these workflows face non-deterministic and input-dependent costs at decode stage, heavy-tailed multi-model requirements with memory fragmentation and over-provisioning, and cross-cluster scheduling trade-offs. We present Maestro, a workload-aware scheduling system designed for LLM-MAS serving under strict GPU budgets. Maestro explicitly leverages agent semantics and roles: it predicts the output length and memory usage of each stage and uses this prediction to drive a hierarchical scheduler. At the node level, Maestro enables dynamic multi-model co-location via hierarchical weight caching and elastic memory provisioning. At the cluster level, it performs latency-aware routing to avoid cold-start delays and memory overloads. At the global level, it enforces workflow-aware prioritization to minimize head-of-line blocking for interactive tasks. Across prototype experiments and trace-driven simulations, Maestro reduces KV-reservation HBM by 67.2% and improves high-contention SLO attainment over EDF by 23.6 percentage points.

Long-running LLM agents accumulate interaction histories far larger than any context window, forcing a standing decision: what to encode deeply, what to forget, and what to retrieve under a fixed memory budget. Production systems answer with semantic similarity or recency -- both mis-specified for the forgetting decision, which is made at consolidation time before the future query is known. We propose a multi-factor memory value function V(m)=\sum_i w_i f_i(m) over seven interpretable factors (emotional intensity, goal relevance, value alignment, self/user relevance, task utility, reliability, and usage history) drawn from cognitive psychology, whose weights are learned from a downstream objective by a gradient-free optimiser, and whose single scalar uniformly controls encoding depth, forget risk, and retrieval rank. We make a methodological point: on LongMemEval, scoring goal relevance against the held-out evaluation question saturates gold-evidence retention at \approx 0.98 -- this measures retrieval, not forgetting. In the realistic blind regime, a learned multi-factor value retains 0.770 \pm 0.011 of gold evidence across 479 usable cases, versus 0.657 for uniform weights, 0.518 for the best single factor, and 0.368 for recency; every paired gap's 95% bootstrap CI is above zero, and a neural network over the same factors ties the linear model. The learned weights are interpretable -- reliability, emotional intensity, and self/user relevance dominate, while query-time goal similarity is correctly down-weighted for the forgetting decision. A controlled synthetic task with planted confounds confirms the learner recovers a separating weighting (1.00 retention) where uniform weighting fails (0.62). The substrate is open-source; all experiments run on a single CPU with no API calls.

Generative listwise ranking with Large Multimodal Models (LMMs) aims to capture global list context in a single forward pass, but its effectiveness degrades in long-context multimodal scenarios. We identify a recurring failure mode, parse collapse, where the autoregressive decoder produces fluent yet incomplete rankings by silently omitting candidates and terminating early. This failure stems from limited context utilization rather than simple formatting mistakes, making prompt engineering and constrained decoding insufficient. We propose PRISMR (Parameterized Representation Internalization for Semantic Multimodal Ranking), a framework that replaces transient in-context list processing with parametric structural conditioning. PRISMR uses a lightweight hypernetwork to encode multimodal candidates in parallel and generate item-specific LoRA weights, which are synthesized into an instance-specific adapter for a LMM. This paradigm enables more robust internalization of list structure while preserving the base model. We further introduce a large-scale multimodal review-ranking benchmark for evaluation. Experiments demonstrate that PRISMR substantially reduces parse collapse, improves listwise ranking performance, and transfers effectively across domains and instruction-tuned backbones.

When a user reveals task-critical information across several conversation turns, LLM accuracy drops by up to 65% despite full context availability. We show that this Lost in Conversation degradation can be substantially mitigated by training models to maintain a compact rolling memory instead of attending to a growing history. To make such training scalable, we introduce a low-cost sharding pipeline that converts single-turn QA datasets into multi-turn fragmented-information episodes, eliminating the need for hours of manual annotation. Training only on sharded GSM8K, our memory-augmented policy significantly improves multi-turn accuracy and generalises zero-shot to harder math and out-of-domain long-context QA. Moreover, memory-trained models outperform full-history baselines even when given the full history at test time, suggesting that learning to compress induces more robust incremental reasoning than full-context exposure alone.

Neural speech codecs based on Vector-Quantized VAEs (VQ-VAEs) are core audio tokenizers for speech LLMs, yet their reconstruction fidelity is bottlenecked by quantization error. Modifying the quantizer or increasing model capacity are common fixes, but they complicate downstream language modeling. Our core idea is to align the decoder's internal feature manifolds when processing both the quantized tokens and their original continuous embeddings, using a lightweight feature-mapping loss. This requires minimal training overhead and no inference-time changes. Applied to XCodec2, self-guidance improves all reconstruction metrics, achieving state-of-the-art low-bitrate performance. Notably, it enables a 4x codebook reduction without fidelity loss, which downstream TTS experiments show significantly improves LLM-based synthesis by simplifying the token modeling space. Multiple statistical observations and visualizations corroborate the enhanced internal manifold alignment in the decoder. Extensive experiments confirm its generality across various inductive biases. Self-guidance thus establishes an efficient, broadly applicable method for high-fidelity neural audio coding.

Wet-lab robots can improve the reproducibility, throughput, and safety of biomedical experiments, but scaling their learning requires customizable simulators for safe and reproducible task generation, open editable laboratory assets, and efficient pipelines that turn limited demonstrations into usable training data. We present Pipette, an embodied simulation platform, benchmark, and data-efficient augmentation framework for wet-lab robot learning. Pipette releases over 43 open-source and re-editable wet-lab assets, together with an extensible asset-building pipeline. A key component of Pipette is its simulation-based data augmentation pipeline, replaying human demonstrations in simulation, applies lighting, camera, speed, and action perturbations, and filters generated episodes with automatic task success checks, rapidly expanding usable training data from limited manual demonstrations. We further introduce an 11-task wet-lab embodied benchmark covering sample handling, culture-ware manipulation, device operation, and precision placement. With only 30 demonstrations per task, ACT achieves 65.5% average success rate, while simulation augmentation improves SmolVLA from 44.1% to 74.7% and π0 from 40.4% to 46.5%, validating the effectiveness of Pipette for data-efficient VLA training and evaluation. Pipette also supports natural-language-driven scene construction and task registration, lowering the barrier for non-expert users to define new wet-lab robotic tasks.

Parallel test-time scaling samples many reasoning traces and majority-votes their answers, improving LLM accuracy but requiring traces to run to completion, incurring substantial computational overhead. We observe that probing partial traces at intermediate checkpoints can extract current answers without disrupting generation, revealing an evolving aggregate vote. Based on this observation, we introduce MARS, a margin-adversarial stopping rule that estimates which active traces are likely to change their answers and stops once the leader remains safe under a conservative bound on future vote movement. The rule separates two sources of uncertainty. It learns the trace-level switch probabilities that determine how much of the current margin is likely to be retained, while handling the harder question of where switching traces land through an adversarial bound calibrated from warmup traces. With true switch probabilities, MARS guarantees with high probability that the early-stopped answer matches the full-budget vote. In practice, a five-feature logistic model closely matches oracle switching behavior. Across three reasoning models and three competition-math benchmarks, MARS saves 25-47% of self-consistency tokens and 14-29% on top of DeepConf Online, a strong confidence-weighted baseline that already filters and truncates weak traces, while matching the accuracy of the corresponding full-budget baselines.

Invariant learning can fail even when the invariant structure is statistically identifiable. We show a conditional computational barrier: under a black-box samplable supervised sparse recovery primitive motivated by average-case sparse-recovery reductions, there exist \emph{samplable} multi-environment instances with a one-dimensional predictive invariant subspace ($k=1$) that are learnable with polynomial samples by exhaustive search, while any polynomial-time constant-accuracy recovery algorithm would contradict the primitive. We further quantify environment diversity by a separation parameter $γ$, which controls identifiability and the curvature of invariance objectives. Under sufficient diversity and local Gaussian regularity, the minimax risk is $\mathbb{E}[\dist(\hat{V},V_{\mathrm{inv}})^2]=Θ(k(d-k)/(n|\mathcal{E}|))$, and under label-induced shifts a phase transition occurs at $n^*\propto k(d-k)/(|\mathcal{E}|γ^2)$ with refined estimation error scaling proportional to $1/γ^2$. Synthetic and real datasets illustrate the predicted gaps and transitions and motivate simple diversity diagnostics.

Multi-label recognition with frozen Vision-Language Models (VLMs) is brittle under distribution shift: standard zero-shot inference scores labels independently, ignoring co-occurrence structure and producing incoherent label sets where dominant concepts suppress weaker but compatible labels. We introduce Bayesian Conditional Priors (BCP) Estimation, a gradient-free test-time adaptation method that injects label dependency without tuning the backbone. BCP views zero-shot logits as a proxy for marginal posteriors under a fixed image-text likelihood and attributes shift-induced errors mainly to a mismatched label prior. For each test image, it selects a high-confidence anchor label and applies an anchor-conditioned Bayesian refinement. This update is closed-form in logit space and admits a pointwise mutual information (PMI) interpretation, explicitly promoting compatible labels and suppressing incompatible ones. BCP operates without target annotations by estimating anchor-conditioned priors online from the unlabeled test stream via lightweight second-order co-occurrence statistics, adding negligible overhead beyond a single forward pass. Across standard multi-label benchmarks and multiple CLIP backbones, BCP consistently outperforms strong TTA baselines, e.g., improving RN50 average mAP from 57.31 to 69.22 and ViT-B/16 from 62.61 to 71.79.

We present a modular two-agent simulation framework for evaluating conversational shopping assistant architectures. An independent buyer agent, configured with personas, missions, and patience levels, is paired with an interchangeable responder that integrates with a real e-commerce search API. Holding the buyer constant across experiments enables controlled comparison of responder designs on identical scenarios. Using 2011 conversations across 14 persona buckets, we establish four empirical findings. First, rolling-window memory outperforms intent-extraction memory on all quality metrics while being 35% faster per query. Second, illustrating rapid evidence-driven iteration, a systematic failure analysis of a responder version enables targeted fixes that reduce failure and near-failure rates by 62% across the full dataset. Third, swapping the responder LLM backbone from Gemini~2.5 to Llama~3.3~70B costs 0.16--0.45 points despite identical architecture. Finally, we document systematic philosophical disagreement between frontier LLM judges: Gemini rewards process correctness while Claude demands concrete outcomes, despite using the same evaluation prompt.

AI alignment, interpretability, steering, and neural perturbation studies identify order-inducing objects. We argue that order is not control. Control requires a receiver-gated response law: a denominator-indexed operator mapping material state, action/drive, bath, and receiver state to response displacement, sinks, effort, and basin projection. We identify it across biological, LLM, adapter, and stochastic-operator panels. The laws are local: an intervention can be admitted, saturated, sign-changing, leaky, or overdriven depending on medium, bath, receiver state, action port, and comparator. Control is assigned when finite effort moves a target or outcome-readout class under the same denominator while damage, null/evasive, invalid format, overdrive, and unnecessary effort stay bounded. Mouse ALM, C. elegans, and zebrafish panels provide physical response-operator evidence while excluding coordinate identity and controller conclusions. LLM panels show generated-output response laws: across four material conditions, response vectors are predictable at 72.8-73.7% component-sign accuracy, rising to 84.3-84.8% on nonzero components; held-out observers predict system-effect and target/oracle families at 93.6% and 91.7% accuracy. Constitution-conditioned adapters reshape susceptibility as prepared media, and stochastic-operator panels separate measured opportunity from deployable action policies. This gives a driven-dissipative response-system account at the mesoscopic control level: drives act through prepared media, baths, and receivers, producing admitted movement, impedance, sinks, or overdrive. The evidence supports local admitted control and measurable stochastic response operators, while leaving deployable pre-generation control, hidden/logit causal sufficiency, biological-to-LLM coordinate identity, and literal thermodynamic quantities outside scope.

Political bias in large language models (LLMs) is increasingly significant, but difficult to measure reproducibly across political and linguistic contexts. We introduce Polar, a 4,026-instance multiple-choice benchmark that measures political bias through option-level likelihoods rather than prompt-based generation. Polar covers two ideological axes and eight issue categories derived from the Manifesto Project, and evaluates models in parallel across U.S. and South Korean political contexts. Across 38 LLMs, measured bias varies systematically with political context, issue category, model group, and presentation language. All models lean left-progressive on U.S. political content, but show more centered and mixed patterns on South Korean content. Translation experiments further show that presentation language alone can shift measured bias. These findings highlight the need for multilingual and cross-contextual evaluation of political bias in LLMs.

Low-Rank Adaptation (LoRA) significantly reduces compute and memory costs for finetuning Deep Learning models but is often harder to tune than dense training: when using factor-wise optimizers such as AdamW, it is sensitive to initialization choices, its optimal learning rates transfer poorly across ranks, and it often fails to beat dense baselines. We derive LoRA-Muon by applying the Muon optimizer's spectral steepest-descent rule to the low-rank setting. Along with our split weight-decay rule, our main claim is that LoRA-Muon is a good low-rank proxy for full-rank Muon and Shampoo-family optimizers. Its optimal learning rates transfer across rank, width, depth, and factor-rescaling. In our compute-matched TinyShakespeare study, a rank-$2$ proxy recovers the dense best tested learning rate, and a rank-$32$ LoRA-Muon run attains lower mean validation loss than the dense baseline in the seed-averaged sweep. We further show that the Spectron optimizer depends on arbitrary factor scaling, so it would likely be a poor fit when finetuning starts from badly imbalanced factors, and that LoRA-RITE's simplified QR-coordinate core implements the same spectral update. LoRA-Muon computes that update without QR-decomposition and avoids storing second moments, making it more accelerator-friendly and memory-efficient.

Hierarchical multi-agent systems (MAS) are rapidly being deployed in high-stakes workflows across domains such as finance and software engineering. In these systems, safety and security are inherently distributed across role-specialized agents, significantly expanding the attack surface, particularly under coordinated adversarial behaviors such as privilege escalation and cross-agent collusion. Existing red-teaming approaches for MAS remain limited: they rely on heuristic selection of target agents and perturb isolated message streams, leaving critical questions unanswered as which agents are most responsible for system safety, and how compromised agents can coordinate to bypass defenses. We propose MAStrike, a closed-loop framework for collusive red-teaming in hierarchical MAS. We propose the first agent-level Shapley value analysis for MAS, quantifying each agent's marginal contribution to system robustness under task-specific distributions. GGuided by this attribution, MAStrike identifies vulnerable agent coalitions and generates coordinated, role-aware adversarial manipulations. These attacks are iteratively refined through structured causal diagnosis, attributing failure cases to uncompromised agents that block adversarial attempts. We further build a comprehensive MAS red-teaming benchmark and controllable environments spanning diverse hierarchical topologies and domains, including finance, software engineering, and CRM. Extensive experiments across MAS built on multiple frontier models show that MAStrike substantially outperforms heuristic baselines. Our analysis further uncovers non-trivial Shapley value distributions and higher-order interaction structures among agents, revealing critical vulnerabilities and coordination patterns that are overlooked by prior single-agent or template-based methods.

We present the first causal mechanistic analysis of a tabular foundation model, investigating how TabPFN 2.5's feature wise attention heads distribute computation across layers. Using activation patching, ablation, and attention entropy across two synthetic regression datasets, we find clear temporal specialisation: one head's causal necessity dominates that of the others by 2 to 5 times at peak layer, with its dominant layer shifting across tasks of different complexity, while the remaining heads exhibit symmetric late layer profiles. Attention entropy and patching provide convergent evidence for the computationally active layers of the dominant head. We additionally investigate inference time steerability via contrastive activation steering, which fails to transfer across samples. We attribute this result to TabPFN's in context learning mechanism, which encodes task structure through context dependent attention rather than the stable parametric directions that make steering tractable in language models.

Molecular dynamics (MD) is the canonical in-silico method for atomistic molecular science, simulating molecular behavior from first-principle physics. Designing an MD pipeline for a new system requires substantial expert knowledge: running it on even one molecule is expensive, ruling out trial-and-error. We automate this expert pipeline-design process with an LLM agent. Unlike existing MD agents that orchestrate a predefined tool set, we treat pipeline design as open-ended code generation in which the agent's behavior is reshaped online by verbal reward. Specifically, we build MDForge, an LLM agent whose in-context update rule densifies the sparse reward via a multi-agent debate among physics experts. On three SAMPL host-guest binding free-energy benchmarks, MDForge automatically designs MD pipelines competitive with human experts. Deployed on a library of unseen candidate guests, its CB[7] pipeline discovers a novel binder that wet-lab competition NMR confirms is a high-affinity, picomolar CB[7] binder. Our data and code are available at https://github.com/Zehong-Wang/MDForge.

The rapid growth of modern training datasets has significantly increased computational cost, motivating dataset pruning~(DP) methods which retain only a subset of informative samples to reduce training cost. Existing pruning criteria typically rely on either intrinsic signals that assess samples independently or extrinsic signals that promote diversity via pairwise relations. While effective in their own specific regimes, each captures only one aspect of sample utility and lacks robustness across different pruning ratios or data distribution. In this work, we present a unified graph-based DP framework. By modeling the dataset as a weighted graph, where node weights encode intrinsic value and edge weights encode extrinsic value, DP can be cast as a Maximum Weight Clique Problem (MWCP). Although MWCP is NP-hard, its structure admits a principled greedy solution based on sample-wise marginal gains. Under a few mild conditions, we further prove that this unified objective enjoys a formal approximation guarantee, which applies to a broad family of importance metrics and provides practical design guidelines. Extensive experiments show that our method outperforms existing DP methods while substantially reducing training cost, reducing training time by over 40\% without sacrificing accuracy on ImageNet-1k with ResNet-50.

Cascaded speech translation (ST) systems suffer from error propagation when Automatic Speech Recognition (ASR) outputs incorrect transcripts. We present the first systematic categorization of ASR errors for Vietnamese ST, classifying substitution errors by phonetic cause and quantifying their impact on downstream Neural Machine Translation (NMT) performance using Linear Mixed-Effects Modelling. We confirm that most ASR substitution errors arise from phonetic confusions rather than random noise, and that these phonetic errors significantly degrade ST quality. Motivated by this finding, we propose Phonetically-Informed Data Augmentation (PiDA), which generates ASR-like corruptions by substituting words with phonetically similar alternatives using phonetic word embeddings. Fine-tuning on a PiDA-augmented version of FLEURS Vietnamese-English improves translation of erroneous ASR outputs (up to +2.04 BLEU over standard fine-tuning) while also slightly improving clean-text performance.

For robotics to be effectively integrated into household or industrial environments, machines must adapt to natural-language prompts in real time. Although Vision-Language Models (VLMs) have enabled zero-shot generalization in robot task and motion planning (TAMP), current state-of-the-art approaches often remain computationally "heavyweight" or require extensive training on thousands of demonstrations. We present GRASP (Grounded Reasoning and Symbolic Planning), a framework designed as a step toward open-vocabulary tabletop manipulation. Our approach leverages a pretrained VLM to translate natural-language queries into neuro-symbolic goal states, grounded in the physical world via a bounding-box detection pipeline. Unlike methods that rely on fixed color lists or hard-coded coordinates, GRASP enables robots to interpret abstract spatial concepts such as "top shelf" and execute tasks without additional fine-tuning. We achieve 73.3% overall success across 90 real-robot trials at three difficulty levels, requiring no task-specific training.

Language model agents are increasingly effective in solving realistic tasks through multi-turn tool use. However, training reliable tool-using agents remains challenging in practice. While reinforcement learning provides an on-policy paradigm for improving agents from their own environment interactions, its effectiveness depends heavily on the training task distribution. When tasks are fixed before training, the task distribution can become increasingly mismatched with the policy's evolving capabilities, causing many rollouts to be spent on uninformative tasks. We propose SENTINEL, a failure-driven reinforcement learning framework that turns the Solver's rollout failures into targeted training tasks. SENTINEL follows a Controller--Proposer--Solver loop: the Controller analyzes failed trajectories and summarizes recurring error patterns, the Proposer generates executable tasks that stress these weaknesses, and the Solver is trained on the targeted tasks. On Tau2-Bench Retail with Qwen3-4B-Thinking-2507, SENTINEL improves Pass\^{}1 from 66.4 to 74.9 and outperforms RL on general synthetic tasks across Pass\^{}k metrics. These results demonstrate that model failures provide an effective and scalable source of targeted training signal for improving tool-using language model agents.

Prior work on a social web highlighter located individuality in selection -- which documents a person chooses to highlight -- but measured it cross-sectionally. We ask the temporal question: is a reader's selection signature a trait or a state? We freeze each reader's first six months of highlighting as a profile and track its own-vs-other advantage on their later selections at growing gaps (to 24+ months), with negatives drawn from the same calendar era -- so supply drift cannot masquerade as personal drift -- at a coarse global level and at a fine level whose negatives and controls come from the reader's own interest neighborhood; the anchor cell reproduces the prior cross-sectional level (+0.188 vs +0.169), validating the harness. Four results. Within the same users, the fine-layer advantage shows no statistically detectable paired decline at any horizon (6-12 month retention R = 1.00 [0.85, 1.18], n = 212; the farthest bin is compatible with a modest decline; the only contrast whose interval excludes zero is the coarse layer at 12-24 months, about 13%). The signal is not reducible to repeated domains (~90% survives excluding all profile sources). Within-person drift is slow (a recent-half profile beats the old half by +0.042). Prospectively, personal profiles -- even one built from a reader's earliest documents, median 20 months before evaluation -- rank their next reads at roughly 3x the AP of every simple non-personal prior tested. We use "trait" operationally (a stable signature under continued engagement); the scope is heavy, long-tenured readers of one platform, and exposure is not separable from choice.

Retrieval-augmented generation (RAG) systems may receive evidence that is not merely noisy but mutually contradictory. This issue becomes particularly salient in multilingual settings, where retrieved Chinese and English evidence may support incompatible answer candidates. We study this problem through X-RAMDocs-ZHEN, a controlled Chinese-English benchmark derived from RAMDocs for diagnosing evidence conflict in RAG. The benchmark contains 300 examples across six balanced conditions, including monolingual support, bilingual agreement, reversed conflict directions, and conflict with optional noise. We further examine X-MADAM-RAG, an interpretable pipeline that decomposes evidence handling into per-document candidate extraction, visible-evidence repair, deterministic candidate grouping, and conflict-aware aggregation. On the original controlled benchmark with Qwen2.5-7B-Instruct, X-MADAM-RAG achieves 0.9667 strict accuracy and 0.9767 conflict-aware success, outperforming an evidence-normalized single-call baseline. However, a zero-call rule-only extractor reaches 1.0000 on the same benchmark, revealing strong template regularity. To probe this limitation, we construct a deterministic naturalized stress test that removes explicit answer templates while preserving candidate strings. On its 100-sample subset, rule-only extraction falls to 0.0000, but X-MADAM-RAG also drops to 0.3000 strict accuracy, below both naive and evidence-normalized baselines. A privileged oracle remains perfect, indicating that document-level extraction is the main bottleneck. These findings position X-RAMDocs-ZHEN and X-MADAM-RAG as diagnostic tools for controlled evidence conflict rather than as evidence of general hallucination detection or robustness to natural retrieval.

Empathetic spoken dialogue systems require not only semantically appropriate responses but also emotionally aligned prosodic expression. However, cascade pipelines often discard acoustic cues during speech-to-text conversion, while end-to-end speech models lack interpretable control over emotion and knowledge integration. To address these challenges, we propose PRISM, a multi-agent framework for empathetic spoken dialogue that decouples speech perception, response generation, and speech synthesis into coordinated components. PRISM introduces a prosody-to-language translation mechanism to stabilize large language model reasoning and enables on-demand invocation of external knowledge tools for empathetic dialogue generation. Experimental results demonstrate that PRISM achieves consistent improvements in empathy, prosodic appropriateness, and text response generation quality across objective and subjective metrics. Our code is available at: https://github.com/Bxzfrm/PRISM.

Large language models (LLMs) often hallucinate by generating factually incorrect or unfaithful content, posing significant risks to their safe use. Detecting such hallucinations is particularly challenging under the zero-source constraint, where no model internals or external references are available, and detection must rely solely on the textual query-answer pair. In this paper, we propose Human-like Criteria Probing for Hallucination Detection (HCPD), a paradigm that emulates the multi-faceted reasoning of human evaluators. Its core is a Human-like Criteria Probing (HCP) mechanism, in which a LLM agent adaptively decomposes its judgment into a weighted set of interpretable criteria and aggregates criterion-specific scores into a final truthfulness measure. To achieve this adaptive capability, we introduce a reward-based alignment scheme using only weak supervision from semantic consistency. At inference, we employ a multi-sampling aggregation strategy to ensure robust decisions while preserving full interpretability. We further provide theoretical analysis supporting the reliability of our approach. Extensive experiments show that HCPD consistently outperforms state-of-the-art baselines, offering an effective and explainable solution for zero-source hallucination detection. Code is available at https://github.com/TRISKEL10N/HCPD.

Visual Text Comprehension (VTC) renders text into images for a vision-language model (VLM) to read, sidestepping LLM context-window limits and powering applications from long-page OCR to multi-page memory QA. Yet existing VTC pipelines treat rendering and layout as a fixed, content-agnostic preprocessing step and offer little mechanistic understanding of how VLMs internally process visualized text. Through a focused empirical study on VTC QA tasks, we reveal that VLMs exhibit a localization-without-utilization regime: evidence-localizing attention emerges sharply in the middle-to-late layers and is largely decoupled from answer correctness, yet simply enlarging the localized spans on the rendered page recovers a large fraction of the failures. Building on these observations, we propose AGAR (Attention-Guided Adaptive Rendering), a training-free, model-agnostic method that leverages a VLM's own middle-to-late layer attention to identify the top-K important visual patches, maps them back to word spans, and re-renders the page with those spans enlarged before re-inferring the answer. Extensive experiments across nine VTC benchmarks (short-form, long-context, and multi-page memory QA) and four VLM backbones show that AGAR (i)consistently improves off-the-shelf VLMs as a plug-and-play enhancement, (ii)composes with VLM post-training to yield further gains, and (iii)remains robust under both visual- and text-side input degradation.

Large language models (LLMs) are increasingly used to access organisational documentation, including standard operating procedures (SOPs), HR policies and institutional guidelines. However, retrieval-augmented generation (RAG) systems that rely on free-form rewriting can introduce hallucinations and unstable trade-offs between completeness and conciseness, particularly in safety- and compliance-critical settings. Objectives: To evaluate extraction as a hallucination-resistant alternative to rewriting-based RAG and compare strategies that balance precision, recall and safety across document types and model scales. Methods: We compare multiple prompting strategies, including line-number-based source selection, extraction of relevant guideline sentences with explicit safety annotations, and a multi-stage pipeline that refines draft answers using supporting evidence from source guidelines. Experiments are conducted on documents of varying length and structure, including local NHS acute care and oncology guidelines and UK-wide NICE guidelines, using both frontier-scale and locally deployable models. Performance is assessed using automatic metrics and human expert evaluation of relevance and completeness. Results: Line-number selection achieves the strongest results, outperforming direct copying and safety-focused strategies across both large and small models while maintaining high term recall (up to 95%) and close alignment with source text. Safety-oriented approaches improve precision but introduce systematic omissions, while multi-stage filtering further amplifies this trade-off. Performance varies with document structure: line-based extraction excels in protocol-like content, whereas alternative strategies perform better on more verbose documents (up to 97% term recall).

While real-world applications of reinforcement learning (RL) are becoming increasingly popular, the security of RL systems deserve more attention and exploration. In particular, recent work has revealed that RL agents are vulnerable to backdoor attacks, where a victim agent behaves normally under standard conditions but executes malicious actions when a specific trigger is activated. Existing backdoor defenses for RL either require access to the agent's internal parameters, operate only at the model or trajectory level, or are limited to specific attack types. To ensure the security of RL agents, we propose \texttt{PolicyGuard}, a \textit{test-time step-level} backdoor defense which leverages Gaussian Process (GP) posterior variance and adapts pseudo trajectories to enable uncertainty computation for individual time step. Besides, we also provide theoretical foundations to explain the efficacy of GP posterior variance. Extensive experiments across seven RL games demonstrate that PolicyGuard achieves state-of-the-art detection performance in most cases, with average AUROC of 0.856 for perturbation-based attacks and 0.859 for adversary-agent attacks.

Spiking Neural Networks (SNNs) are well-regarded for their biological plausibility and energy efficiency in processing sequential data. However, dominant SNN architectures typically rely on first-order Ordinary Differential Equations (ODEs) to govern neuronal state transitions. This first-order assumption imposes a "memoryless" bottleneck, limiting the model's capacity to capture the complex, long-range dependencies inherent in long-sequence tasks. In this work, we propose LongSpike, a novel SNN framework that integrates fractional-order State-Space Modeling, or f-SSM, from control theory into the spiking domain. By extending traditional integer-order SSMs to the fractional-calculus regime, LongSpike enables the hierarchical integration of neuronal dynamics with long-memory kernels. To mitigate the computational overhead and parallelization challenges typically associated with fractional operators, we leverage a state-space formulation that supports efficient, parallel training. Empirical evaluations on challenging benchmarks, including Long Range Arena (LRA), large-scale WikiText-103, and Speech Commands, demonstrate that LongSpike outperforms state-of-the-art SNNs in accuracy while preserving sparse synaptic computation. The code is available at https://github.com/xinruihe389-commits/LongSpike.

This study investigates semiparametric efficient estimation of causal and structural parameters in a semi-supervised setting. In our setting, unlabeled auxiliary regressors are available in addition to labeled observations consisting of outcomes and regressors. Our goal is to construct estimators of causal and structural parameters whose asymptotic variances are smaller than those of estimators constructed using only labeled data. We refer to this framework as prediction-powered causal inference (PPCI). We first derive the efficient influence function and the efficiency bound, which imply that the use of auxiliary regressors can attain a smaller asymptotic variance than the efficiency bound attainable from labeled observations alone. Then, by combining the efficient influence function with the debiased machine learning (DML) framework, we propose methods that we call DML-PPCI. If we construct an estimating-equation estimator, we refer to the method as EE-DML-PPCI; if we construct a targeted-learning estimator, we refer to the method as TMLE-DML-PPCI. The asymptotic variances of both estimators match our derived efficiency bound. In the construction of the estimators, estimation of the efficient influence function plays an important role. In our study, the efficient influence function is also a Neyman orthogonal score, which depends on the Riesz representer and the regression function. For Riesz representer estimation, we develop semi-supervised generalized Riesz regression with convergence rate guarantees.

Bitcoin's Lightning Network (LN) can be exploited as a covert, low-cost command-and-control (C&C) channel for botnets, as demonstrated by the LNBot and D-LNBot designs. However, both remain proof-of-concept prototypes evaluated only through simulation, leaving key questions about real-world topology formation, propagation complexity, and resilience to takedowns unanswered. We present LNTest, the first reusable testbed for LN-based botnets, built from Core Lightning nodes containerized with Docker over a shared Bitcoin Core regtest chain. LNTest supports three overlay topology modes (a deterministic chain, autonomous peer discovery, and user-supplied graphs), enabling controlled experiments across different botnet structures. Using LNTest, we report three main findings. First, D-LNBot's autonomous formation protocol does not produce the uniform chain from its design; instead, it creates a clustered chain in which cliques are linked by bridge nodes whose removal fragments the network. Second, command propagation scales linearly with botnet size ($Θ(n)$), not the $O(m \log n)$ previously claimed, and gains nothing from higher neighbor connectivity. Third, the overlay topology determines the effectiveness of takedown strategies: uniform-degree chains resist targeted removal but fragment under random failure, scale-free topologies show the opposite pattern, and the autonomous clustered chain is fragile under both, making it the most vulnerable of the three. LNTest is released as open source, with a script that reproduces all our experiments, to support reproducible research on LN-based botnet defenses.

Interleaved thinking, where a unified multimodal model alternates between textual reasoning and visual generation, has shown promise on spatial and physical tasks. However, in complex long-chain scenarios, we identify a fundamental failure mode: generated images diverge from the textual context while subsequent text ignores the visual evidence, causing the two modalities to alternate without genuinely informing each other. We term this Modal Isolation and attribute it to compounding information loss at modality boundaries. We decompose each reasoning cycle into atomic operations and define modality transition loss, quantifying cross-modal hallucination (text-to-image) and visual utilization deficit (image-to-text) at each boundary. We propose MoTiF (Modality Tiransition Fidelity), a two-stage training framework that directly optimizes these transitions: Reflective SFT trains the model to detect and recover from erroneous visual outputs; Flow-GRPO improves image generation fidelity via reinforcement learning. All training signals in MoTiF derive from transition-level fidelity rather than end-task accuracy. Across four visual puzzle benchmarks, this transition-level supervision substantially improves both cross-modal coherence and final task accuracy. The results demonstrate that effective interleaved reasoning requires explicit structural supervision at modality boundaries, not merely scaling or end-task optimization.

Mixed categorical-continuous optimization arises in many practical domains, yet remains challenging. In the black-box setting, evolution strategy-based approaches have shown promise in extending the efficiency and robustness of the CMA-ES to mixed-variable spaces. However, these methods exhibit worsened performance when strong categorical-continuous interactions are present, as their underlying search distributions assume independence between categorical and continuous variables. To address this limitation, we propose a bilevel optimization framework that explicitly captures such interactions by optimizing over categorical variables in an outer loop, and over continuous variables conditioned on each categorical configuration in an inner loop. We formulate each level of the bilevel problem as a stochastic relaxation under information-geometric optimization. To mitigate the high computational cost inherent to bilevel optimization, we introduce a warm-starting strategy that accelerates the lower-level search by selecting the best among multiple cached configurations and updating the cache after each iteration. Experimental results on binary-continuous domain demonstrate that the proposed method outperforms existing state-of-the-art approaches in interaction-handling capability while also being more computationally efficient across benchmarks encompassing both previously reported and newly proposed types of interaction.

In Low-Rank Adaptation (LoRA), the scaling factor $α$ is often treated as a mere complement to the learning rate, yet its role in optimization remains poorly understood. In this paper, we reveal that the scaling factor $α$ and the learning rate function differently, with $α$ emerging as the dominant driver of effective optimization, delivering gains that cannot be replicated by learning rate scaling alone. Through the synergy of extensive empirical analysis and a theoretical Signal-Drift framework, we uncover three findings into LoRA's scaling mechanism: First, LoRA's spectral suppression smooths the optimization landscape, rendering standard hyperparameters overly conservative and creating an optimization gap. Second, when leveraging this smoothness to accelerate convergence, $α$ outperforms the learning rate by amplifying the task signal without increasing the drift ratio. Third, the optimal scaling factor follows a sublinear relationship with the rank, well characterized by a square-root law with an unexpectedly large coefficient, revealing the insufficient scaling of existing rank-tied heuristics. Based on these insights, we propose LoRA-$α$, a minimalist framework that restores $α$ to its principled regime, making LoRA compatible with standard small learning rates. Extensive evaluations across diverse tasks demonstrate that LoRA-$α$ consistently improves performance while streamlining hyperparameter search, unleashing the learning potential of LoRA.

Large language models are increasingly deployed as agents for long-horizon tasks, yet their performance is shaped not only by model capability and environment design, but also by the harness that mediates agent--environment interaction. Existing harnesses are largely manually engineered, making them difficult to scale as trajectories grow longer and interactions become more complex. In this work, we ask whether harness can be generated by a learnable plug-in module that can be trained in an end-to-end fashion. We introduce HarnessBridge, a lightweight learnable harness controller that parameterizes the agent--environment interface as a bidirectional projection. HarnessBridge learns two bidirectional projections: observation projection, which distills raw trajectories into compact, decision-relevant states, and action projection, which converts proposed actions into executable transitions or trajectory-grounded rejections. We train HarnessBridge on a harness supervision dataset via unified instruction tuning. On Terminal-Bench~2.0 and SWE-bench Verified, HarnessBridge matches or surpasses strong specialized harnesses while substantially reducing token usage and trajectory length, and generalizes from smaller generators to larger commercial models.

We present an approach to fine-tuning large language models using Direct Preference Optimization (DPO), a reinforcement learning technique. Our experimental results demonstrate that DPO simplifies the training pipeline, improves computational efficiency, and achieves competitive performance. The evaluation using BLEU, ROUGE, and cosine similarity metrics indicates effective learning and convergence, though further investigation is needed to address observed training instability.

As large language models (LLMs) are increasingly deployed across heterogeneous hardware with varying resource constraints, the ability to adaptively manage the trade-off between performance and efficiency without retraining is critical. We propose Drop-by-Drop, a novel multi-bitwidth post-training quantization framework that enables inference-time precision control over LLM weights from a single trained model. Our method is theoretically grounded in information theory and successive refinement. We establish that LLM weights, which commonly follow a Gaussian distribution, can be optimally reconstructed with increasing fidelity as additional bits are incorporated, under a weighted mean squared error distortion motivated by LLM loss functions. To realize this in practice, Drop-by-Drop incorporates Matryoshka-style supervision into the loss function, exploiting the structure of additive codebooks. Drop-by-Drop produces a single model where ordered subsets of codebooks yield accurate partial reconstructions at each precision level. This approach significantly reduces storage and memory overhead by allowing a single checkpoint to serve multiple bitwidths, while maintaining competitive perplexity and accuracy across major architectures, such as Qwen, LLaMA, Gemma, and Mistral.

Search Agents (SAs) typically leverage large language models (LLMs) to support complex information-seeking tasks by autonomously exploring web sources and synthesizing information into comprehensive responses. For SAs evaluation, prior benchmarks mainly focus on specialized tasks that are unlikely to arise in real-world user scenarios. Moreover, their reliance on coarse task-level rubrics often limits evaluation interpretability. To bridge this gap, we introduce DailyReport, an open-ended benchmark to evaluate SA capabilities on daily search tasks. It contains 150 open-ended tasks with 3,546 associated rubrics, capturing widely discussed and timely information demands of real-world users. Each task is decomposed into subtasks and evaluated with cascade rubrics across disentangled dimensions. Through cascade performance attribution and user-centric aggregation, we derive highly interpretable scores for each dimension, along with a user preference score. Our results on 17 agentic systems show that current systems still fall short of users' expectations. To facilitate future research, our dataset and code are made publicly available at https://github.com/AGI-Eval-Official/DailyReport.

Multimodal-attributed graphs (MAGs) couple graph topology with node semantics from text, images, and other modalities. Traditional graph learning contextualizes node semantics by coupling topology with node features. However, this coupling design becomes troublesome in MAGs, where structure-induced and modality-intrinsic semantics may contribute differently to downstream tasks. Structure-induced semantics promote relational consistency through smooth topological variation, whereas modality-intrinsic semantics often encode local, fine-grained distinctions that should not be uniformly smoothed or aligned. Therefore, the key challenge is to identify semantic roles before cross-modal fusion. To this end, we leverage graph-frequency variation as a prior, where low-frequency components capture topology-consistent semantics and high-frequency components preserve modality-specific semantics. Based on this intuition, we propose SMGFM, a spectral multimodal graph pretraining framework that decomposes each modality-specific node signal into graph-frequency bands and assigns band-level semantic roles before cross-modal interaction. Concretely, SMGFM constructs frequency-resolved modality tokens with scalable Chebyshev filters, estimates their coupling reliability through topology-conditioned routing, and performs band-modality interaction before fusion. Its frequency-routed objectives align smooth consensus routes while preserving modality-specific routes, mitigating spatial-domain entanglement and uniform cross-modal alignment. Extensive experiments conducted on the MAG datasets demonstrate that SMGFM achieves state-of-the-art performance across graph-level and modality-level tasks.

Despite strong performance in competitive programming, the role of Large Language Models (LLMs) in supporting human learning in the same setting remains largely unexplored. In this work, we introduce UOJ-Bench, a benchmark designed to evaluate not only the problem-solving ability of LLMs, but also their ability to identify errors in human-written code -- a crucial educational activity traditionally supported by running test cases over online judge systems. UOJ-Bench consists of three distinct tasks: code generation, code hacking, and code repair, all constructed from real-world code submissions on the Universal Online Judge (UOJ) and evaluated through UOJ's native judging infrastructure. Our results show that under one-shot evaluation, even the strongest models fail to identify errors in more than 50% of a set of submissions that have been found to be incorrect by UOJ users. While test-time scaling improves success rates to above 90%, the substantial computational costs incurred from model inference limit its practicality for large-scale deployment. Despite these limitations, we find that the best-performing models under test-time scaling can uncover errors in over 5% of full-score submissions across roughly 30 problems, suggesting that frontier LLMs can already provide complementary signals beyond standard judging systems.

Multimodal attributed graphs (MAGs) integrate graph topology with heterogeneous modality attributes, such as text and images, thereby enabling richer modeling of complex relational systems. However, such expressiveness also makes learning on MAGs depend on multiple semantic sources, including structural topology, textual and visual attributes, each of which can be regarded as a branch for node representation. Node-level branch semantic imbalance arises when these branches differ across nodes in semantic informativeness and reliability: a branch that provides discriminative semantics for one node may mislead another due to bias in modality quality or structural context. Existing methods often mitigate such heterogeneity through cross-branch agreement or alignment, implicitly treating the dominant prediction as reliable supervision. When the dominant branch is biased, forced imitation may propagate its bias to other branches and suppress original semantics that are useful for classification. We propose GraphMNL, a graph-aware multimodal negative learning framework that addresses this issue by using Negative Learning as cross-branch guidance. Instead of forcing inferior branches to imitate a teacher prediction, the model teaches them which classes a node is unlikely to belong to. GraphMNL builds a branch library, identifies dominant and inferior branches via graph-aware reliability arbitration, gates unstable transfer, and applies target-preserving negative learning over non-target classes. This design decouples target supervision from branch guidance so that supervised losses learn the correct class, while Negative Learning suppresses unlikely alternatives when branch agreement is unreliable. Through the comprehensive experimental evaluation, GraphMNL achieves the best performance on Grocery datasets with 72.47% accuracy and 76.60 F1 score on Reddit M datasets.

Conventional communication systems, including both separation-based coding and learning-based joint source-channel coding (JSCC), are typically designed under Shannon's rate-distortion theory. However, relying on generic distortion metrics fails to capture complex human visual perception, often resulting in blurred or unrealistic reconstructions. In this paper, we propose Joint Source-Channel-Generation Coding (JSCGC), a generative communication paradigm that replaces the conventional decoder with a generative model at the receiver. The received signal is treated as a condition that controls the sampling process into the learned conditional distribution, reformulating communication from deterministic reconstruction for distortion minimization to controlled generation for mutual information maximization under perceptual constraints. Based on this formulation, we develop a unified joint training and efficient stochastic sampling framework, and provide theoretical analysis of its effectiveness in both learning and inference stages. Extensive experiments on latent-space image transmission demonstrate that the JSCGC consistently improves feature-based, semantic-level, and distributional quality across diverse channel conditions, while exhibiting a distinct error behavior characterized by semantic inconsistency rather than distortion.

Large Language Models such as GPT-4o and GPT-5 achieve strong zero-shot performance on biomedical claim verification, but cost and opacity limit scalable use. We fine-tune three small LLMs: Phi-3-mini (3.8B), Qwen2.5-3B, and Mistral-7B, via QLoRA on SciFact and HealthVer, providing the first study of QLoRA models against GPT-4o and fine-tuned BioLinkBERT encoders. Mistral-7B QLoRA surpasses both GPT-4o and GPT-5 (up to 12% F1 gain) at a fractional cost using just 1,008 training examples. We conduct extensive in-domain and cross-domain evaluation: models trained on SciFact tested on HealthVer and vice versa, at matched sizes to isolate dataset structure from data quantity. We identify a previously unreported structural artifact in SciFact that inflates in-domain scores, and show through bidirectional out-of-domain evaluation that training on structurally sound data enables robust cross-domain transfer. We plan to release all code and adapter checkpoints.

Rapid advances have been made in developing general-purpose embodied agent in environments like Minecraft through the adoption of LLM-augmented hierarchical approaches. Despite their promise, low-level controllers often become performance bottlenecks due to repeated execution failures. We argue that a key limitation is not only the lack of episodic memory, but also the decoupling of \textit{what-where-when} memory from \textit{which-why} reasoning. To address this, we propose \textbf{WISE} (Which-Why Informed Semantic Explorer), a long-horizon agent framework with an enhanced low-level controller equipped with a Causal Event Graph that augments episodic memory with explicit causal structure linking observations to task relevance. Unlike prior work such as MrSteve, which relies on feature similarity for retrieval, WISE enables robust recall under viewpoint changes and supports opportunistic task reordering through causal reasoning. Building on this memory, we propose an Opportunistic Task Scheduler that dynamically re-prioritizes subtasks when causally relevant opportunities are detected. We further equip WISE with a multi-scale progressive exploration strategy to provide spatially comprehensive observations for downstream reasoning. Experiments show that WISE largely improves task success and efficiency on long-horizon sparse tasks, particularly in settings requiring adaptive decision-making.

Wave propagation based on the spectral element method (SEM) is a representative HPC workload, but existing SEM implementations are not well matched to emerging ARM multicore CPUs with Scalable Matrix Extension (SME). We present an SME-enabled optimization of \textsc{SPECFEM3D} on the emerging LX2 processor that combines an SME-aware batched small-matrix kernel for SEM tensor-product operators, a memory-aware hybrid MPI+OpenMP execution scheme for limited-HBM systems, and a dispersion-based iso-accuracy study of the $(h,p)$ tradeoff. At fixed polynomial order, the optimized implementation improves full-application performance by 4--6$\times$ over the original code and delivers clear gains over optimized non-SME CPU baselines. Beyond these implementation-level gains, our results suggest that SME shifts the performance-favorable operating point toward higher polynomial orders along the dispersion-based iso-accuracy frontier, further reducing time-to-solution and working-set size. These results indicate that SME affects not only kernel efficiency, but also the practical discretization tradeoff for SEM on modern ARM multicore platforms.

Semantic mapping is a core service that enables grounded interactions in emerging Extended Reality (XR) applications such as AI assistants and spatial object search. Deploying this capability on mobile XR devices requires a system that is open-vocabulary, real-time, and low-power. Existing approaches are compute-intensive and assume server-class resources. Cloud offloading offers a practical path, but no existing system splits semantic mapping across the device-cloud boundary or manages its communication, execution, and memory footprint. We present SemanticXR, the first device-cloud system for real-time, open-vocabulary semantic mapping and querying under XR power, bandwidth, and memory constraints. Our key insight is to elevate semantically identifiable objects to first-class units of communication, execution, and memory across the device and server. On the server, object-level parallelism and geometry downsampling improve mapping latency, while object-level depth-mapping co-design reduces upstream bandwidth. On the device, an object-level sparse local map with incremental updates and update prioritization enables network-robust querying with bounded memory and downstream bandwidth. Object-level configurable resource usage vs. quality trade-offs let applications and the system adapt mapping to application requirements and operating conditions, respectively. Against a device-cloud baseline with the same perception models, object-level organization improves server-side mapping latency by 2.2X at equal semantic quality. Depth-mapping co-design maintains upstream bandwidth under 2.5 Mbps. On the device, SemanticXR sustains sub-100 ms query latency for up to 10,000 objects even under network drops, supports tens of thousands of objects within 500 MB, and scales downstream bandwidth with map changes, not total scene size. The system adds only 2% device power during normal operation.

Large language models (LLMs) are increasingly used for tasks once reserved for trained researchers, including hypothesis generation, specification choice, and drafting conclusions. We argue that the reliability of AI-assisted research depends not only on model capability, but also on how cognitive labour is structured between humans and machines. We study this problem through Human-in-the-Loop Economic Research (HLER), a decision architecture based on pre-commitment, decision sequencing, accountability, and attention allocation. In a pre-specified 2*4 factorial experiment with 280 complete research runs across four datasets, an unconstrained multi-agent baseline produced critical failures in 72% of runs. Using the same underlying model, the same agent decomposition, and identical prompts for the shared reasoning agents, HLER reduced the failure rate to 16% by imposing three architectural commitments: LLMs reason but do not execute data work, data and estimation are handled deterministically, and three human decision gates bind the workflow. Fisher's exact test rejects equality of failure rates at p<0.001. Reliability gains were largest on the least publicly represented dataset, a Qing-dynasty population register, consistent with a task-based production model with Frechet-distributed output quality. An 80-run ablation suggests that deterministic computation and human gates contribute independently, with exploratory evidence of complementarity. We interpret HLER as a research harness rather than an autonomous AI scientist: it sharply reduces failures, makes residual weaknesses more visible, and prevents unreliable claims from being advanced as publication-ready outputs.

This study explores privacy-preserving machine learning (PPML) techniques using the PySyft platform to enable collaborative prediction of student retention between institutions. We developed a remote data science (RDS) framework with a semi-air-gapped architecture consisting of high-side and low-side servers, allowing researchers from three universities to build predictive models on sensitive student data without direct data access. Using historical data from a small private university (N=720), we evaluated three synthetic data generation approaches and validated the framework through inter-institutional collaboration. The results demonstrate consistent classification performance across institutions (Macro F1: 0.690--0.695) while maintaining strict Family Educational Rights and Privacy Act (FERPA) compliance. We also propose Data-Type-Aware Templates, a novel synthetic data method that prioritizes privacy over distributional fidelity. Our findings confirm that RDS-based PPML is technically feasible for educational settings and offers a practical alternative to federated learning for small-scale inter-institutional collaborations. The code is available at https://github.com/jtfields/NAIRR240195-Privacy-Preserving-Machine-Learning.

We present an interpretable machine learning pipeline to decompose Cross-Sectional Equity Return Predictability into auditable factor contribution. We apply an XGBoost model with TreeSHAP attribution and conduct stress testing on 3632 Chinese A-share stocks from 2009 until 2019. Using 60-month, rolling windows over 55 months of out-of-sample data, XGBoost obtains a mean AUC of 0.547 and +2.38%/month (Newey-West t = 5.94; Annualized Sharpe 2.23) long-short spread for the top vs bottom quintiles. This alpha is persistent after adjusting for the Carhart four-factor model (+2.31%/month; t = 7.48). SHAP Decomposition indicates that behavioral signals (turnover and momentum) account for 58.2% of predictive attribution compared to 10.7% for valuation ratios, on average, across 55 industry groups. Ablation analysis serves to cross-validate this ranking and provides evidence that SHAP and ablation diverge in a manner that highlights feature substitutability structure that is largely invisible to either method used in isolation.

Masked diffusion language models (MDLMs) such as LLaDA now rival autoregressive (AR) LLMs, but every existing knowledge-editing and unlearning method (ROME, MEMIT, etc.) targets AR transformers and either makes assumptions that fail under iterative denoising, or requires gradient updates whose backward-pass activations cost tens of GB of extra VRAM and which collapse MDLMs at standard learning rates. We introduce TimeROME-DLM, the first training-free, gradient-free, inference-time knowledge-editing framework for MDLMs. It couples two components: a Temporal Indirect Effect (TIE) causal-tracing protocol that identifies, for each fact, the coordinate whose intervention most strongly drives the object prediction at later denoising steps; and a closed-form, low-rank residual edit memory that aggregates subject keys and target deltas across all forget facts and applies a single ridge-regularised update at that coordinate at every diffusion forward, with sparsification to limit utility spillover. Backbone weights stay frozen; only three hyperparameters (alpha, lambda, q) are tuned on a small validation split. On TOFU forget01 with TOFU-finetuned LLaDA-8B-Base, TimeROME-DLM cuts forget-set log-probability by roughly 83 nats. The same configuration transfers to LLaDA-8B-Instruct, Dream-7B, MMaDA-8B, DiffuLLaMA-7B, and LLaDA-MoE-1.4B. It keeps retain-set log-probability nearly flat (within ~1 nat at the utility-safe operating point) across 50 sequentially inserted facts, delivers a four- to fourteen-fold wall-clock speedup with zero additional VRAM over the strongest converged training-time baseline, and scales sub-linearly to 400 facts. TimeROME-DLM closes the locate-then-edit gap between AR LLMs and MDLMs at a fraction of the computational cost.

Regression trees are among the most interpretable yet expressive model classes in machine learning. Historically, greedy induction has been the dominant approach for constructing well-performing regression trees. While optimal methods based on dynamic programming and branch-and-bound exist, they are computationally prohibitive for general linear regression trees, despite often achieving substantially better performance than greedy approaches. Recent work has shown that specialized lookahead strategies can dramatically improve runtime while maintaining near-optimal performance, primarily in classification settings. In this work, we develop a novel algorithm for near-optimal, sparse, piecewise linear regression trees that combines a lookahead-style search strategy with efficient rank-one Cholesky updates of the Gram matrix. We demonstrate, both theoretically and empirically, that our method achieves a favorable trade-off between computational efficiency, predictive accuracy, and sparsity, and scales significantly better than the current state of the art.

Predicting single-cell transcriptional responses to genetic, chemical and cytokine perturbations is a fundamental challenge in computational biology and AI Virtual Cell (AIVC) modeling, with direct implications for drug discovery and the elucidation of gene regulatory networks. Existing approaches often rely on auxiliary cell-state encoders, hierarchical variational autoencoders, dedicated Transformer encoder-decoder modules, or gene-interaction priors to compress high-dimensional expression profiles into latent representations. While effective, these designs increase architectural complexity and may limit scalability and generalizability. This paper introduces OCOO-T, a minimalist flow-matching-based AIVC model for transcriptional perturbation response prediction. OCOO-T utilizes a vanilla Transformer stack that operates directly on continuous gene expression profiles and formulates perturbation response prediction as a continuous-time denoising process. Perturbation embeddings, dosage information, and cell-line/cell-type specificity are integrated through adaptive layer normalization and in-context tokens. Comprehensive evaluations on Tahoe100M, Replogle, and PBMC benchmarks demonstrate that OCOO-T achieves state-of-the-art performance across diverse perturbations and cell types while effectively scaling to long transcriptional profiles through patching and depatching of cellular contexts. By leveraging the simplicity of Transformer-based denoising for single-cell omics, OCOO-T provides an effective and scalable framework for in-silico cellular simulation.

Search agent benchmarks exemplified by BrowseComp have rapidly saturated over the past year, with the strongest models surpassing 90% accuracy. Since these benchmarks are predominantly human-authored, annotators lack a global perspective on entity statistics and cannot systematically maximize search space size and structural complexity. This creates a difficulty ceiling that is hard to break. To address this, we introduce LoHoSearch (Long-Horizon Search Agents), a challenging benchmark comprising 544 human-verified questions across 11 domains. LoHoSearch is constructed via an automated pipeline built upon a knowledge graph covering over 7 million Wikipedia entities, which selects relations with large search spaces and assembles them into structurally complex questions with KG-verified unique answers. Our evaluation demonstrates that even the strongest model achieves only 34.74% accuracy, and existing context management strategies (best +6.8%) yield far smaller gains than on prior benchmarks. LoHoSearch provides a more demanding standard for evaluating long-horizon reasoning and context management in search agents.

The rapid emergence of autonomous AI agents is transforming artificial intelligence from isolated model inference into distributed systems of reasoning, communication, and action. This paper develops the vision of the Internet of Agentic AI (IoAI): an open ecosystem in which heterogeneous agents discover one another, negotiate responsibilities, exchange context, invoke tools, and execute workflows across cloud, edge, device, organizational, and cyber-physical environments. We synthesize foundations from single-agent agentic AI, multi-agent systems, distributed computing, communication networks, game theory, and security engineering to characterize the architectures and mechanisms required for scalable agent ecosystems. The paper examines agent deployment models, workflow lifecycles, communication protocols, interoperability layers, resource-management challenges, and trust architectures, with case studies in adaptive manufacturing and distributed operational coordination. The resulting framework highlights the central research challenges of controlled emergence, semantic interoperability, secure identity, incentive-compatible coordination, resource-aware orchestration, and governance for large-scale networks of autonomous agents.

As scientific workflows shift from deterministic executables to LLM-based agents, the development practices on offer, such as fine-tuning, reinforcement learning, and prompt-and-go, bury the scientist's judgment. We propose treating agent construction as a workflow stage and introduce AgentBuild, which builds a scientific agent from a contract the scientist authors. The contract is a version-controlled rubric, a difficulty-graded curriculum, and a curated external knowledge base. A rubric-driven judge gates a meta-optimizer coding agent that edits the agent within a declared boundary, so the build compiles the agent, not the scientist's judgment. We instantiate this for Rietveld refinement of X-ray diffraction data through GSAS-II behind MCP and A2A, where a blank-harness construction run progresses through a lithium lanthanum zirconium oxide (LLZO) signal-to-noise ladder, reaches the 4 hour scan as a frontier case, and exposes the workflow-scope limits that remain. The same rubric that rewards credible fits also scores trajectory scope, making the frontier a contract failure rather than a pattern-fitting failure. As base models evolve, re-running AgentBuild is a re-tune, not a rebuild, and the scientist's authored contract remains the durable asset.

While recent vision-language models (VLMs) demonstrate strong multimodal understanding, they remain limited in spatial reasoning tasks that require active evidence acquisition and multi-step visual interaction. This limitation suggests that relying solely on implicit visual representations from vision encoders is insufficient for recovering fine-grained spatial evidence. We introduce PERception-Interaction-reason Agent (PERIA), a tool-augmented visual agent for spatial reasoning tasks across map reasoning, visual probing, and vision reconstruction. PERIA uses two lightweight tool families: vision perception tools for exposing textual, symbolic, and spatial evidence, and vision interaction tools for manipulating visual context, tracing paths, and verifying spatial relations. To train PERIA, we develop a unified recipe that combines supervised tool-use trajectory synthesis, composite rewards, and Observation-Relaxed Group-in-Group Policy Optimization (OR-GIGPO) for effective multi-tool behavior. Experiments on 13 benchmarks from 8 datasets show that PERIA-8B improves over the Qwen3-8B backbone by 10.0% on in-distribution benchmarks and 4.4% on out-of-distribution benchmarks, while outperforming previous state-of-the-art baselines of similar size by 7.0%-14.8%. It also achieves performance comparable to much larger models such as Qwen3-VL-235B-A22B-Thinking and GPT-5, demonstrating the effectiveness of PERIA in enhancing spatial reasoning capabilities.

Do research topics in artificial intelligence grow gradually, or do they advance through abrupt, detectable jumps? Analyzing 80,814 accepted main-track papers from five premier AI conferences (ACL, CVPR, ICLR, ICML, NeurIPS) spanning 2017 to 2025, we show major AI topics advance through topical phase transitions: remaining marginal for years, then surging across venues within one to three years. Large language models became the dominant cross-venue topic by 2025, diffusion models rose with comparable abruptness, and language-model methods crossed into computer vision via vision-language models, whereas reinforcement learning compounded smoothly, distinguishing genuine phase transitions from ordinary growth. This structure is our primary contribution: a large-scale, cross-venue characterization of how AI research reorganizes. We then ask whether a transition leaves a detectable footprint before it peaks. We define an early-warning signature, four publication-dynamics criteria frozen on 2017-2021 data, and evaluate it out of sample on 2023-2025 transitions, obtaining a precision of 27% and recall of 63% against a 13.5% base rate. Applied to 2025 data, the signature flags reasoning and test-time compute, agentic AI, multimodal LLMs, retrieval-augmented generation, and world models as topics to monitor over 2026-2028. The source code is also publicly available on GitHub at https://github.com/KurbanIntelligenceLab/ai-phase-transitions.

Moving instance segmentation (MIS) attracts increasing attention due to its broad applications in traffic surveillance, autonomous driving, and animal tracking. Event cameras record asynchronous brightness changes, providing high temporal resolution and dynamic range, which makes them highly sensitive to motion information. By fusing event and image features, motion cues from events can complement spatial details from images, enhancing the performance of MIS. However, current multimodal MIS methods still struggle to segment small moving instances, as event cameras often yield sparse features under limited resolution. Moreover, event features entangle appearance attributes with motion cues, which further restricts effective cross-modal fusion. To address these challenges, we first propose a dual-disentangling feature extraction framework that separates and extracts appearance and motion information within both image and event modalities, thereby improving feature density. Subsequently, a multi-granularity cross-modal alignment is introduced to align distributionally and semantically consistent features across modalities, enabling more effective fusion with rich spatial and temporal details. The experiment results demonstrate that our method achieves state-of-the-art performance in multimodal MIS, especially for small instances under challenging conditions such as fast motion and low-light settings.

AI governance for medical imaging is formalizing: the 2026 ACR-SIIM Practice Parameter recommends local acceptance testing and ongoing drift monitoring, and the ACR Assess-AI registry monitors AI outputs using DICOM metadata for context. We argue that a necessary, currently unmonitored layer sits beneath output metrics: whether incoming studies remain within the acquisition envelope a model was validated on. Using a LUNA16-trained MONAI RetinaNet lung-nodule detector, we test whether acquisition state behaves as a structured, measurable variable. On real paired CT differing only in reconstruction kernel (NLST B30f vs B80f), kernel alone shifted AI-measured diameter and flipped a Fleischner size category in 5.2% (8 of 155) of nodules at fixed patient and acquisition, while detection confidence was unchanged (Wilcoxon p=0.22). Under controlled LIDC-IDRI perturbations the effects dissociated by axis: the noise axis degraded detection confidence (p=5.9e-32, concentrated in nodules under 6 mm) but not measurement, while the frequency/kernel axis corrupted measurement (p=8.6e-13) but not detection. A 4-feature pixel fingerprint recovered reconstruction identity (patient-level AUC about 0.95 on real CT, 0.995 on a QIBA phantom) where the ConvolutionKernel DICOM tag was uninformative (identical labels across reconstructions). The kernel axis transported across four manufacturers (leave-one-vendor-out AUC 0.94-0.98, matching the within-vendor ceiling). Acquisition state thus maps to distinct AI failure modes, frequency content to measurement reliability and noise to detection sensitivity, and is not recoverable from metadata. Acquisition-aware, input-side validation is the missing layer for the acceptance-testing and drift-monitoring requirements now entering imaging-AI accreditation.

Environmental scientists spend disproportionate effort on data wrangling rather than analysis, and AI agents that automate geospatial workflows remain unvalidated: no benchmark evaluates agents operating through structured tool calling against real APIs. We introduce the GeoNatureAgent Benchmark, the first benchmark for environmental analysis agents that operate via structured tool calls to a production-style geospatial API. It comprises 93 tasks across 18 categories, covering municipality analysis, multi-turn conversation, spatial reasoning, cross-indicator synthesis, error handling and recovery, ranking, comparison, multilingual understanding, habitat analysis, and task rejection. Tasks are evaluated against an open, self-hostable API serving three environmental indicators across Spain and Portugal via sixteen tools. We evaluate seven LLMs (Claude Sonnet 4, DeepSeek V3.2, GLM-5, Gemini 2.5 Pro, Qwen3-235B, GPT-OSS-120B, Llama 4 Scout) under three temperature-1.0 seeds, reporting capability and per-case cost as orthogonal axes. We find: (1) Claude Sonnet 4 leads at 60.8% +/- 0.8%, followed by DeepSeek V3.2 at 56.3% +/- 3.1%, with no other model above 51%; (2) the cost-accuracy Pareto frontier is occupied mostly by open-weight models, with DeepSeek V3.2 offering 93% of Claude's capability at 11x lower cost ($0.011/case); (3) comparison tasks remain universally unsolved (0% on close-value comparisons), exposing systematic reasoning limits; and (4) structured tool calling against a real API is more discriminative than general-purpose GIS benchmarks, with accuracies 25-35 points lower. We further show extensibility by integrating BigEarthNet V2 land cover for Portugal alongside Spanish CO2 and erosion indicators. The benchmark, harness, and self-hostable API are publicly available.

Irrelevant numbers in a prompt can shift language model judgments, producing anchoring effects in numerical reasoning. We study where this anchor-sensitive signal is carried inside language models using a controlled multiple-choice setup with shared answer options. We define a logit-difference metric comparing the correct answer option with the answer option corresponding to the anchor, and validate that it tracks behavioral anchoring. Using attribution-based circuit localization on 7B--8B Qwen and Llama base and instruction-tuned models, we find that edge-level methods recover this signal more faithfully than node-level methods. Low- and high-anchor circuits transfer strongly within a model, suggesting shared pathway structure across anchor direction. However, sparse transfer across base and instruction-tuned variants is less reliable, indicating that post-training changes which pathways matter most. Overall, our results provide a mechanistic account of how anchoring-related decision signals are carried inside language models.

Understanding the digital world on mobile devices is shifting from static UI perception to dynamic action comprehension. This capability enables models to convert visual state transitions into operational knowledge, defined as short natural-language sentences that describe action types, target UI elements, textual arguments, and execution orders. However, due to the highly diverse and heterogeneous UI designs across applications, existing vision-language models (VLMs) struggle to accurately infer these underlying operations. To bridge this gap, we introduce Teach VLM, a core model designed to translate mobile screen trajectories into step-wise operational knowledge by extracting and analyzing operation-related keyframes from demonstration videos. To address the scarcity of aligned training data, we develop a systematic data flywheel for scalable data acquisition. We further introduce a novel Chinese Mobile Screen Teach Benchmark for fine-grained evaluation. Building upon Teach VLM, we propose the Teach-and-Repeat paradigm, where the generated operational knowledge serves as an interpretable procedural reference to guide downstream screen-based execution agents. Extensive evaluations demonstrate that Teach VLM significantly outperforms strong VLM baselines, achieving state-of-the-art performance in operation semantics prediction. Furthermore, experiments in Android World show that our paradigm yields consistent Task Success Rate improvements for downstream agents. Together, Teach VLM and the Teach-and-Repeat paradigm offer a practical pathway from raw demonstrations to reusable task automation.

Quantum circuit routing is a key step in compiling programs for noisy intermediate-scale quantum processors. Routes that appear efficient by standard overhead metrics can still lose fidelity when they pass through poorly calibrated couplers. We study a calibration-aware graph reinforcement-learning router that uses same-day IBM Heron r2 calibration data to choose hardware-edge SWAPs. We train the policy with proximal policy optimization and evaluate it with exact simulated fidelity across nine Munich Quantum Toolkit (MQT) Bench circuits and three calibration snapshots. Across these evaluations, pooled mean exact fidelity is $0.727$, compared with $0.440$ for SABRE-best20 and $0.481$ for target-aware SABRE. Fidelity gains come with higher routed two-qubit counts and are concentrated in the 5q and 8q circuit families; under the fixed tree action graph, all 10q families favor SABRE-best20. Overall, our results show that calibration-aware learned routing can improve fidelity beyond gate-count-driven compilation.

Recent reinforcement learning approaches have shown great promise in improving humanoid motion tracking performance and achieving fall recovery under disturbances. However, most existing works treat motion tracking and fall recovery as different tasks and require multi-stage training with specialized recovery rewards and/or separate recovery policies. Moreover, existing reinforcement learning-based methods often terminate training episodes immediately after severe tracking failures, limiting recovery-oriented exploration in unstable or fallen states. To address the above issues, we propose Stubborn, a streamlined and unified reinforcement learning framework to achieve robust humanoid motion tracking and fall recovery. Specifically, Stubborn uses an asymmetric Actor-Critic architecture and consists of three major components. First, a yaw-aligned tracking representation is adopted to reduce sensitivity to global drift and heading disturbances while preserving gravity-related balance information. Second, we introduce a Bernoulli-based probabilistic termination mechanism that enables the policy to encourage exploration of fall-recovery behaviors under varying failure modes. Third, we propose a probabilistic termination and tracking-error-driven strategy that dynamically reshapes the sampling distribution based on tracking performance, increasing the training efficiency for difficult motion segments and unstable states. Extensive comparisons with SOTA methods and ablation studies show that Stubborn achieved competitive performance, and the proposed probabilistic termination mechanism and adaptive sampling strategy contributed to the performance and robustness gains. For real-world demonstrations, please refer to https://aislab-sustech.github.io/Stubborn/.

Multimodal large language models (MLLMs) are trained on massive multimodal data, making data unlearning increasingly important as data owners may request the removal of specific content. In practice, these requests often arrive sequentially over time, giving rise to the challenging problem of MLLM Lifelong Unlearning. However, most existing benchmarks are limited in scale and scope, failing to capture the complexities of MLLM lifelong unlearning. To fill this gap, we introduce the MLUBench, a large-scale and comprehensive benchmark featuring 127 entities across 9 classes under lifelong unlearning requests. We perform extensive experiments using MLUBench and reveal that existing unlearning methods suffer from severe, cumulative degradation. More critically, we further identify the unique challenge of this problem: unlike in unimodal models, MLLM lifelong unlearning is constrained by the need to preserve multimodal alignment. Continually unlearning from one modality could degrade the entire model. To alleviate this challenge, we propose LUMoE, an effective method. Experiments demonstrate that LUMoE significantly mitigates the degradation problem faced by baselines. The source code and the MLUBench dataset are open-sourced in https://github.com/lihe-maxsize/Lifelong_Unlearning_main.

Adaptive Hamiltonian learning is central to calibrating and characterizing quantum devices. In an adaptive controller, choosing the next experiment is itself a computation. Bayesian design rules are recomputed after every posterior update, and that step can take seconds. Across hundreds of shots, those seconds become a significant wall-clock cost for adaptivity. We introduce SymQNet, an amortized reinforcement-learning approach for low-latency adaptive Hamiltonian learning. SymQNet learns a posterior-conditioned acquisition policy offline, then uses a fast policy forward pass online while retaining Bayesian posterior feedback. On transverse-field Ising benchmarks, SymQNet substantially reduces acquisition latency relative to bounded Fisher-information search and bounded two-step Bayesian active learning by disagreement (BALD). At five qubits, it reduces acquisition-only decision latency by $47.1\times$ and $72.6\times$ relative to these online baselines; at twelve qubits, full simulated steps take $1.02$ s for SymQNet versus $13.27$ s for bounded two-step BALD. Overall, we show that learned acquisition can make adaptive Hamiltonian learning practical for repeated low-latency workloads.

Summaries of real-world events can become outdated as contexts evolve and new information arrives. A common response is to generate a new summary from the updated context, but full regeneration discards the previous draft, can obscure what changed, and may be unnecessary when only a few claims are unsupported. We study localized faithfulness repair: updating outdated spans in an existing summary while preserving supported content. We propose DETECT-REMASK-REPAIR, a diffusion-based framework that identifies, remasks, and repairs outdated regions with masked diffusion language models. To evaluate evolving-context summarization, we introduce StreamSum, a benchmark of synthetic event timelines. Experiments on DialogSum and StreamSum show that localized diffusion repair provides a controllable alternative to full rewriting: faithfulness-steered repair improves early drafts, one-step repair reduces repair cost to under half a second, with the framework enabling faithfulness-speed-preservation tradeoffs across datasets. We also find that the framework can provide a post-hoc correction step that improves faithfulness for autoregressive systems.

Short-term load forecasting is essential for reliable energy management, but practical deployment on edge devices requires models that remain accurate under limited memory, finite measurement budgets, and hardware noise. This work proposes a hardware-efficient Quantum Reservoir Computing (QRC) framework for energy load forecasting, where a fixed quantum reservoir transforms temporal input windows into high-dimensional features and only a classical Elastic Net readout is trained. To reduce deployment cost, the trained readout is compressed using post-training fixed-point quantization at bit widths from 8 to 2 bits. The framework is evaluated on the Tetouan and Spain energy load datasets under exact statevector simulation, 512-shot finite sampling, and realistic hardware-noise models from IBM FakeTorino and IBM FakeMarrakesh. Results show that 6-bit readout precision preserves full-precision forecasting performance while reducing readout memory by 81.2%. Below this point, degradation becomes dataset dependent, with Tetouan showing stronger sensitivity and Spain degrading more gradually. Hardware-noise validation further shows that the trained readout transfers to noisy reservoir states without retraining. These findings support quantized QRC as a resource-aware forecasting approach for near-term quantum time-series applications.

Collaboration is widely recognized as a cornerstone of 21st-century education, yet teachers still encounter persistent challenges in fostering productive peer interaction. LLM conversational peer agents introduce new possibilities for mediating in-person group work, raising questions about how persona design, particularly their voice characteristics, shapes learners' perceptions, trust, and interactional dynamics. While prior work has examined agent accent effects in one-to-one settings, little is known about how these effects manifest in groups. We conducted a between-subjects mixed-methods study with 33 teachers examining how a GenAI voice agent with different accents (British, Indian, and African American) influenced collaboration and agent perception. Across surveys, group interaction analyses, and artifacts, we find that accent shaped participants' mental models and the roles the agent assumed in group interaction. The British-accented agent was largely treated as a tool and engaged in detached, utility-based ways, whereas Indian- and African American-accented agents were more readily anthropomorphized and integrated as peers. These role expectations influenced trust, engagement, and reliance over time. This work advances understanding of how GenAI's sociolinguistic design features shape group dynamics in CSCL, with implications for designing culturally inclusive AI partners in group learning.

Agentic large language model systems that autonomously invoke tools, maintain persistent memory, and execute multi-step plans are increasingly deployed in public-facing domains, including government services, healthcare triage, and financial advising. We ask whether the frameworks used to build these systems provide architectural-level structural safety guarantees. Applying six containment principles derived from a compositional model of agentic architectures, we audit three dominant frameworks (LangChain, AutoGPT, and OpenAI Agents SDK) and find no native compliance in any of them. Memory integrity, a defense against one of the most prevalent vulnerability classes, is not observed in any of the three evaluated frameworks. We validate these findings empirically: in a simulated government benefits agent built on LangChain, a single memory-poisoning write induces persistent targeted corruption across all tested seeds and backends, increasing the wrongful denial rate for targeted applicants to 88.9%. Under a complex five-factor policy, the same attack preserves aggregate accuracy while increasing targeted wrongful denials by 3.5x, rendering the corruption difficult to detect through standard monitoring. We then introduce two lightweight containment mechanisms: a memory integrity validator and a policy gate, which eliminate both attack vectors with sub-millisecond overhead (<0.2ms per call). We conclude that the current agentic framework ecosystem may not yet meet secure-by-default expectations for public-facing deployments and outline priority architectural interventions to enable trustworthy deployment in high-stakes, socially impactful applications.

Large Language Models have consistently demonstrated a lack of creativity and diversity across tasks. Prior work has focused on addressing whether models are capable of generating creative outputs. Here, we aim to consider novelty and investigate what makes model-generated content novel or not novel in a task-specific manner. We propose a fine-grained evaluation metric GENIE to measure the novelty of responses along task-specific features with respect to a population of responses. We show that unlike GENIE, holistic metrics struggle to capture the high-dimensionality of novelty and do not provide insight on which properties they target. Finally, we use GENIE to measure the effectiveness of mitigation methods that address creativity to better understand where these methods can improve novelty.

Evaluating retrieval-augmented generation (RAG) systems requires benchmarks that capture diverse question characteristics, yet practitioners lack empirical guidance on which dimensions to vary and at what granularity. We present HieraRAG, a hierarchical framework for studying granularity in RAG benchmark construction, defining optimal granularity as the level that maximizes discriminative power (the standard deviation of generation quality across categories) within a given RAG configuration. As a case study, we generate 5,872 synthetic question-answer (QA) pairs from FineWeb-10BT across 3 dimensions (Question Complexity, Answer Type, Linguistic Variation) at 3 granularity levels (2, 4, and 8 categories). With a BM25+Falcon-3-10B pipeline, optimal granularity varies by dimension: complexity benefits from fine-grained distinctions (discriminative power: 0.053) while answer type and linguistic variation peak at medium granularity. We introduce a Coherence Ratio metric to quantify whether fine-grained splits cleanly subdivide parent categories, revealing structural differences across dimensions (Question Complexity: 0.40 vs. Answer Type: 1.44). Human evaluation of 110 stratified QA pairs confirms synthetic quality. While these specific findings reflect a single configuration, HieraRAG provides a portable procedure and validation metric for practitioners to determine evaluation granularity within their own RAG settings.

As the EU Carbon Border Adjustment Mechanism (CBAM) approaches, the global semiconductor value chain faces growing structural tensions between regulatory transparency and data sovereignty. This article proposes a RegTech reference architecture using the International Data Spaces (IDSA) framework to orchestrate trustworthy environmental telemetry across the semiconductor-petrochemical nexus. The framework distinguishes the mandatory CBAM requirements from voluntary Science Based Targets initiative (SBTi) frameworks, while addressing the additive complexities of the Safe-and-Sustainable-by-Design (SSbD) framework. Moving beyond standard linear technology stacks, we introduce a prospective roadmapping methodology that transforms upstream physical vulnerabilities into circular, negative feedback loops. Focusing on the Taipei and Penang technology corridor, the article details how sovereign data exchange enables Digital Product Passports (DPPs) to drive Global Business Services (GBSs) capability demands. Finally, we discuss the integration of Agentic AI for autonomous compliance and FinTech green financing, providing a scalable blueprint for global industrial clusters to achieve sovereign, sustainable, and transparent value chains.

World modeling is emerging as a central principle for building intelligent systems capable of prediction, reasoning, and decision making. A central distinction can be drawn between explicit world models, which learn structured dynamics for rollout-based reasoning and planning, and implicit world models, which encode predictive structure within scalable learned representations. These complementary paradigms provide a foundation for physical AI in domains such as robotics and autonomous driving, enabling intelligence beyond reactive control under real-world constraints. Recent foundation models further suggest a pathway toward unified systems integrating perception, prediction, and action. Despite rapid progress, major challenges remain in hierarchical reasoning, long-horizon planning, and autonomous goal formation, which are critical for advancing toward artificial general intelligence. This tutorial presents a coherent framework in which diverse world modeling approaches are unified through shared predictive structure and differentiated by how such structure is represented and exploited.

Self-evolving agents are expected to improve through interaction without external supervision, but this remains difficult in partially observable environments where agents must explore actively, learn from limited feedback, and decide when to trust prior experience. Existing LLM-agent methods often rely on memory or planning modules, yet they rarely close the loop between them to continually refine an internal understanding of environment dynamics. We introduce ProPlay, a procedural world model that supports procedure-level preplay, where agents can rehearse future procedural paths using the learned world knowledge. Rather than representing experience as isolated rules or low-level action constraints, ProPlay abstracts successful trajectories into procedures and organizes them in a procedure graph that captures causal transitions among task stages. Each transition is associated with a reliability record embedding to estimate its task-specific contribution from past outcomes. Before each episode, ProPlay simulates future procedural trajectories over known graph structures as structured soft guidance; after execution, it refines the graph using environment feedback. Experiments on public benchmarks show that ProPlay consistently improves environment understanding and self-evolution capability over strong baselines. Our code has been released in https://github.com/antman9914/proplay.

While MPC effectively handles structured, diverse, and low-level specifications, it lacks the capability to dynamically incorporate high-level contextual information such as social norms, user intent, or natural language instructions. To address this limitation, this manuscript introduces an agentic MPC framework that enables context-aware, semantically adaptive control synthesis by integrating with large language model-based agents. The agent interprets heterogeneous inputs, including natural language messages, environmental observations, and external knowledge, to resynthesize the control specifications. The effectiveness of the framework is demonstrated in an autonomous driving scenario, where the system aligns with personal preferences or responds to social situations such as emergency vehicle yielding.

Evaluating procedural reasoning in AI-supported learning systems requires question-answer datasets that are both learner-like and grounded in the instructional knowledge the system is expected to use. We study how TMK-based question generation strategies affect dataset quality for procedural and multi-hop reasoning. We compare three strategies: strict generation from Task-Method-Knowledge (TMK) models, transcript-first generation with post-hoc TMK filtering, and TMK-aware generation that combines transcripts with structured guidance. To evaluate generated items, we introduce a grounding validation framework based on closed-set evidence units extracted from TMK models. The framework measures whether answers are supported by the underlying representation, whether questions are self-contained, and whether they target multi-hop procedural reasoning. Across 23 instructional topics and 690 generated question-answer pairs, strict TMK generation achieves the strongest overall quality, with 96.5% grounded questions and 92.6% usable questions. Transcript-first generation produces more learner-like questions but more context-dependent or weakly grounded items, while TMK-aware generation yields high raw multi-hop coverage but lower grounding. These results show that procedural richness and natural phrasing do not guarantee representational grounding, motivating explicit representation-aware validation for evaluation datasets in AI-supported learning.

Apple's Metal 4.1 exposes a tensor compute path: the Metal Performance Primitives (MPP) matmul2d operation over cooperative_tensor fragments, whose interface is documented but whose hardware behavior is deliberately hidden. The specification states which data-type rows are supported, never whether they are hardware-accelerated, where the operation physically executes, what its accumulator width is, or how it partitions matrix fragments across threads. We present Rigel, an empirical characterization of this path on a single Apple M4 Max (a pre-neural-accelerator generation). Using a checksum-gated, provenance-tracked microbenchmark harness, Rigel recovers eleven facts the v4.1 specification hides or contradicts. The headline finding: the Metal 4.1 fp8 (E4M3) matmul2d is emulated, not accelerated: it sustains 0.94x the throughput of fp16 despite reading half the operand bytes, so on M4 it is a memory-footprint feature, not a performance feature. We further show, via a three-signal triangulation (throughput ceiling, comparison against simdgroup_matrix, and per-rail power attribution), that matmul2d executes entirely on the GPU shader cores with no dedicated matrix datapath and no evidence of Apple Neural Engine routing; that it accumulates in >=fp32; and we reconstruct the opaque 8x8 cooperative_tensor fragment layout Apple documents nowhere. Acting on the characterization, a hand-fused GEMM + bias + GELU kernel beats the decomposed path by +6.5-12.9% in the cache-resident regime. All findings are reproducible from committed MIT-licensed code and per-cell CSVs.

Large language models (LLMs) are increasingly used to generate code at scale. Meanwhile, prior work has investigated whether training data may be recoverable from model outputs, by auditing the textual overlap between training examples and model generations. Code, however, can be functionally equivalent while textually dissimilar. In this work, we study functional memorization: extraction of functional logic beyond what verbatim metrics detect. We construct a counterfactual setup for Olmo-3-32B, comparing a midtrained model (exposed to target code) against a pretrained reference (not exposed). We prompt both models with Python function signatures and measure both textual and functional similarity (i.e., LLM-as-a-judge, execution-based). Our results show clear evidence of functional memorization, highlighting the need for auditing metrics that go beyond textual overlap.

We study the problem of selecting the largest among $n$ unknown values $x_1,\dots,x_n$ given only a single unbiased estimate $y_i$ for each $x_i$. We design strategies that are simultaneously admissible (not uniformly dominated by any other strategy) and also never worse than a given baseline such as uniform random selection. We provide an application to stochastic optimization, where we obtain online-to-batch conversion bounds with a desirable "no-compromise" guarantee: they are never worse than standard random iterate selection, and yet can be significantly better in benign settings.

Are large language models (LLMs) bad at capturing human judgment? Two commonly stated limitations are that LLMs fail to capture full distributions of responses, and that their judgments are unstable across wording variations. We demonstrate simple prompting strategies that mitigate these limitations. Across two datasets--a U.S.-representative set of 144 moral scenarios and 38 moral beliefs from the International Social Survey Programme's Family and Changing Gender Roles module covering 32 countries--we show how simple elicitation techniques help improve AI-human alignment. First, prompting models to report standard deviations and response proportions recovers the full range of human responses better than common strategies. Second, ensuring scenarios are clear to human participants--as reflected in human confusion ratings--boosts model alignment, and LLMs can track human confusion ratings. At the same time, we find that LLMs' estimates of their own error are poorly calibrated, though they can predict human variability relatively well. These results suggest that asking better questions to LLMs can yield better answers.

The transition of scientific applications to GPU-accelerated exascale systems is constrained by trade-offs between performance, portability, and productivity. This work evaluates the performance portability of directive-based GPU programming by porting gPLUTO, a production-grade magnetohydrodynamics code for astrophysical simulations, from OpenACC to OpenMP, and analyzing its performance on NVIDIA A100 (Leonardo Booster) and AMD MI250X (LUMI-G) devices. On NVIDIA platforms, OpenACC and OpenMP achieve comparable performance due to a shared compiler backend, providing a consistent baseline for assessing algorithmic efficiency. In contrast, the same OpenMP implementation is approximately three times slower at the application level on AMD MI250X with respect to the NVIDIA A100 OpenACC baseline, with kernel-level slowdowns reaching up to an order of magnitude, driven by sensitivity to strided memory-access patterns and compiler limitations. Kernel-level profiling shows that the dominant contributors to run-time are memory-latency-bound rather than limited by peak band-width. In low-parallelism kernels, C++ abstraction layers increase register pressure and spilling, leading to extreme slowdowns of up to 47x in specific cases. These results indicate that portable performance across GPU architectures requires not only application-level changes but also continued advances in compiler backends and architecture-aware optimization strategies

Why is the past of English "go" the apparently unrelated "went"? Such alternations are frequent in languages. They neither aid communication nor learnability, yet they can be persistent, surviving over centuries or millennia. We present a multi-agent simulation of the emergence of morphological stem and inflection alternations. Alternate forms arise by phonological changes or, as with "go/went", from lexical alternatives associated with a subset of the population. When an agent 'hears' another agent use a novel form for a slot in the paradigm of a word (say, the past tense of go), they will with some probability adopt that form, possibly spreading its use to other slots in the paradigm that shared the same original form. Thus alternative forms can spread through the population and become entrenched as stem or inflectional marker alternants. Unlike many previous computational studies, our system allows for naturalistic lexical forms, realistic phonological rules, lexicons with hundreds or thousands of entries, and agent populations in the tens or hundreds. It supports several network topologies, diffusion patterns and agent adoption policies. One issue with such simulations is evaluation: how realistic is the resulting morphology compared to those of real languages? We introduce the AI Historical Linguist, a novel Large Language Model-driven system that models a debate between two historical linguists. We use this to compare a set of real language morphologies, disguised morphologies, and experimentally evolved morphologies. The results suggest that among the factors that favor more plausible morphologies are scale-free social networks and random Bernoulli adoption of forms. We also present three case studies modeling attested historical changes, allowing us to test what might have happened if history had been different. All code and data are released.

Safety-relevant studies of language models, including alignment and jailbreaking evaluations and AI control protocols, often rely on prefilling model outputs. If AI models can recognize and act on the fact their prior assistant messages have been inserted or edited, the effectiveness and validity of these methods could be compromised. We investigate whether frontier language models can distinguish between tampered and untampered assistant-side context, a capability we call prefill awareness. To do so, we construct a binary preference benchmark across three prefill mechanisms, filtering for cases where models show consistent stances. We find that frontier models show substantial prefill awareness: Claude Opus 4.5 detects prefills opposing its preferences in 9-35% of cases with a 0% false positive rate when prompted; additionally, models often revert towards baseline behavior without explicitly reporting that the prefill was foreign. Controlled ablations later also show that detection and resistance rely on different cues, where stylistic mismatch mainly affects whether models flag a prefill as foreign, while preference mismatch mainly affects whether they revert toward their baseline answer. We also examine more realistic agentic settings such as misalignment-continuation evaluations and SWE-bench trajectories, where frontier models sometimes disavow prefilled assistant turns in ways that depend strongly on dataset, task success, and hidden formatting artifacts. Our results indicate that prefill awareness is already a substantial confound for some prefill-based methods. We recommend that model developers track this capability in frontier systems.

Wearable healthcare devices are the fastest-growing Internet of Things (IoT) sector. Many automated healthcare services rely on two crucial biological signals, namely ECG and EEG, which reflect the activity of the heart and brain, respectively. Although deep neural networks are considered the primary way to process and analyze these signals, the very tight energy and computational power constraints in wearable devices are far below the computational, energy, and memory bandwidth demands of DNN models, thereby impeding the deployment of deep learning in many practical wearable services. This paper investigates the feasibility of deploying state-of-the-art DNN models in resource-constrained wearable devices. Notably, we explore the trade-off between accuracy and computational complexity of DNNs when parameter quantization and electrode reduction methods are used. Our investigation centers on several state-of-the-art DNN models designed for EEG signal analysis, specifically for detecting epileptic seizures. Our findings demonstrate that, when applied judiciously, these techniques can significantly reduce the complexity of the DNNs under consideration with minimal adverse effects on accuracy. These results reveal the explicit trade-offs between accuracy and complexity reduction encountered when adapting DNN-based online EEG analysis for wearable devices.

The convex Latent Optimal Partition (LOP)-l2/l1 approach enables block-sparse signal recovery with unknown partitions but relies on manual hyperparameter tuning. Additionally, numerical instability in differentiating its proximal operator prevents its automatic parameter tuning via Deep Unfolding (DU). To address these limitations, we propose two architectures: a stable framework utilizing implicit differentiation and a flexible variant leveraging Deep Weight Factorization (DWF). The DWF-based approach also supports nonconvex smooth data fidelity terms. Numerical experiments demonstrate that DU-LOP-l2/l1 yields competitive performance and high resilience against impulsive noise.

Large Language Models (LLMs) are rapidly evolving into agentic systems that interact with external tools and environments, introducing new security risks such as indirect prompt injection attacks through untrusted external sources. Existing defenses mainly focus on blocking malicious content at inference time, and current red-teaming methods primarily optimize attack success. As a result, developers have limited visibility into how latent prompt injections emerge and propagate through agents. We propose PI-Hunter, an automated agentic auditing framework for proactive vulnerability exposure in LLM agents. PI-Hunter constructs realistic source-aware test cases and iteratively evolves them through feedback-driven exploration to induce agents to retrieve and reveal latent malicious instructions embedded within external environments. Extensive experiments across multiple benchmarks, agent architectures, attacks, and defenses demonstrate that PI-Hunter substantially improves vulnerability exposure and attack-surface coverage over strong automated red-teaming baselines, while remaining effective under existing prompt injection defenses.

AI agents are increasingly being developed to accelerate scientific discovery, yet their practical capabilities in real research settings remain poorly understood. Existing benchmarks for AI agents rarely capture the complexity, heterogeneity, and extended reasoning required by scientific work, whereas benchmarks for scientific tasks often reduce research to static, direct problems and provide limited support for interactive evaluation. Here, we introduce SciAgentArena, a systematic benchmark for evaluating AI agents in real-world scientific research scenarios drawn from emerging needs across multiple domains. SciAgentArena comprises approximately 200 tasks with stepwise verification and an interactive, agent-agnostic environment for assessing diverse AI agents. Using this benchmark, we find that current agents can contribute effectively to well-specified data-analysis workflows, particularly when the task structure and evaluation criteria are clear. However, their performance remains uneven across scientific contexts: agents struggle to generate genuinely novel insights, sustain self-directed exploration, and formulate robust solutions for open-ended research questions. We further characterize common failure modes across agents and identify opportunities for improving their reliability, autonomy, and scientific reasoning. Together, SciAgentArena provides a practical framework for measuring progress in AI agents for science and for guiding the design of future agents capable of addressing complex scientific challenges. Full codes, tasks, and datasets can be accessed via this link: https://sciagentarena.github.io/.

Physics-Informed Neural Networks (PINNs) are a machine learning method for solving forward and inverse Partial Differential Equations (PDEs). When applied to PDEs with Dirac delta functions in the forcing terms, boundary conditions, or initial conditions, PINNs require approximating them with smooth surrogate functions, a practice that can introduce significant modeling errors. In this work, we exploit the interpretation of PINNs as Residual Least Squares (RLS) methods and show that this perspective enables direct treatment of Dirac delta terms by integrating the weak-form equation. Among RLS formulations other than PINN, we focus on the Radial Basis Function (RBF) expansion (also known as a single-layer RBF Network). We show that while integrating out the Dirac delta in PINNs causes residuals to fail to converge to zero, RBF-RLS consistently provides good forward and inverse solutions to transport problems. We explain this finding using the Neural Tangent Kernel (NTK) theory. We test both approaches on linear PDEs that represent groundwater flow and transport in porous media and rivers. We solve inverse problems to fit synthetic data, noisy synthetic data, and real-world measurements.

Privacy auditing provides an important safeguard by estimating the actual information leaked by a model, thus ensuring that theoretical privacy guarantees hold in practice. We study empirical privacy auditing for differentially private (DP) machine learning, focusing on efficient one-run methods for mechanisms such as DP-SGD. Prior one-run approaches threshold training examples or "canaries" into binary membership guesses, which discards useful information. We show that, in the white-box DP-SGD setting, canary-aligned signals naturally form a sequence of random variables whose normalized sum is asymptotically Gaussian. Leveraging this distributional perspective, we develop a DP-auditing framework that leads to tighter privacy lower bounds from a single training run.

As LLMs increasingly serve in advisory and deliberative roles, users rely on them for non-verifiable reasoning in domains lacking objective ground truths. However, traditional evaluations of LLM reasoning focus almost exclusively on fact-based domains, such as mathematics and science, leaving uncertainty over whether and to what degree models can handle ambiguous, subjective, or value-laden problems over time. To address this concern, we propose moral reasoning as a paradigmatic subdomain of non-verifiable reasoning. We define moral robustness as a model's capacity to exhibit sound moral reasoning across time and contexts, and we introduce a scalable, adversarial, multi-turn evaluation framework to empirically measure this capability. We simulate 48,000 user-agent moral deliberations across four frontier LLMs, varying premise relevance, premise order, conversation duration, and the user's stated moral view. We find that models successfully ignore morally-irrelevant distractors, but shift their reasoning by up to 6.5%, on average, towards the user's stated preferred moral view, and varying their reasoning depending on factors such as order (altering moral judgments by order in 13-22% of the cases) and duration (altering moral judgments between single-turn and multi-turn in 10-24% of the cases). Our analysis indicates that models tailor not just their final verdicts but their underlying justifications to align with a user's moral viewpoint - a failure mode we characterize as moral deliberative sycophancy.

Anticipating LLM behavioral tendencies from low-cost psychometric probes is critical for safe deployment, but only if self-reports (SR) reliably predict behavior. Recent work documented substantial SR-behavior dissociation in LLMs, but relied on broad personality traits (Big 5) that predict specific behaviors weakly, even in humans. Furthermore, the isolation of conversational sessions combined with weak context matching left open whether LLMs truly lack coherence or whether the conditions needed to detect such coherence were not met. We contrast Big 5 with the Theory of Planned Behavior (TPB), which measures intention targeted to a specific behavior and predicts human behavior substantially better than broad traits. We run experiments across four behavioral tasks and 11 frontier LLMs, while also varying session context and identity induction. We find that SR-behavior coherence exists but is selective. 1) Within a shared conversation, the Theory of Planned Behavior reaches human-level coherence; Big 5 does not. 2) Across separate conversations, coherence survives only for behaviors anchored outside the immediate prompt, such as implicit bias shaped by training, and collapses when behavior is strongly primed by context, as with sycophancy. 3) Persona prompting makes self-reports more consistent across conversations, but does not bring behavior into alignment. These findings suggest that coarse personality frameworks, such as Big 5 may not be the best tools for testing deployment behavior. More task- and behavior-specific instruments are needed, and even these must be evaluated across tasks and contexts.

Most learned dexterous grasp generators relegate contact forces to a downstream verification step, so a kinematically-plausible pose can still violate the conditions for a stable physical grasp. We address this with EquiDexFlow, an SE(3)-equivariant flow-matching model that jointly predicts wrist pose, joint angles, fingertip contacts, surface normals, and contact forces from an object point cloud. Our architecture projects contacts onto the object surface and forces into the Coulomb friction cone by construction, so placement and friction compliance hold without loss penalties. We prove end-to-end SE(3) equivariance and verify it empirically over 200 rotations, with wrist residuals below $0.04^\circ$ and exactly zero joint deviation. Trained on 8,100 force-closure grasps across 81 objects for the 16-DoF Allegro Hand, our model achieves zero friction violations, the best composite score, and the lowest wrench residual among all ablation variants. We retarget decoded fingertip contacts to a 16-DoF LEAP Hand via per-finger inverse kinematics, and our hardware-feasible refinement places every joint at least 5% inside its actuator envelope while preserving wrench balance. On the physical robot, retargeted EquiDexFlow-decoded grasps complete open-loop pick-and-hold trials on all six test objects, with every asymmetric object succeeding at both the canonical pose and a $120^\circ$ co-rotation. Videos, code, and checkpoints are available at https://equidexflow.github.io.

Inferring others' beliefs requires more than reading surface signals; it requires tracking who told them what, in what order, and how credibly. The Theory of Mind Utility (ToM-U) formalizes this epistemic state inference problem at the computational level of analysis, specifying what mentalizing computes and why without commitment to algorithmic or neural implementation. ToM-U achieves this by constructing Local Epistemic World Models (LEWMs) -- directed typed graphs that represent agents, state nodes, and the epistemic relationships among them -- and evaluating discrete candidate LEWMs against observed behavior until one achieves sufficient confidence. Five formal definitions specify the LEWM structure, agent node properties including ordered information access history, a bounded proliferation mechanism for recursive mentalizing, three inference procedures, and a residue function that captures the structured trace left by failed mentalizing attempts. ToM-U differs from Bayesian Theory of Mind and adjacent formal accounts, which presuppose rather than derive belief states, and from simulation theory and theory-theory, which lack a formal apparatus for epistemic state inference. The architecture generates directional, falsifiable predictions about mentalizing failure that follow from structural properties of the model rather than auxiliary assumptions, and positions ToM-U as a domain-agnostic mechanism upstream of goal inference and other downstream social cognitive processes.

Radio-frequency (RF) fingerprinting systems must operate in open-world environments where signals from unknown transmitters and temporal drift introduce distribution shift at test time. Out-of-distribution (OOD) detection provides a natural framework for this problem, yet its application to RF fingerprinting (RFF) remains limited. A key barrier to their adoption is that most OOD detectors require auxiliary OOD data for parameter tuning, an assumption that is difficult to satisfy in RF environments where representative OOD data is impractical to collect. In this work, we introduce a promising set of OOD detection methods from the machine learning literature to open-set RFF domain. We present these methods within a unified mathematical framework based on information theory, which is a natural framework for communication systems. Our framework allows for the systematic analysis of methods and development of new methods. We further demonstrate the applicability of recent work on tuning OOD detectors without given OOD tuning data for open-set RFF. We evaluate on the POWDER RF fingerprinting dataset, showing that detectors tuned without any given OOD data achieve performance comparable to baselines with access to true OOD tuning data and greatly out-perform baseline approaches without access to true OOD tuning data, showcasing the practical viability for the RFF problem.

The integration of Large Language Models (LLMs) and Multimodal LLMs (MLLMs) into scientific peer-review workflows introduces novel and significant risks for adversarial manipulation, especially given the multimodal nature of scientific papers where figures, not just text, convey core evidence. This creates a significant gap: current robustness studies on AI peer-review are overwhelmingly text-only. Moreover, the problem is distinct from standard jailbreaking, as a peer-review attack seeks to induce a domain-specific, targeted failure (e.g., "inflate this score") rather than a general safety policy violation, for which no practical defenses exist. To address this, we introduce PaperGuard, the first comprehensive benchmark designed to systematically evaluate and defend AI-generated peer-review against these domain-specific, cross-modal attacks. Our framework is built on three pillars: (1) a new multimodal peer-review dataset spanning multiple scientific domains; (2) a unified suite of attacks, including black-box prompt injections and white-box perturbations, specifically designed to target both text (GCG) and figures (PGD); and (3) a practical defense, motivated by the long-context challenge of academic papers, that uses chunk-based embedding search to efficiently localize and mitigate harmful instructions. Our extensive experiments, conducted across state-of-the-art models, confirm that AI reviewers are pervasively vulnerable. PaperGuard establishes the foundational benchmark, protocols, and actionable defense necessary to pioneer trustworthy, attack-resilient AI-assisted scholarly reviewing.

Claims that artificial general intelligence has already arrived and claims that it remains decades away are often defended from overlapping evidence. "AGI" lacks a single shared and stable referent and competing operationalizations can return different verdicts on the same system. This article treats that under-specification as a design and governance problem. Following Design Science Research Methodology, it develops DAF-AGI, a second-order conceptual artifact with two coupled components: five ordinal criteria for assessing the adjudicative fitness of candidate definitions and a structured governance audit of authorship, interest, certification, external verification and revision authority. The artifact is demonstrated on five prominent measurement families and one deflationary boundary position in a documented corpus and then stress-tested against a stylized strong arrival claim: that current generative systems constitute AGI because they outperform a well-educated adult on many cognitive tasks. On evidence from the cited 2024-2025 sources, the claim was certifiable only under a performance-based operationalization; capability-ontology, psychometric and skill-acquisition approaches did not certify it, the economic family remains indeterminate and the deflationary position refuses binary adjudication. The contribution is a novel integration and operationalization, not an empirical validation: independent application, inter-rater testing and author-external cases remain necessary. The paper further proposes definitional sovereignty as an enabling component of algorithmic sovereignty: the institutional capacity to contest, certify and revise imported technological categories under public accountability.

Diffusion models provide expressive data-driven priors for Bayesian inverse problems, but many diffusion posterior samplers rely on heuristic guidance approximations that can fail for nonlinear operators and multimodal posteriors. In this work, we develop a stabilized path-space framework for diffusion-based posterior sampling. Starting from a base diffusion process whose terminal marginal represents the prior, we define a likelihood-weighted target measure on trajectories and cast posterior sampling as learning a controlled stochastic process whose path measure matches this target. This formulation connects diffusion posterior sampling to stochastic optimal control while preserving the Bayesian structure needed for uncertainty quantification. We introduce a time reparameterization that makes the path-space control problem well posed by removing the bias induced by the unknown initial value function, without auxiliary training. We then learn the control via a trust-region path-space optimization method with log-variance objectives. The path-space perspective also unifies our learned control approach with existing guidance-based samplers, quantifies the sampling error induced by approximate controls, and yields importance sampling corrections for asymptotically exact posterior expectations. We evaluate the proposed framework on a suite of benchmark inverse problems with analytically characterized or high-quality reference posteriors, enabling principled assessment of sampling accuracy and uncertainty quantification. These experiments provide insight into the behavior of diffusion-based posterior samplers and demonstrate improved accuracy and robustness over leading approaches.

As LLM-based multi-agent systems (MAS) are deployed in the wild, the resilience of their collaboration structures against adversarial compromise becomes a critical safety concern. Attackers may leverage prompt-injection or jailbreaking to sabotage individual agents within MAS workflows, but the interaction between model scaling and system-level resilience remains poorly understood. This paper investigates how model scale affects the security of linear multi-agent workflows. Our experiments across scales of two open-weight model families on the HumanEval benchmark reveal a compliance-correction symmetry: larger models are far more likely to faithfully execute malicious instructions, with the control-to-malicious performance drop reaching 53.7pp at 27B in uncorrected pipelines. However, appending a lightweight terminal Fixer stage collapses this to 0.6pp and restores statistical parity with control-level performance, demonstrating that strictly linear collaboration structures can be viable and resilient to adversaries at this scale, and suggesting that the brittleness previously attributed to linear topology may stem from a lack of correction.

Despite their linguistic diversity and global significance, African languages remain underrepresented in research and resources to support NLP. We aim to bridge this gap by introducing AfriSUD, the first large-scale collection of syntactically annotated treebanks for nine diverse African languages spanning major language families and regions across Sub-Saharan Africa. Using the Surface-Syntactic Universal Dependencies (SUD) framework, our community-led effort provides high-quality, native-speaker verified data that capture typological key features such as agglutination and tone. We evaluate a range of models on AfriSUD for part-of-speech tagging and dependency parsing including non-transformer baselines, multilingual pretrained encoders, and LLMs. Our results reveal a significant syntax gap, where models still show clear limitations across the nine languages, suggesting that existing architectures may not fully capture the structural diversity of African-language syntax.

Retrieval-augmented generation (RAG) agents increasingly run with persistent memory that accumulates across user sessions. This creates a new attack surface: an adversary interacting only through normal channels can inject crafted memories that, once retrieved, steer the agent's responses for future users, without touching model weights or code. We call this Multi-Session Memory Poisoning (MSMP) and show that no existing defence certifies against it; static-corpus defences (RobustRAG, ReliabilityRAG) assume a fixed knowledge base, and heuristic filters are bypassed by fluent enterprise-style text. We present Signed Memory with Smoothed Retrieval (SMSR), the first defence with a certified robustness bound for this setting. Component 1 adds HMAC-SHA256 provenance at write time, blocking unsigned injection. Component 2 applies randomised memory ablation with verdict-based majority voting at query time, bounding the influence of authenticated adversaries. We prove that no provenance-free retrieval-time filter can certify against adaptive injection, derive a hypergeometric certificate for Component 2, and formalise the Consistent Minority Effect, whereby a consistent adversarial answer wins string-based voting as a numerical minority while verdict-based voting removes it. Across 15 enterprise scenarios (3,150 repeated trials), Component 1 cuts attack success from 93-100% to 0% for all unsigned variants. For an authenticated adversary with a single injection, Component 2 holds success to 8.0% (95% CI [5.8, 10.9], n=450), below the certified worst case. In an end-to-end query-only attack where the agent itself writes the poison rather than it being pre-seeded, SMSR reduces success from 65.3% to 5.3% (n=150, non-overlapping CIs) on a live agent stack. Clean-query utility is 90% (Component 1) and 85% (combined).

Large language models (LLMs) are increasingly integrated into clinical systems, making it essential to evaluate the real-world utility of these systems. However, static benchmarks tend to measure correctness rather than user acceptance, aggregate performance across queries, and require densely annotated datasets -- leading to major blind spots for evaluating clinical systems. In this work, we perform a deployment-centered evaluation of an LLM system embedded within electronic health records at an academic medical center, where user feedback is sparse but closely reflects the deployment conditions. Specifically, we train a pre-response classifier that estimates the risk that a future interaction will result in the user rejecting the LLM response, based on query content and deployment-specific context available before generation. We conduct a prospective analysis of our model over 4.5 months of user feedback, finding that our prediction model achieves an AUROC of 0.719. Further, we estimate the benefit of such predictions in two downstream use cases (guardrail triggering and abstention). Our key conceptual insight is that making use of deployment-specific context (i.e., the provider type, department name, language model used for response), as opposed to only query content, improves the ability to predict whether the user will reject the system output. Altogether, our empirical case study demonstrates the feasibility of predicting user rejection using deployment-specific context, opening the door to targeted guardrails.

Type 2 Diabetes (T2D) poses an increasing global health threat, demanding effective glycemic assessment to support personalized and improved diabetes care. Wearable sensors such as continuous glucose monitors (CGM) and fitness trackers offer many valuable insights for glycemic assessment. However, effectively analyzing these data requires integration with essential individual-level context. Existing methods are often based on traditional machine learning (ML) and rely primarily on historical blood glucose measurements and overlook personalized information, which limits their performance across diverse diabetes populations. Recent advances in large language models (LLMs) have demonstrated their ability to integrate diverse data modalities while modeling sequential dependencies, motivating the exploration of their potential for personalized glycemic assessment. In this paper, we propose GlyLLM, an LLM-powered framework for modeling CGM-based glycemic dynamics through the integration of wearable sensor data and structured metadata. GlyLLM can leverage the extensive prior knowledge of pre-trained LLMs and achieve sensor-text semantic abstraction at decision time. Experiments on two related tasks on the AI-READI dataset demonstrate that our model outperforms traditional ML methods by an average of 13.66\% in Root Mean Squared Error (RMSE) for glucose forecasting and 13.08\% in Area Under the Receiver Operating Characteristic (AUROC) for diabetes categorization. Additionally, our ablation study shows that diabetes surveys and biometric tests are more critical than other health information for glycemic assessment. Our work presents a promising step toward harnessing the power of LLMs to advance personalized glycemic assessment in T2D care.

We give a simple, unified, and nearly tight bound for sampling arbitrary logconcave distributions from a warm start using the In-and-Out algorithm along with exponential lifting. The main new ingredient in the analysis is an improved bound on the Poincaré constant of a lifted distribution. As a consequence, the resulting convergence rate is nearly tight for both constrained settings (e.g., Gaussian restricted to a convex body) and well-conditioned settings (e.g., strongly logconcave and smooth densities).

Auto-regressive models have emerged as powerful tools for sequential data, from language to video. Understanding how and why these models learn latent representations remains an open theoretical question. In this work, we demonstrate that when trained by empirical risk minimization on data from partially observed linear dynamical systems, two-layer linear auto-regressive models naturally learn to approximate Kalman filtering. In particular, we show that the learned hidden representation coincides, up to a similarity transformation, with the state estimates produced by the optimal (Kalman) filter, even though the model has no explicit knowledge of the underlying dynamics or state. The result follows from three main insights. First, we establish that the Kalman filter is well approximated by an auto-regressive model with bounded truncation error. Second, we show that despite non-convexity, the two-layer optimization landscape is benign, i.e., all stationary points are either strict saddles or global minima. Finally, as our main contributions, we provide finite-sample guarantees on prediction error, parameter estimation error, and latent state recovery. Numerical simulations support the theoretical results and demonstrate that the latent representations of auto-regressive models recover state estimates.

In this paper, we propose the Enhanced World Action Model (EWAM), a closed-loop online adaptation architecture built upon a pretrained and fully frozen Cosmos3 backbone network. Evaluated entirely under a zero-shot task protocol, EWAM is centrally focused on reducing the amount of additional deployment data required to adapt to new task layouts. Notably, no extra task-specific demonstration sets were introduced in any of the evaluations, and no fine-tuning was performed on the backbone network. Its performance gains stem entirely from an inference-time co-reasoning mechanism composed of four inserted lightweight neural layers: the Neural Experience Memory Layer located in the intermediate layers of the Diffusion Transformer (DiT) provides task-relevant execution context; the Neural Anomaly Detection Layer after the state prediction head monitors the divergence between predicted and actual states in real time; the Neural Policy Routing Layer dynamically selects direct execution, conservative replanning, or rollback recovery based on the anomaly severity; and the Neural Action Correction Layer refines the generated action chunks using execution diagnostics. Unlike naive feature fusion, the memory, anomaly detection, and correction modules are deeply integrated into the Cosmos3 forward path in a differentiable manner, with only the final routing decision being a discrete supervised one.

Latent reasoning models (LRMs) replace explicit chain-of-thought with continuous thoughts. Recent work treats observable latent-state patterns, such as BFS-like frontiers and decodable arithmetic computation, as evidence for internal reasoning mechanisms. Evaluating two LRMs (Coconut and CODI) against controls lacking the proposed recurrence or curriculum, we find these patterns also appear in the controls and do not always causally affect behavior. Causal interventions reveal that latent-thought utilization is not binary but graded, scaling with a thought's causal effect on model behavior. Geometric analyses reveal this effect concentrates in low-rank directions whose step-to-step geometry grows more structured as their behavioral influence increases. Latent thoughts should therefore be treated as hidden computation, not hidden explanation: decodability, attention, or static structure alone cannot establish mechanism. LRM interpretability thus requires matched controls and causal tests.

We are entering a new era of composite model architectures that integrate diverse components such as vision encoders, language backbones, diffusion and flow heads, audio codecs, action generators, and world-model predictors. Such architectures underpin a broad class of multimodal models, including unified multimodal models, omni models, speech-language models, vision-language-action policies, and world models. However, existing model serving frameworks were built on narrow assumptions about model structure, making them ill-suited to accommodate this new architectural diversity. Here we present M*, a universal serving system for efficient serving of composite AI models. M* represents models as dataflow graphs, processing requests spanning diverse modalities and tasks as traversals over these graphs. The core insight is a modular abstraction that supports arbitrary composition of model components, flexible placement onto a physical cluster, and model-agnostic optimizations within a distributed runtime. We call this abstraction the Walk Graph and show how it can concisely capture composite models from a broad range of families. We instantiate M* on representative models and find that it achieves, on average, 20% lower end-to-end latency than vLLM-Omni for text-to-image workloads on BAGEL, while delivering up to 2.9x lower real-time factor and 2.7x higher throughput for text-to-speech workloads on Qwen3-Omni. M* also outperforms the V-JEPA 2-AC rollout baseline for robotic planning by up to 12.5x. Thus, our work paves the road towards more efficient serving of complex models with minimal developer effort.

Marketing mix models are used to forecast business outcomes and to attribute those outcomes to marketing channels, but these goals are not equivalent. We study a failure mode in graph-based neural MMM called attribution bypass: a high-capacity decoder can obtain low forecasting error through target autoregression, dense communication, co-movement, context, or latent memory while failing to route counterfactual sensitivity through the graph used as the attribution object. We introduce DICE-MMM as a bounded diagnostic and training framework. We do not claim that observational neural MMM identifies causal effects. Instead, DICE separates three questions often conflated in graph-based MMM: graph recovery, forecasting accuracy, and whether the trained decoder's perturbation-induced influence is graph aligned. Stage 1 trains a graph encoder with a restricted graph-mediated decoder. Stage 2 freezes the selected encoder and trains a graph-safe latent decoder whose cross-node communication must pass through the supplied graph. Decoder use is evaluated with CIG, AR-CIG, and graph-swap tests. Across controlled R/d/T swaps and an external multi-graph rawlog stress test, DICE improves stable graph recovery over CausalMMM. The experiments show that forecasting accuracy is not an attribution certificate: in a sparse-target benchmark, no-graph and full-graph decoders achieve MSE@7 around 0.004 while AR-CIG nAUPRC remains near or below zero, whereas an oracle graph reaches 0.807 +/- 0.129 at comparable MSE. Frozen graph-swap localizes the bottleneck: the same DICE-hard-trained decoder moves from nAUPRC -0.044 +/- 0.006 under learned graph inputs to 0.894 +/- 0.027 with the oracle graph. The contribution is a stress test and failure-localization framework showing that low MSE can hide attribution bypass and that the unresolved bottleneck is graph-support selection, not forecasting or decoder capacity.

Over the last decade, building human-level artificial general intelligence has moved from far-fetched speculation to being a concrete next-decade target for many of the largest AI organisations. Achieving this goal would have profound and far-reaching impacts on human society, which raises many complex questions for the decade ahead. This report investigates how AI itself might continue to develop in a post-AGI world along the continuum of machine intelligence. The endpoint of this continuum, Universal AI, is theoretically well understood, which provides some formal grounding for the main focus of this report: the transition from human-level AGI to artificial general superintelligence, which, intuitively, can be understood as a system that is more intelligent and cognitively capable than large organisations of humans. After characterizing ASI, the report discusses four potential pathways from AGI to ASI: scaling AGI, AI paradigm shifts, recursive improvement, and ASI emerging from large-scale multi-agent collectives. The report then discusses possible frictions and bottlenecks along these pathways. Determining whether the impact of these frictions will be negligible or substantial raises a number of concrete open research questions. Due to large uncertainties for predicting ASI progress, it cannot be ruled out that AI progress might continue to accelerate over the next years. This could imply that the image of a single transformative step change, caused by the introduction of human-level AGI into our society, could be inaccurate. More apt might be the prospect of a series of transformative societal changes caused by AI-enabled progress and breakthroughs across many areas of science and technology. Preparing for this prospect requires a massively interdisciplinary endeavour of global scope and interest.

Machine learning models often degrade when they are deployed on a target distribution that differs from the source distributions they were trained on. Recent work in causality-based domain generalization has shown how shared causal structure between domains can induce invariant predictors, e.g., models on a subset of features which have stable risk across structured domain shifts. However, the extent to which such population-level causal invariances can lead to gains in finite-sample settings remains underexplored. In particular, in practice we often have access to a few labeled target samples, a setting called supervised domain adaptation (sDA). In this paper, we explore when (full or partial) causal knowledge can provably improve supervised domain adaptation. As a first step, we study linear regression, where full or partial causal knowledge specifies a collection of invariant or possibly invariant feature subsets, each yielding a source-trained candidate predictor. We derive matching upper and lower bounds showing that finite-sample gains are governed by the target-risk margins separating the candidates, together with the finite-source estimation error. When these margins are sufficiently large relative to $n_Q$, an adaptive aggregation procedure can match the best candidate predictor while avoiding negative transfer relative to target-only learning. On the other hand, when the margins are too small, no algorithm can reliably exploit the candidate collection to obtain faster finite-sample rates. We further connect these margins to structural shift magnitude in linear SCMs and validate the theory on real-world causal benchmarks.

Federated learning (FL) enables collaborative model training without sharing raw patient data, but standard approaches such as FedAvg treat each client as a black box and provide no mechanism for isolating an adversarial contributor, auditing per-client influence, or honoring a departed participant's right to be forgotten. We present Fed-FBD (Federated Functional Block Diversification), a modular federated architecture that decomposes a ResNet backbone into six functional blocks (the stem, four residual groups, and the classification head) and maintains a warehouse of N color variants, each assembled from independently tracked and contributor-stamped blocks. Fed-FBD provides three capabilities absent in FedAvg: (i) architecturally guaranteed block-level isolation, so that an adversarial or mislabelled client cannot contaminate the clean colous; (ii) privacy-by-design, where membership inference advantage is already indistinguishable from chance before any privacy mechanism is applied; and (iii) surgical machine unlearning of a departed participant's contribution at sub-second cost and without retraining. Experiments on six MedMNIST-2D datasets, PathMNIST at 224x224, and CIFAR-10 show that Fed-FBD trades a modest 0.3%-3.1% IID accuracy gap on the adequately sized datasets for these guarantees, remains within 0.8%-4.0% of FedAvg at Dirichlet alpha=1.0 on three of four datasets, and confines all six adversarial attacks we study to the poisoned client's own blocks with at most +/-0.01 AUC drift on the clean colors.

We study the communication complexity of convex decentralized optimization over time-varying networks, where $n$ nodes hold private functions and must agree on the global minimizer using only synchronous exchanges with neighbors. The cost is the number of communication rounds to reach accuracy $\varepsilon$ -- a measure akin to round complexity in the LOCAL model, but constrained by nodes sharing only oracle responses. We prove a new lower bound of $Ω\!\left(χ_{\mathcal G} \sqrt{κ_g}\,\log\frac{n}{χ_{\mathcal G}}\log\frac1\varepsilon\right)$ communication rounds, where $χ_{\mathcal G}$ is the condition number of the network Laplacians and $κ_g$ that of the global objective, showing the round complexity attainable under uniform regularity cannot be matched in the nonuniform regime. The construction rests on spectral graph theory: we embed time-rotating star gadgets into the edges of an expander and patch them to preserve spectral connectivity.

Compact language models (LMs) reduce cost, latency, and deployment risk for tool agents. Yet MCP-style tool use requires more than isolated function calling: an agent must discover tools from live catalogs, satisfy schemas, preserve dependencies across intermediate outputs, and ground final responses in executed evidence. Small planners often generate plausible workflow graphs that fail under tool resolution, parameter validation, dependency tracking, or execution. We argue that this failure mode is poorly handled by small-corpus distillation. A few hundred teacher traces can teach workflow format, but rarely cover the recovery behavior needed to repair failed plans over changing tool catalogs. We introduce Evoflux, an inference-time evolutionary search method that treats compact tool use as the repair of executable tool workflows. It evolves typed workflow graphs through structured edits, execution feedback, adaptive intensity, meta-guided redesign, and diversity pruning. On held-out MCP-Bench tasks spanning live MCP servers and 250 tools, Evoflux raises execution feasibility from roughly 3% to 17-24% across small planners. In contrast, SFT and SFT+DPO on the same search-mined data match, underperform, or collapse below zero-shot performance; ReAct reaches higher peaks, but with higher variance and token cost. These results show that execution-grounded search is more reliable under scarce teacher-trace budgets.

Cross-domain graph anomaly detection (GAD) aims to identify abnormal nodes in unseen target graphs, showing strong potential in real-world applications with heterogeneous graph data. However, existing methods often depend on dataset-specific feature semantics and structural patterns, which limits their ability to generalize across different domains. To address this challenge, we propose AlignGAD, a zero-shot generalized graph anomaly detection framework. Our framework is built upon three key components: a Global Unification Module that aligns heterogeneous node features and normalizes graph signals in the spectral domain; a Clustering Module that constructs cluster-aware graph views to capture group-level abnormal patterns; and a Node Discrepancy Scoring Module that measures reconstruction discrepancy and aggregates anomaly evidence from different graph views. Experiments on multiple real-world datasets demonstrate the effectiveness of AlignGAD under the zero-shot GAD setting.

Rapidly expanding low Earth orbit satellite constellations are placing increasing demands on terrestrial ground networks, motivating the development of more efficient ground station network designs. Current approaches select sites from predefined locations, limiting optimization to existing infrastructure and constraining performance. In contrast, free-placement optimization operates over a continuous spatial domain on Earth, broadening the search space and allowing higher-throughput configurations at the cost of potentially requiring new infrastructure deployment. In this work, we introduce SCORE (Sequential Cyclic Optimization via Refinement & Evaluation), a two-stage free-placement method for ground station design. SCORE combines sequential coordinate selection with cyclic refinement to manage high-dimensionality, non-convexity, and local minima that challenge global optimizers. We benchmark SCORE against one-shot methods such as differential evolution (DE) and integer programming approaches using locations from Kongsberg Satellite Services and the World Teleport Association. Tests across two commercial Earth observation constellations (Capella Space and ICEYE) and one synthetic Walker-Star constellation show that SCORE requires up to 5x fewer function evaluations to converge relative to DE while improving downlink throughput by up to 13%. Compared to fixed-site methods, unconstrained SCORE achieves up to 15% greater total downlink, establishing a strong empirical performance benchmark for flexible placement; infrastructure-constrained SCORE retains over 92% of this gain while restricting placement to within proximity of existing fiber and power infrastructure. We also explore trade-offs between expanding existing stations and deploying new sites, informing future ground network design for operational constellations.

Screenshot-based mobile GUI agents can operate ordinary smartphone apps through the same visual interface as a human user, but this capability also turns every screen observation into a privacy boundary. During normal task execution, screenshots may expose contacts, messages, photos, files, recommendations, health cues, and other sensitive context that is unrelated to the user's request. We call this problem incidental visual privacy exposure. It is difficult to address with existing defenses: text anonymization misses many visual and inferential cues, while generic privacy masking can remove the evidence and controls that a GUI agent needs to complete the task. This paper presents CAPED, a context-aware pre-upload exposure control layer for mobile GUI agents. CAPED is designed as a phone-side protection layer: before screenshots are released to a remote multimodal agent, it extracts task requirements, uses screen context as a privacy prior, parses visible UI elements, and selectively exposes only content needed for the current task while masking incidental private content. We evaluate CAPED on AndroidWorld for broad task utility and with a controlled 28-task seeded privacy evaluation used as a measurement instrument for trajectory-level incidental leakage. In this seeded evaluation, Full CAPED reduces success-conditioned weighted seeded leakage from 0.766 under raw screenshots to 0.268 while preserving high task utility. A broader AndroidWorld run shows a remaining prototype-level utility cost, but the results support the central claim that screenshot upload should be treated as an explicit device--cloud boundary decision, governed by task-driven selective exposure rather than all-or-nothing screen sharing.

Speech enhancement models typically apply uniform capacity across all frequencies, disregarding the non-uniform spectral resolution of human hearing. We propose BASENet, a frequency-adapted architecture that partitions the spectrum into Bark-scale bands and assigns each a scaled-capacity encoder derived from critical-band density, automatically granting deeper branches to perceptually dense low frequencies and lighter ones to high frequencies. A cross-band attention module captures harmonic dependencies across bands through compact frequency-pooled representations at linear complexity. Built on inverted residual blocks with dense connectivity and a convolutional recurrent network, BASENet achieves 3.55 PESQ and STOI~96% on VoiceBank+DEMAND with only 0.83M parameters and 7.3 G~MACs, the fewest parameters among all methods with PESQ > 3.50. A causal variant (3.44 PESQ) surpasses several non-causal baselines, confirming suitability for real-time streaming on resource-constrained devices.

Physics-Informed Neural Networks (PINNs) are an attractive tool for partial-observation problems in biology, where the governing dynamics are known but some compartments cannot be measured. Chemotherapy pharmacokinetics (PK) is a clean instance: drug concentration in plasma is routinely measured, but concentration in tissue -- which determines tumour kill and off-target toxicity -- is not. We benchmark a PINN against the standard clinical baseline (nonlinear least-squares on the analytical biexponential plasma solution, hereafter NLS) and a physics-agnostic neural baseline (a data-only MLP) on two PK problems. On the linear two-compartment problem, NLS is near-optimal; the PINN matches it to within a small constant factor while also producing the tissue curve in a single training pass, whereas the data-only MLP fails on tissue by roughly 10x. On a Michaelis-Menten extension (saturable elimination), the biexponential closed form no longer exists, so NLS is mis-specified and silently returns meaningless rate constants. The PINN instead exposes a deeper fact: the Michaelis-Menten two-compartment model is non-identifiable from plasma alone, and the PINN reports this honestly by converging to a basin with k12 -> 0. Adding two sparse tissue observations largely resolves identifiability: across five seeds the PINN recovers k21 to within 1% of truth and Vmax, Km to within one standard-deviation bar, while k12 moves in the correct direction (0.02 -> 0.82) but remains ~2 sigma below truth -- a recovery the closed-form NLS estimator cannot attempt at all, because its biexponential ansatz describes only plasma. Our claim is not that PINNs beat NLS. It is that PINNs offer a uniform recipe that ties the textbook estimator on the textbook problem, exposes structural identifiability that the textbook estimator hides, and absorbs heterogeneous measurements within a single loss.

Human mobility data is important for transportation, urban planning, and epidemic control, but large-scale trajectory collection is often costly and privacy-constrained, motivating realistic synthetic trajectory generation. Existing LLM-based generators typically rely on either prompt engineering, which preserves zero-shot reasoning but lacks fine-grained spatiotemporal grounding, or trajectory-level fine-tuning, which improves statistical precision but incurs substantial computational cost and may weaken general reasoning. We propose TrajGenAgent, a semantic-aware hierarchical LLM-agent framework for human mobility trajectory generation without model fine-tuning. TrajGenAgent uses a two-stage orchestrator-worker design: an LLM first synthesizes an individual- and weekday-conditioned activity chain from historical evidence via in-context learning, and a deterministic workflow then grounds each activity into a complete visit using personalized POI retrieval, distance-aware location selection, kinematics-aware travel-time propagation, and LLM-based duration estimation. To evaluate realism beyond aggregate spatiotemporal statistics, we introduce an anomaly-detection-based evaluation framework using two complementary detectors to assess behavioral and semantic plausibility. Experiments on benchmark and large-scale simulation datasets show that TrajGenAgent improves spatiotemporal fidelity, semantic coherence, and individual-specific behavioral realism over representative neural and LLM-based baselines, while avoiding parameter updates.

We develop a new, differentially private mean estimator called the balloon mean. The main features of the balloon mean are that it is computationally tractable and enjoys robustness to outlying observations. It is based on an iterative clipping procedure over expanding Mahalanobis balls, or ``balloons.'' The method satisfies zero-concentrated differential privacy and depends on a small number of interpretable tuning parameters. We provide theoretical guarantees under heavy-tailed and contaminated elliptical models, characterizing its statistical performance and robustness to outliers. Extensive simulations demonstrate that the balloon mean is robust to heavy-tailed and contaminated data, and outperforms existing differentially private mean estimators in contaminated settings.

Machine-learning drug-discovery pipelines increasingly rely on generative models that propose molecules far from the data used to train downstream synthesizability filters. Existing filters (SAScore, SCScore, RAscore, DeepSA) are purely statistical and degrade in exactly this out-of-distribution (OOD) regime. We ask whether cheap, closed-form physical priors, used as auxiliary supervision on a graph neural network (GNN), improve OOD generalization. We add two auxiliary losses to a GINE backbone: a topological complexity regression supervised by the Bertz index, and a strain-energy soft penalty supervised by MMFF94 force-field energy. On a 65,177-molecule corpus (HIV, Tox21, COCONUT) labeled by SAScore thresholds we reproduce a strong in-distribution baseline, then evaluate a 4-way ablation (baseline / +complexity / +strain / +both) on a single-source OOD split (train on drug-like HIV+Tox21, test on COCONUT natural products), repeated over 5 seeds with paired bootstrap confidence intervals. All three physics-aware variants give a small but statistically significant OOD improvement over the baseline (mean OOD AUC 0.9774): +complexity Delta = +0.0060 (95% CI [+0.0023, +0.0102]), +strain Delta = +0.0032 ([+0.0008, +0.0052]), +both Delta = +0.0066 ([+0.0038, +0.0093]); every interval excludes zero, and the combination is best. The variants are indistinguishable in-distribution, so the effect is visible only under OOD evaluation. We are explicit that the effects are modest, and we report a cautionary methodological finding: a single-seed version of this experiment produced a qualitatively different (non-monotone) story that did not survive multi-seed evaluation.

Detecting mental health disorders from Arabic social media text remains challenging due to dialectal variation, informal language, limited high-quality annotated resources, and severe class imbalance. While English mental health natural language processing (NLP) has progressed substantially, Arabic multi-class disorder classification remains insufficiently studied. This study proposes a two-phase framework for Arabic mental health text classification. In phase 1, three Arabic pre-trained language models, AraBERT, CAMeLBERT, and MARBERT, undergo Domain-Adaptive and Task-Adaptive Pretraining (DAPT and TAPT) using a large-scale corpus of unlabeled Arabic mental health tweets. The adapted models are evaluated under a unified protocol to identify the most effective backbone model. In phase 2, the selected model is assessed across four configurations combining single-stage and hierarchical two-stage classification architectures with full fine-tuning and Low-Rank Adaptation (LoRA). To support this study, we constructed a novel annotated Arabic mental health dataset comprising 50,670 tweets across six categories, with strong inter annotator agreement (Krippendorff's Alpha = 0.733, average pairwise agreement = 0.797). Experimental results show that the domain-adapted MARBERT (MentalMARBERT) achieves statistically significant improvements over baseline models in both accuracy and macro-F1. The hierarchical two-stage architecture combined with full fine-tuning achieves the best overall performance, reaching a macro-F1 of 0.861 and an accuracy of 0.877. These findings demonstrate the effectiveness of domain-specific adaptive pretraining and hierarchical classification for Arabic mental health disorder detection.

AI-augmented computing delegates natural language queries, code generation requests, and other open-ended tasks to a cluster of AI models that processes queries and generates responses. This paradigm introduces a resource dimension that neither classical time nor space complexity captures: the cost of sending queries to and receiving responses from such a cluster. We introduce token complexity, a formal resource measure defined as the minimum expected token cost to achieve a specified level of output quality on a task, and develop a taxonomy classifying AI systems by the strength of their probabilistic properties. We develop token complexity within the framework of AI-Oracle Turing machines, in which a probabilistic Turing machine interacts with a stochastic oracle via dedicated query and response tapes. We prove basic theorems establishing that token complexity behaves as expected: monotonicity (higher quality costs more tokens), convexity (quality improvements become progressively more expensive), price sensitivity (small price changes produce bounded cost changes), and price-relativity of task ordering (the token complexity ordering of tasks can reverse depending on the query-to-response cost ratio). We prove that the complexity frontier, defined as the set of all feasible resource bounds in tokens, time, and space, is non-empty, upward-closed, and convex.

The standard taxonomy of predictive uncertainty defines epistemic uncertainty as the part removable by collecting more data, while the standard measure identifies it with a mutual-information term. We prove the definition and the measure are extensionally inconsistent. On an explicit construction, the measure assigns all uncertainty to the epistemic class, yet no quantity of training data reduces it. Reducibility is instead a property of the pair (uncertainty, acquisition class), and the dichotomy resolves into three parts: aleatoric, sample-reducible epistemic, and mechanism-reducible epistemic uncertainty. An exact identity for the value of an observation shows that in-distribution data never reduces mechanism-irreducible uncertainty and generically increases it. Ensemble disagreement, the deployed epistemic estimate, tracks the training procedure rather than the epistemic term. It collapses to zero beneath a positive truth under consistent training, and equals hyperparameter-scaled initialization noise under interpolation. A finite-sample falsification test and seed-swept experiments confirm the theory.

When working with real-world temporal data, it is common to encounter features whose distribution is changing over time. The naive employment of Machine Learning models on this unstable data might lead to rapidly degrading performance, especially if the new distribution is much different from what was previously seen during training. In order to cope with this problem, it is critical to automatically identify features that are changing over time. With these features detected, data scientists and other practitioners will be able to mitigate the issue (for instance, by applying data transformations), deploying more robust models that retain high performance for longer periods of time. In this paper, we describe which temporal changes a feature should not suffer from, and propose TEDD, a technique to a) identify when a dataset might lead to an unstable Machine Learning model and b) automatically detect which features cause such lack of robustness. In order to achieve it, we leverage a regression model to highlight which features contribute to a good prediction of an instance's timestamp. We compare our approach to other methods in real and synthetic data, testing their detection capability on all simple change patterns. We show that our method: detects all types of basic changes, both for numerical and categorical features; can detect multivariate drifts; returns a comparable value measuring the amount of change of each feature; requires no parameter tuning; and is scalable both on number of features and instances of the dataset.

Offline reinforcement learning allows control policies to be learned directly from data without online interaction, making it suitable for safety-critical tasks. Recent studies have applied diffusion models to offline reinforcement learning to leverage their strong capacity for modeling complex data distributions. However, existing approaches primarily focus on single-agent settings, leaving the safety challenges in multi-agent environments largely unexplored. In this work, we propose a safe offline multi-agent reinforcement learning algorithm that embeds neural individual control barrier functions into the diffusion model to enhance safety during trajectory generation, with control policies recovered through inverse dynamics. We evaluate our algorithm across diverse benchmarks, demonstrating substantial safety improvements while maintaining competitive rewards.

Predicting how a cell's transcriptome responds to a drug it has never seen is a core, hard problem in computational cell biology: recent benchmarks show complex models often fail to beat trivial baselines once test compounds are held out by chemistry. We study one cell line and assay, THP-1 cells profiled by DRUG-seq, scored by the active-compound weighted MSE(wMSE) of the VCPI prediction contest. We propose a staged approach: dumb baselines (untreated control and mean training-compound response) that the field keeps failing to beat; non-parametric retrieval (a Tanimoto-weighted average of a held-out compound's nearest training compounds); and a fusion stage combining a frozen chemistry embedding with retrieval-support features to predict the residual over the mean, with an uncertainty head and gene programs. On the released VCPI THP-1 drug-seq data (14,026 training compounds), under a Bemis-Murcko scaffold split, the model ranking inverts depending on the metric. Under an inverse-variance per-gene proxy, a regularized linear regression on Morgan fingerprints appears to win over the deep models, retrieval, and ChemBERTa -- the textbook "simple baselines win" result. But under the contest's true active-set metric (per-(gene, compound) Mejia weights, validated against the official scorer; mean baseline 0.535 vs the organizers' 0.507 reference), that reverses: the deep models win, our fusion decoder significantly beats the linear fingerprint baseline (-0.012 wMSE, paired bootstrap p < 10^-4), and the proxy's winner becomes the worst chemistry-aware predictor. Picking the metric picks the winner -- to our knowledge the first demonstration on real held-out drug chemistry of the metric-calibration effect established largely on genetic perturbation. We release a reproducible pipeline wired to the official scorer that emits a valid submission over the real 1064 x 12,995 grid.

As distributed AI workloads grow in scale, multi-GPU systems have become essential for training large models. Although techniques like kernel fusion and overlapping communication with computation help reduce delays, they also introduce irregular and transient traffic patterns that are difficult to model using existing tools. These techniques rely heavily on fine-grained synchronization and peer-to-peer communication, which place significant pressure on interconnect bandwidth and latency. In this work, we introduce Eidola, a scalable extension to the gem5 simulation framework that enables detailed modeling of inter-GPU communication traffic. The extension is scalable as our GPU model serves as a succinct eidolon, emulating the minimal characteristics needed for traffic modeling. Eidola uses annotated timing profiles from real applications to emulate peer-to-peer GPU writes with cycle-level precision. This allows researchers to simulate and analyze synchronization behavior across large multi-GPU configurations. The simulator supports configurable per-GPU traffic patterns and enables isolated performance analysis under different communication scenarios. We demonstrate Eidola's effectiveness by reproducing variability in fused kernel execution and by implementing a SyncMon-inspired synchronization mechanism, confirming reductions in polling-related memory traffic. Our results show that Eidola provides a flexible and scalable platform for studying inter-GPU communication and supports architectural exploration in modern distributed GPU systems.

Long-horizon tool-use reinforcement learning can learn from outcome verification, but its trajectory-level advantage is broadcast across many reasoning, API, and answer tokens. Self-distillation promises a denser signal by reusing a policy's own rollouts or a privileged teacher. We show, however, that direct token-level self-distillation can silently destroy tool use: it rehearses teacher behavior without knowing which actions the verifier rewards, so useful skills and harmful shortcuts are amplified together. We introduce Sibling-Guided Credit Distillation (SGCD), which uses distillation for credit assignment rather than as a competing actor loss. Dynamic sampling produces mixed successful and failed sibling rollouts; an external LLM summarizes their contrast into a training-only stepwise credit reference; dense teacher/student divergence drives credit reassignment; and bounded detached credit weights reshape GRPO token advantages. The deployed student sees no external LLM, sibling evidence, or oracle. Across AppWorld and $τ^3$-airline, SGCD improves over matched GRPO comparators: AppWorld TGC $42.9 \to 45.6$ on test_normal and $24.7 \to 27.0$ on test_challenge, and $τ^3$-airline pass@1 $0.583 \to 0.602$.

Incremental Learning (IL) for Open-ended Image-to-Text Generation (OpenITG) enables models to continuously generate accurate, contextually relevant text for new images while preserving previously acquired knowledge. Unlike prior studies, this paper addresses a more practical scenario in which the predominant category of visual data shifts over time as environments evolve. In this context, we introduce a new notion of continual alignment, which incrementally adapts the alignment module within pre-trained VLMs to preserve high-quality cross-modal representations. Based on this idea, we propose Efficient Continual Alignment (ECA), a novel exemplar-free IL approach for OpenITG. The key challenge is enabling the model to acquire new, task-specific features while minimizing interference with the established alignment without accessing raw data from previous tasks. To address this, ECA employs three core mechanisms: a Mixture of Query (MoQ) module that adapts task-specific query tokens, a Fisher Dynamic Expansion (FeDEx) that dynamically expands model structure based on a Fisher Information Matrix (FIM)-based metric, and an embedding dictionary with Dictionary Replay (DR) to retain past knowledge. To evaluate ECA's performance, we construct four new IL OpenITG benchmarks that better reflect real-world scenarios. Experimental results demonstrate that ECA significantly mitigates catastrophic forgetting and improves IL performance compared to baseline methods. Code and benchmarks are available at https://github.com/Snowball0823/ECA.

We show that the standard basis of transformer hidden states already provides a training-free, architecture-general feature basis. Individual dimensions encode semantic content via their signs and confidence via their magnitudes, functioning as independent binary registers. We validate this Bag of Dims framework across three model families (Qwen 3.5-4B, Gemma 3-4B, Mistral 7B) through four progressive experiments. Sign patterns alone carry predictive content: replacing all magnitudes with unity achieves 72-93% top-5 next-token accuracy through the LM head, and pure Hamming scoring without any decoder reaches 80-90% top-4096. These sign patterns organize into semantic features: using a single-token type cache (one forward pass per vocabulary token, no context), we discover 175 categories via per-dimension sign consistency (mean AUC 0.80) from 50 anchors with zero training. A trained probe adds only +0.018 AUC and converges to axis-aligned weights, confirming negligible cross-dimension structure. This structure extends to attention: all 175 categories remain discoverable in K and V projections. On the write side, static FFN weight inspection links 20% of features to individual writer neurons (>0.70 agreement; random controls: 0%), with top-200 neuron coalitions achieving >0.70 agreement on 99.9% of prototypes via majority vote. Fully unsupervised discovery (random seeds, no labels) scales to 1500 features at 100% yield and 99% sparsity across all three models, with pairwise MI of 0.0014 bits confirming low inter-dimension coupling. These results establish that the standard basis already suffices for feature reading throughout the transformer compute pathway, requiring no training, no optimization, and no GPU-days beyond a single forward pass per vocabulary token.

Personalized medicine in acute ischemic stroke requires moving beyond average treatment effects (ATE) to individualized treatment effect (ITE) estimates to support treatment decisions. In acute ischemic stroke, mechanical thrombectomy has been shown to be more effective on average than lysis in randomized controlled trials (RCTs), such as the MR CLEAN study. We aim to identify which individual patients benefit most from mechanical thrombectomy compared to lysis. The outcome of interest is the modified Rankin Scale (mRS) at three months, an ordinal measure of functional disability (0: no symptoms, 6: death). We demonstrate that causal transformation models on directed acyclic graphs (TRAM-DAG) can be used for ITE estimation after being fitted on observational MAGIC multi-center stroke patient data. To ensure comparability with the MR CLEAN population, which we use for validation, we train the TRAM-DAG on a MAGIC sub-population with NIHSS at admission >= 6, corresponding to one inclusion criterion of MR CLEAN. The fitted model is then used to estimate ITEs for stroke patients in the MR CLEAN population. While these ITE estimates cannot be confirmed experimentally, we show that their average is consistent with the trial's reported ATE. Furthermore, the ITE estimates correctly rank trial patients by their observed frequency of a good outcome (mRS at three months <= 2). These findings support the use of causal models like TRAM-DAG for personalized decision-making in stroke care and highlight their ability to bridge the gap between observational evidence and clinical trials.

Thanks to the rapid adoption of AI code assistants powered by large language models (LLMs), industry codebases are, increasingly, a hybrid of AI- and human-authored code. For risk management and productivity analysis purposes, it is crucial to enable fine-grained location detection of AI-generated code. To develop algorithms for this task, quality benchmarks are needed to assess performance. However, existing benchmarks tend to comprise academic, LeetCode-style problems and presume a code snippet is either completely human-authored or completely AI-authored, which is not reflective of the diverse intents and styles of industry codebases utilizing AI code assistants. To fill these gaps, we introduce HybridCodeAuthorship, a novel benchmark of Python code files with interleaved human- and AI-authored lines of code to simulate authentic utilization of AI code assistants. In this paper, we first present our dataset construction pipeline, which leverages CodeSearchNet, a massive collection of links to open sourced repositories on GitHub. We then benchmark the performance of two state-of-the-art AI-generated code detection algorithms at both the line- and chunk-level. Experimental results demonstrate that HybridCodeAuthorship is a challenging benchmark with a top-scoring algorithm, AIGCode Detector, obtaining a highest F1 score of 0.48 and 0.56 on chunk-level and line-level code detection tasks, respectively.

Robust lie detectors for language models could enable powerful techniques for auditing, monitoring, and post-hoc investigation of model behaviour, but evaluating them requires testbeds where models verifiably believe the opposite of what they say. We show that existing trained model organisms often fail this requirement, leaving prior positive and negative detection results difficult to interpret. We address this with 13 reasoning model organisms whose hidden beliefs are verified in chain-of-thought and shown to generalise to held-out tasks, alongside Varied Deception, a prompted-lying testbed covering a broad range of lie-inducing motivations. On these testbeds we evaluate four detectors: a chain-of-thought judge, a logprob classifier, and two activation probes, including Did-You-Lie (DYL), a new method for training follow-up probes. On prompted lying, across 31 open-weight models spanning 2B to 1T parameters, all four detectors show positive scaling with model capability. However, every activation- and logprob-based detector drops sharply on our trained model organisms, with DYL retaining the most signal; only the chain-of-thought judge remains strong, with 0.82 balanced accuracy, partly as an artefact of our verification process favouring CoT-readable beliefs. Current lie detectors therefore cannot support high-confidence claims about model beliefs, and we suggest research directions that may address some of their current limitations. We release our datasets, model organisms, and trained detectors.

Closed-loop driving simulators typically populate their environments with non-ego traffic agents that behave largely the same way, produced either by rule-based traffic managers or by learned models trained toward a single behavioral mode. Recent work introduces style variation through post-hoc labels on observational data or LLM-inferred reward weights, but these signals act as proxies for what a style should reward rather than demonstrations of humans explicitly asked to drive in that style. We introduce PersonaDrive, a pipeline that conditions a vision-language-action (VLA) driving agent on retrieved demonstrations from a style-instructed human driving dataset, in which participants drive CARLA leaderboard routes under aggressive, neutral, and conservative instructions on a driver-in-the-loop rig. The pipeline has three stages: (i) offline triplet mining over per-style human driving data using a combined image-text similarity score; (ii) training a lightweight retrieval head that fuses frozen visual features with a small control encoder over per-style databases; and (iii) fine-tuning a single VLA backbone to treat retrieved context points as in-context behavioral demonstrations during waypoint prediction. At inference, the same backbone is conditioned on any style by swapping which per-style database the retrieval head queries, so selecting a style requires no per-style retraining while enabling human-style, style-diverse non-ego agents for closed-loop simulation. On Bench2Drive, PersonaDrive (no style) improves the driving score by 4.6% over SimLingo and 2.5% over HiP-AD, and under style conditioning attains the highest driving score in every style within a roughly 2% band (its weakest style surpassing the strongest baseline, DMW, by 5.4%), while average speed and acceleration rise by 18% and 25% from the conservative to the aggressive instruction.

ML classifiers deployed in high-stakes domains produce predictions whose quality varies systematically across subgroups. For granular subgroups defined by intersections of multiple features, predictions are often inconsistent with the observed data: the model's outputs contradict the evidence available for that subgroup. This problem is exacerbated by regularisation, which improves aggregate performance by collapsing small subgroups into larger groups, disproportionately affecting demographic minorities. We define two requirements for consistent prediction: determinism (identical individuals receive identical predictions) and statistical consistency (we cannot reject, at significance level alpha, the hypothesis that the predictions for a subgroup were drawn from the Bayesian optimal target distribution inferred for that subgroup). From these requirements we derive the Fair Bayesian classifier, which enforces both across every group and subgroup simultaneously and abstains whenever no consistent deterministic prediction is possible. On three benchmark datasets (Adult, COMPAS, and Bank Marketing), standard classifiers produce statistically inconsistent predictions for a substantial proportion of subgroups. Our classifier achieves zero consistency error by construction while exceeding baseline accuracy and multicalibration on every dataset tested. Statistical consistency provides a principled foundation for prediction quality with direct implications for algorithmic fairness. Minority demographics are disproportionately concentrated in small subgroups, precisely where frequentist inference is least reliable; addressing this inference problem is therefore a necessary step toward fair ML. By enforcing Bayesian consistency at the finest resolution the data supports, the our classifier demonstrates that exhaustive subgroup fairness with principled abstention is achievable in practice.

This work investigates the impact of severe class imbalance on the performance of automated machine learning (AutoML) frameworks for multiclass network intrusion detection using the NSL-KDD dataset. Unlike previous studies that simplify the problem through binary classification or minority-class removal, we preserve the original five-class distribution, including highly underrepresented attacks such as R2L and U2R, enabling a realistic evaluation of imbalance-sensitive learning behavior. Nine open-source AutoML frameworks were analyzed under a unified and reproducible experimental protocol, considering differences in architectural design, ensemble strategies, validation procedures, hyperparameter optimization, and imbalance-handling mechanisms. The results demonstrate that frameworks incorporating ensemble learning and imbalance-aware optimization achieve better minority-class discrimination. PyCaret obtained the best overall performance, reaching 66\% macro-F1, followed by AutoGluon with 55\%, whereas frameworks lacking native balancing support exhibited significant degradation in minority-class detection capability. The analysis further shows that accuracy-oriented optimization alone is insufficient for highly imbalanced IDS scenarios, since high-weighted metrics may coexist with poor generalization on rare attack categories. As a contribution, this work establishes a standardized benchmark for AutoML-based intrusion detection under severe multiclass imbalance, highlighting current architectural limitations and the need for native integration of imbalance-aware optimization, resampling, and stratified evaluation strategies into automated learning pipelines. The source code is publicly available.

Two major periods of reduced funding and confidence in artificial intelligence research, commonly called the first and second AI winters, are usually explained through engineering failure, commercial disappointment, and inflated expectations. This article develops a complementary thesis: that the dominant paradigms of those periods also met genuine formal barriers, including limitations of representation, optimisation, computational complexity, statistical learnability, and high-dimensional approximation. The contribution is synthetic rather than archival. We do not claim that particular theorems mechanically caused the winters; rather, we show that several central disappointments of early AI were aligned with mathematically precise bottlenecks. We analyse these bottlenecks through the perceptron impossibility results of Minsky and Papert, the complexity-theoretic hardness of exact neural-network training established by Blum and Rivest, minimax rates for nonparametric estimation in high dimension due to Stone, vanishing-gradient analyses by Hochreiter and by Bengio and collaborators, and classical statistical learning theory in the tradition of Vapnik and Chervonenkis, Valiant, and Blumer and collaborators. We then relate these barriers to the later breakthroughs that mitigated, rather than eliminated, them.

Protein language models are trained on highly imbalanced datasets, raising the question of how they represent underrepresented biological sequences. Using viral proteins as a case study across ESM model families, we identify a dominant nativeness axis in embedding space, aligned with masked reconstruction perplexity, that orders sequences from well-modeled cellular proteins through viral proteins to shuffled and random sequences. Scaling contracts this axis unevenly across viral families. Despite this, protein language model embeddings retain viral-specific signal: viral proteins remain linearly separable beyond zero-shot perplexity and shallow sequence features. Together, these results suggest that pLM representations are structured by a general notion of nativeness while preserving information specific to distinct biological groups.

Conversational shopping assistants now serve hundreds of millions of customers, yet no existing benchmark jointly evaluates the open-ended multi-turn reasoning, domain expertise, and criterion-level quality that real shopping conversations demand. Shopping reasoning is unique among language model applications. Unlike factual question answering or verifiable code generation, it requires balancing subjective preferences, budget constraints, and cross-product trade-offs across multi-turn dialogue, capabilities absent from previous e-commerce and general-purpose benchmarks. We introduce the Shopping Reasoning Bench, an expert-authored benchmark of 525 missions (232 single-turn, 293 multi-turn) with 10863 importance-weighted binary rubrics authored by retail domain experts. These criteria are organized under a taxonomy of five reasoning categories and fifteen subcategories covering diverse demands such as preference refinement, trade-off analysis, and compatibility assessment. An evaluation of nine models across three families (GPT, Claude, Gemini) shows that pass rates reach only 57--77% overall. On multi-turn missions, all models score 13--29 points lower on optional above-and-beyond criteria than on required ones, and performance degrades 4--18 points as conversations progress. These gaps show that current models handle basic shopping assistance but fall short of expert-level advice, making Shopping Reasoning Bench a challenging testbed for future shopping assistant development.

Autonomous long-horizon sidewalk navigation is essential for micro-mobility applications such as robotic food delivery and assistive electronic wheelchairs. Unlike autonomous driving on the road, long-horizon sidewalk navigation requires precise maneuvering through unpredictable sidewalk terrains and pedestrians, with a lightweight perception stack as minimal as a single monocular RGB camera. While imitation learning (IL) from demonstrations offers a practical solution, the resulting autopilot policy often suffers from compounding errors, a lack of social compliance on sidewalks, and deficiencies in counterfactual reasoning to handle complex situations. To address these challenges, we introduce FlowPilot, a mapless navigation policy that achieves robust and efficient long-horizon navigation performance using only a monocular RGB camera. We first propose to use anchored flow matching as an action representation for policy pre-training on large-scale robot fleet data and to capture the diverse, complex, multimodal distribution of sidewalk navigation behaviors. To bridge the gap between imitation and alignment, we further design a human-in-the-loop preference learning scheme to tune the policy on a small amount of human intervention data. It strengthens the model's counterfactual reasoning and social compliance on sidewalks. We evaluate FlowPilot through extensive simulation and real-world experiments in diverse sidewalk environments. FlowPilot achieves 42% success rate and 66% route completion in simulation, while FlowPilot-HP further improves real-world robustness and social compliance, reducing IR by 40.0% and NIR by 52.1% relative to the base model.

Transforming a dense, abstract proverb into an engaging and morally faithful narrative requires deep cultural understanding and robust semantic grounding. We frame this problem as a \emph{constrained semantic decompression} task and study proverb-conditioned story generation as a testbed for abstraction-to-realization in large language models (LLMs). Focusing on Persian, we introduce the Proverb Aligned Narrative Dataset (PAND), pairing proverbs with human-written stories and explicit meanings. By a hybrid evaluation framework that combines human-calibrated LLM-as-a-Judge with structural metrics, we analyze model behavior across multiple prompting regimes. Our findings reveal a persistent \emph{decompression gap}: current LLMs often achieve strong surface-level fluency while failing to faithfully instantiate the underlying moral and causal structure encoded in proverbs. We further show that explicit reasoning and iterative refinement can partially mitigate these failures, suggesting that many decompression errors arise from difficulties in translating abstract meaning into narrative form rather than a complete lack of relevant knowledge. Our proposed task naturally extends to other forms of compressed cultural knowledge.

Foundation models are rapidly transforming Earth observation by enabling scalable pretraining across diverse unlabeled geospatial modalities. However, their architectural diversity ranging from encoder-only to encoder-decoder and masked autoencoding paradigms makes it challenging to assess performance trade offs in a consistent manner. In this work, we present an apples-to-apples comparison of leading FM architectures designed for geospatial multimodal reasoning, with a particular focus on flexibility across varied spectral band configurations. We standardize pretraining using identical self supervised learning objectives and training datasets, and evaluate all models under consistent parameterization on the GEOBench benchmark across classification and segmentation tasks. Our results offer new insights into the design trade-offs between model flexibility, modality alignment, and downstream task performance. By highlighting architectural strengths and limitations under controlled conditions, this study provides practical guidance for building next generation geospatial foundation models capable of robust multimodal reasoning.

Modern Lean theorem provers achieve strong performance only with substantial training and inference compute, driven in part by scarce verified proof data and the long reasoning traces of formal proof search, making both supervised fine-tuning (SFT) and sampling expensive. We introduce Pythagoras-Prover, a compute-efficient open-source family of Lean theorem provers built for practical compute budgets. The family spans two generation paradigms: autoregressive models at 4B and 32B parameters, and a first proof-of-concept diffusion-based prover (4B) that iteratively refines Lean proofs at inference time. For training efficiency, we build a Lean-verified corpus stratified into easy, medium, and hard problems for curriculum SFT, so models acquire proof skills progressively from shorter, simpler proofs to longer, harder ones. During SFT, a dynamic proof-reasoning filtering scheme preserves informative proof traces while keeping each instance within an 8k-token context budget. We also introduce Augmented Lean Formalisation (ALF), which expands scarce verified corpora into variants of formal statements, populated via self-distillation for extra training signal without formally verifying every mutated instance. By perturbing known problems while preserving their formal character, ALF reduces reliance on any statement's surface form. Empirically, Pythagoras-Prover-4B surpasses DeepSeek-Prover-V2-671B at pass@32 on MiniF2F-Test (86.1% vs 82.4%) with ~167x fewer parameters, while Pythagoras-Prover-32B sets the open-source state of the art at 93.0% on MiniF2F-Test and solves 93 of 672 PutnamBench problems. We release MiniF2F-ALF, an ALF-mutated contamination-sensitive benchmark on which every evaluated model loses accuracy; here our 32B remains strongest and our 4B matches the prior state of the art, Goedel-Prover-V2-32B.

The Internet of Things (IoT) is experiencing rapid growth. Smart devices are emerging in smart homes and industrial applications, performing mission-critical tasks. Bugs in IoT software can lead to severe consequences. For example, a buggy smart lock can allow unauthorized access to a private property. Testing is a primary practice to expose software bugs and ensure software quality. However, little is known about how IoT software is tested. To bridge this gap, we conducted the first empirical study on test cases in open-source IoT software. Specifically, we evaluated the effectiveness of test cases in IoT software, explored the challenges inherent in testing IoT software, and analyzed the usage of mock objects. Our results indicate that while IoT software often contains a considerable number of tests, their effectiveness remains limited. We identified the primary challenges in testing IoT software as managing complex interactions with various external dependencies, such as other network-reliant IoT components, file systems, operating systems, and databases. We also observed that the use of mock objects in IoT software closely aligns with our identified testing challenges. This alignment demonstrates the potential of mocking as a solution to enhance test coverage and address the complexities of IoT software testing.

Large Vision-Language Models (LVLMs) have achieved strong performance across medical imaging tasks, yet they remain prone to factual inconsistencies, poor visual grounding, and misalignment with clinically meaningful feedback. Existing post-training alignment approaches, including Direct Preference Optimization (DPO) and its variants, face three critical limitations in the medical domain: (1) sequence-level reward signals treat clinically critical tokens identically to generic filler text; (2) reliance on static supervised fine-tuning references as preferred responses introduces an off-policy distribution shift, steering optimization toward stylistic artifacts over clinical correctness; and (3) alignment objectives lack explicit visual grounding constraints, leaving models insensitive to subtle yet diagnostically decisive pathological features. Our method leverages a bidirectional token-wise KL regularizer alongside a visual-contrastive grounding objective that pairs clean and lesion-corrupted images to penalize responses generated without adequate visual evidence. Together, these components form a fine-grained, on-policy alignment framework that constructs preference pairs by minimally editing model-generated outputs, correcting only clinically erroneous spans while preserving the original linguistic style. Extensive experiments across medical imaging tasks and clinical text generation benchmarks validate the effectiveness of our approach.

Traditionally, decision support studies how humans use machine learning models to make better decisions. In modern agentic systems, this division of roles is increasingly reversed: AI agents act on behalf of users, while humans and tools becomes support mechanisms around them. This role reversal brings reliability concerns to the forefront, since agentic errors can be consequential and agent behavior must remain aligned with human goals and constraints. Departing from the classical view of decision support, we revisit its two basic principles, the cost--value tradeoff of seeking support and the role of uncertainty quantification, in a setting where AI agents are the central actors. We propose a framework for strategic decision support for AI agents through an optimization problem that minimizes support usage subject to controlling a counterfactual missed-support error: the probability that the agent acts alone on instances where support would have materially improved its output. At the population level, we show that the optimal policy is a threshold rule on the value of support. Building on this structure, we develop an online algorithm that adaptively thresholds such a score and uses randomized exploration to control missed-support error without distributional assumptions. We further introduce a calibration-on-the-fly method that reduces unnecessary support calls online. We instantiate this framework across diverse scenarios, including information gathering, human--AI collaboration, and tool use, showing how each can be modeled through the same strategic decision-support lens. Experiments across these settings show that our method reliably controls the target error while substantially reducing support usage in practice.

Real-world networks are inherently incomplete, noisy, and dynamically evolving, making it difficult to capture all actors and their relationships. Their scale often renders direct analysis computationally demanding. While influence maximisation (IM) has been widely studied, the role of graph reduction as a preprocessing step, and its impact on IM accuracy, remains underexplored. In this work, we introduce the Spreading-Oriented Reduction Benchmark (SORB), an open-source, standardised framework for systematically evaluating IM models across diverse task settings. SORB provides an extensible pipeline operating on a representative collection of real-world networks, including single- and multilayer structures, and accounts for graph reduction directly into the evaluation process. This design shifts the focus from analysing IM algorithms in isolation to quantifying how graph reduction alters predictive performance. Using SORB, we study the effects of sparsification and coarsening across multiple IM scenarios. Our results show that the impact of reduction is strongly dependent on both the network type (single-layer vs. multirelational) and the downstream task ($Gain@k$ vs. $\mathrm{AUC}_{\mathrm{cutoff}}$): sparsification preserves seed set quality on single-layer networks, whereas flattened multilayer networks exhibit systematic ranking degradation regardless of reduction strategy. These findings highlight the importance of reduction-aware, multi-task evaluation when studying spreading processes in complex networks.

Mechanism-level drug-drug interaction (DDI) prediction requires identifying which enzyme or pharmacodynamic axis is implicated, in which direction, and with which evidence -- not merely whether two drugs interact. We introduce a reproducible mechanism-level DDI labelling and evaluation protocol with a structured 7-family/147-subtype taxonomy, leakage-safe cold-split protocols, and auditable reasoning metrics for evaluating pharmacological prediction beyond flat interaction classification. We propose a pipeline that produces a 7B reasoning MARD (Mirror-Augmented Reasoning Distillation), combining three training innovations: a single-token KL divergence on direction tag that ties the model's prediction, per-loss PRM-weighted DPO with programmatic hard negatives, and a leakage-safe mechanism-aware retrieval channel. Process-reward step labels are automatically verifiable against DrugBank-structured fields, requiring no human or LLM judges. On the April-2026 DrugBank release, our MARD-7B is the only system in a 32-system comparison whose accuracy survives drug-pair novelty, beating the best baseline by +13.9 pp and GPT-4o by +6.7 pp at ~1% of frontier API cost. Further analysis reveals an anti-memorisation signature where accuracy improves on rarely seen drugs, suggesting that gain comes from structured pharmacological reasoning rather than drug-frequency memorisation. We release corpus, DDI-PRM, retrieval index, and training code.

Scientific figures compress complex pipelines into a single canvas, yet understanding them requires paper-grounded, step-by-step narration aligned with visual highlights a capability missing from current video generation systems and benchmarks. To address this, we introduce paper-grounded figure-to-video generation: generating narrated, region-grounded walkthrough videos from a figure and its paper. We propose MINARD (Multimodal Interpretation of Narrated Architecture via Region Decomposition), a pipeline that generates paper-grounded narrations and sequentially grounds them to figure regions. We also release FigTalk, a benchmark with new sequential and component-level grounding metrics derived. On FigTalk, MINARD generates humanlike, paper-faithful narrations and outperforms narration-conditioned figure spatial grounding compared to existing approaches in both automatic and human evaluation

We present EDEN (Emergency Department Electronic Notes), a new and unique large-scale corpus of clinical notes produced in Emergency Departments of Italian hospitals. The corpus, in its current version, is composed of approximately 4 million clinical notes fully anonymized, covering diverse phases of patient care during the stay in the emergency department. In addition, a subset of about six thousand notes has been manually annotated by clinical experts through a structured Case Report Form (CRF) containing 132 items relevant for two patient situations in emergency departments, dyspnea and loss of consciousness. Items may assume numerical values (e.g., for blood saturation), categorical (e.g., for level of consciousness ), binary (e.g., for presence of traumas), and mixed value types. The annotation process involved multiple clinicians and underwent iterative revision to resolve ambiguities in item formulation, resulting in a richly structured (although high imbalanced) resource. The dataset aims to fill a relevant gap of data able to support both the development and the use of Large Language Models in concrete medical applications. We describe the data collection protocol, the on-site anonymisation pipeline, corpus statistics, and the annotation scheme. Finally, we propose CRF-filling as a novel structured information extraction benchmark, and provide zero-shot baseline resulting from Gemma-27B and MedGemma-27B. To the best of our knowledge, the EDEN dataset is the largest freely available corpus of clinical notes existing for the Italian language.

Arbor is a multi-agent framework that introduces structured tree search as a cognition layer for autonomous agents operating in large, stateful action spaces. Prior autonomous optimization systems operate on isolated targets with stateless evaluation. Arbor instead maintains an explicit search tree of scored hypotheses that serves as the shared working memory across agents, evolving with every measurement, treating failures as diagnostic signal that reshapes subsequent exploration, and expanding as prior successes shift the bottleneck distribution. We validate Arbor on full-stack LLM inference optimization, a domain where achieving peak performance has historically required coordinated effort from engineering teams across the application, framework, compiler, kernel, and hardware stack. Arbor pairs an Orchestrator agent, which drives optimization by delegating to Domain Specialists across the inference stack, with a Critic agent that safeguards stability through root-cause analysis, introspection, and measurement validation -- a checks-and-balances architecture where neither agent can unilaterally drive the system. Agent capabilities are decomposed into hard skills (domain expertise) and soft skills (coordination protocols that determine how contributions compose), enabling fully autonomous multi-day campaigns. Arbor achieves up to 193% inference throughput-latency Pareto improvement over vendor-optimized baselines, while a single agent without the harness plateaus at +33% throughput improvement and crashes irrecoverably within hours. Arbor generalizes to multiple generations of hardware platform, and run-to-run variance is within 2 percentage points demonstrating that the method is hardware-agnostic and reproducible.

The ensemble Kalman filter (EnKF) is widely adopted for sequential data assimilation, but fails for solutions with discontinuities, such as shocks in compressible flows. Uncertainty in shock location induces multimodal ensemble statistics that violate the Gaussian assumptions underlying the EnKF, producing large-scale spurious oscillations in the analysis state. We introduce a feature-preserving latent-EnKF that performs the ensemble update in a learned low-dimensional latent space, where shock and flow features admit a smooth manifold representation, thereby preserving sharp features during EnKF analysis. The updated latent state is mapped back to physical state through a shared decoder for all ensemble members. The algorithm eliminates the member-specific ordered training and positivity flooring used in prior approaches. Numerical experiments on a Sod shock tube and Mach 2 shock interaction with a 2D cylinder, using sparse and noisy observations, show accurate feature recovery of shocks and contact discontinuities without spurious oscillations.

The rapid shift toward agentic and long-context workloads in Large Language Models (LLMs) is pushing the industry beyond the capacity of individual servers toward disaggregated shared storage to handle TB-scale context states. This movement has led to the emergence of specialized shared context layers designed to externalize and share cumulative inference states across distributed clusters. While offloading to a data processing unit (DPU) within just-a-bunch-of-flash (JBOF) architectures accelerates NVMe-over-fabrics (NVMe-oF) target processing, the need for sophisticated software-level optimization and cost-efficiency burdens remain significant. Consequently, the ideal architecture for scaling this shared context infrastructure is still an active area of exploration. In this paper, we propose ITME (Inference Tiered Memory Expansion), which leverages a CXL-hybrid memory to present a massive, TB-scale byte-addressable remote memory expansion. This approach enables cost-efficient scaling and simplifies the software stack through direct byte-addressability, effectively addressing the challenges of shared context infrastructure. Our key insight is that the deterministic access patterns of voluminous model weights and prefix caches enable the system to proactively manage data movement across the memory-storage hierarchy. We validate ITME by evaluating its performance potential with production-grade SK Hynix CMM and PCIe Gen5 NVMe SSDs, while further demonstrating its functional feasibility through an FPGA-based hardware prototype. Overall, ITME enhances conventional CPU-offloading by providing additional remote memory expansion to accommodate large KV cache footprints beyond host memory limits, achieving up to a 35.7\% throughput improvement.

Modern machine learning progresses through empirical work, benchmarking new methods to evaluate relative performance. However, the statistical variability inherent to evaluation - exacerbated by the stochastic nature of many algorithms - often makes performance estimation unreliable due to the limited test samples available, leading to a validation crisis in which genuine advances are difficult to discern. In this work, we show that cross-validation improves markedly confidence when evaluating and comparing learning algorithm performances. We introduce the concept of sample gain, which quantifies the virtual data augmentation achieved by using multiple cross-validation splits to reduce benchmarking variance. Experiments on both synthetic and real-world datasets (histopathologic scans and NLP fine-tuning) demonstrate that multiple splits can substantially improve the reliability and stability of performance estimates, with diminishing returns often setting in later than expected. We also introduce a procedure to dynamically early-stop cross-validation by estimating from the first few folds if subsequent folds will bring large sample gains. Our findings highlight the value of pushing cross-validation on available samples to achieve robust and reliable benchmarking.

Open-world mapless navigation from sparse language instructions requires resolving underspecified goals and inferring which environmental cues are relevant for reaching the goal. For instance, reaching an out-of-view destination may require interpreting ramps, signs, or detours that reveal where to go or which route to take. Prior works are limited by their reliance on known navigation factors and closed-set factor categories, or identify cues before motion planning and miss plan-dependent cues. We argue that pretrained Vision-Language Models (VLMs) can discover novel instruction-relevant cues, but require adaptation to focus on which cues matter and how they should influence motion planning. We realize these ideas in Foresight, a test-time framework in which a finetuned VLM alternates between proposing image-space motion plans and critiquing them using the language goal and visual context. Subsequent plans are conditioned on prior critiques, enabling iterative motion refinement before execution. To align plan critiques and refinements with open-set behavior preferences, we learn a reward model from human feedback and use it to post-train the VLM with reinforcement learning in the plan-critique loop. In offline evaluations and 6 real-world environments, Foresight improves average task success by 37% and reduces interventions per mission by 52% relative to state-of-the-art test-time reasoning and foundation-model baselines, while running in real-time on a Jetson AGX Orin. We will release code, data, and training details to support future work on test-time reasoning for robot motion refinement. Additional videos at: https://amrl.cs.utexas.edu/foresight

Vision-language models (VLMs) project images into hundreds to thousands of visual tokens, making decoder inference expensive in both attention computation and KV-cache memory. Existing visual-token reduction methods largely follow a rank-and-remove paradigm: they score visual tokens, keep a compact subset, and permanently discard the rest. We show that this irreversible action is fragile because visual-token importance changes across decoder depth; tokens ranked low at one stage may become relevant in later layers, especially for grounding-sensitive queries. We propose Reroute, a training-free plug-in that replaces removal with recoverable routing. At each routing stage, selected vision tokens pass through decoder blocks, while deferred tokens bypass the stage and re-enter the candidate pool at the next routing decision. Reroute reuses existing attention-score ranking rules and stage-wise schedules, preserving the theoretical TFLOPs and KV-cache budget class of the pruning method it augments. Across FastV, PDrop, and Nüwa variants on LLaVA-1.5 and Qwen backbones, reroute improves grounding under aggressive token reduction while maintaining general VQA performance. These results suggest that VLM token reduction should not be viewed only as irreversible pruning, but also as recoverable routing. The code can be found here: https://github.com/elmma/mllm-reroute/

Modern conversational agents condition on an ever-growing dialogue history at each turn, incurring redundant attention and encoding costs that grow with conversation length. Naive truncation or summarization degrades fidelity, while existing context compressors lack cross-turn memory sharing or revision, causing information loss and compounding errors in long dialogues. We revisit the context compression under conversational dynamics and empirically present its fragility. To improve both efficiency and robustness, we introduce Context-Driven Incremental Compression (C-DIC), which treats a conversation as interleaved contextual threads and stores revisable per-thread compression states in a single, compact dialogue memory. At each turn, a lightweight retrieve, revise, and write-back loop shares information across turns and updates stale memories, stabilizing long-horizon behavior. In addition, we adapt truncated backpropagation-through-time (TBPTT) to our multi-turn setting, learning cross-turn dependencies without full-history backpropagation. Extensive experiments on long-form dialogue benchmarks demonstrate superior performance and efficiency of C-DIC; notably, C-DIC shows stable inference latency and perplexity over hundreds of dialogue turns, supporting a scalable path to high-quality dialogue modeling.

Contact-rich manipulation requires force sensitivity, but many robot arms lack dedicated force sensors due to their high cost. We present Neural External Torque Estimation (NEXT), a data-driven method that estimates external joint torques without needing any dedicated force sensors. NEXT trains in 1 minute from only 10 minutes of free-motion data, yet achieves estimates comparable to dedicated joint-torque sensors. NEXT enables force-feedback teleoperation on low-cost arms and improves policy learning through Force-Informed Re-Sampling Training (FIRST), which up-samples pre-contact and contact segments during behavior cloning. Across five long-horizon tasks, FIRST outperforms prior force-aware policies by over 17% in task progress. Together, NEXT and FIRST bring force-aware teleoperation and policy learning to off-the-shelf robots without additional sensing hardware. Video results and code are available at https://jasonjzliu.com/factr2

Vision-Language Models (VLMs) are increasingly deployed as high-level planners for embodied agents, with an emerging strategy of scaling test-time compute to improve capability. However, we observe that doing so increases latency, token usage, and FLOPs while yielding uneven, often diminishing gains in downstream success, limiting where embodied agents can be deployed. We argue that choosing when and where to spend test-time compute is central to bringing frontier performance to the real world. We introduce DIRECT, a routing framework that uses multimodal scene context to allocate compute per prompt, improving the success--cost Pareto frontier over fixed model selection. Across three dominant scaling axes, namely chain-of-thought depth, model size, and memory history, our experiments on VLABench and RoboMME show that test-time compute is not a uniform lever: different axes yield qualitatively distinct capability gains. We validate these insights on a physical Franka arm in a DROID setup spanning zero-shot manipulation and long-horizon chaining, where our router matches or exceeds a stronger model's success rate at up to 65% lower average latency. Ultimately, our results show that naively scaling test-time compute is wasteful, and that DIRECT can provide frontier-level embodied planning in robotic systems at a fraction of the cost. Project page can be found at jadee-dao.github.io/direct/.

Long input sequences are central to document understanding and multi-step reasoning in Large Language Models, yet the quadratic cost of attention makes inference both memory-intensive and slow. Context distillation mitigates this by compressing contextual information into model parameters, and recent work such as Doc-to-LoRA amortizes context distillation into a single forward pass that generates one LoRA adapter per document. However, producing a single monolithic adapter for all queries leads to irrelevant-query interference, limited compositional recall, and poor scalability to long-document reasoning. To address these challenges, we propose Doc-to-Atom (Doc2Atom), a compositional parametric memory framework that decomposes each document into semantically typed knowledge atoms. Each atom is compiled into an independent micro-LoRA adapter and a provenance retrieval key. At inference time, a lightweight query router selects and assembles only the relevant atoms into a query-specific adapter, which is then injected into a frozen base model. The entire system is trained end-to-end through a multi-objective distillation framework. Experiments on six diverse QA benchmarks demonstrate that Doc2Atom outperforms Doc-to-LoRA baselines while reducing the memory cost of document internalization.

Router is the cornerstone component to the Mixture-of-Experts models. Serving as expert proxies, the rows of the router matrix compute their similarity to the MoE inputs to determine which subset of experts is activated. Ideally, each router row is designed to encode the expert matrix into this representative vector, such that its dot-product with token can better reflect token-expert affinity. However, there exists no design principles to enforce this condensation. In this paper, we propose to align each router row with the principal singular direction of the associated expert, as this direction provides the most expressive mathematical description of a matrix. Based on this principle, we propose a router redesign with Manifold Power Iteration (MPI). Specifically, it introduces a "Power-then-Retract" paradigm, where a power iteration step is performed on the router weights, followed by a retraction to impose a norm constraint to ensure both efficiency and stability. Theoretically, we show that MPI drives router rows to converge toward the principal singular directions of associated experts. Empirically, we pretrain MoE model across scales from 1B to 11B parameters to confirm that this alignment facilitates more effective MoE models.

Recently, large language models (LLMs) have achieved promising progress in the fields of classical Chinese translation and the generation of classical poetry. However, domain-specific research on precise translation and affective-semantic understanding of classical poetry remains limited. The main challenge is that most studies treat the poetic appreciation task as a general-domain problem, neglecting the distinctive features of poetic appreciation, while high-quality and domain-specific datasets are extremely limited. To address this limitation, we decompose the task into three subtasks: term interpretation, semantic interpretation, and emotional inference. Based on multiple open-source datasets, we perform data cleansing and alignment to construct the Classical Chinese Poetry Instruction Pair Dataset (CCPoetry-49K), which comprises 49,404 high-quality instruction-response pairs explicitly optimized for this domain. We then propose a domain-specialized LLM, called PoetryQwen, by applying Low-Rank Adaptation (LoRA) to fine-tune the Qwen2.5-14B model. Experimental results on the CCL25-Eval Task 5 benchmark demonstrate that PoetryQwen achieves a score of 0.757, representing a 9.7% improvement over the Qwen2.5-14B-Instruct baseline (0.690). These findings clearly indicate that PoetryQwen significantly enhances performance in precise translation and emotional understanding of classical poetry. We present new dataset and methodological considerations intended to support the domain-specific optimization of LLMs.

Rubrics have emerged as an alternative to RLVR in open-ended domains where a single ground-truth final answer is not available. Existing rubric-based training methods rely on an LLM verifier that scores each rollout against rubrics. This introduces substantial training-time overhead, exposes optimization to verifier-specific biases, and reduces rubric feedback to a sparse end-of-trajectory signal. We propose Rubric-Guided Self-Distillation (RGSD), a verifier-free training method in which the base policy, conditioned on the rubric, serves as the teacher for the unconditioned student. RGSD distills the rubric-conditioned teacher distribution into the student token-by-token, replacing sparse trajectory-level rewards with dense per-token learning signals and removing the LLM judge from the training loop entirely. Across Qwen-2.5 (3B, 7B) and Qwen3-Thinking (4B, 8B) models on medical and science domains, RGSD achieves rubric satisfaction comparable to judge-based GRPO while using one on-policy rollout per prompt and no training-time verifier calls. Ablations show that raw rubrics provide a stronger teacher enrichment signal than self-generated reference responses, while a stronger GRPO judge can outperform RGSD in some settings, positioning RGSD as a complementary verifier-free alternative when verifier cost or reliability is the bottleneck.

Large Language Models (LLMs) have democratized database access through Text-to-SQL, but moving from prototypes to production remains difficult. Real deployments must handle strict SQL dialects, massive schemas, and evolving user preferences, while supervised fine-tuning is costly and rigid and agentic test-time scaling is expensive. We present Tahoe, a system that treats prompt optimization as a dynamic data management problem. Tahoe uses an error-driven hint learning pipeline across Development and Deployment to consolidate debugging traces into a structured Hint Bank. Compiler feedback is distilled into reusable Syntax Hints for dialect-specific rules, while execution and user feedback are converted into Semantic Hints for schema- and user-specific logic. Tahoe further introduces a Strategy Layer that models conflicting user intents as competing strategies under shared natural-language triggers, with recency signals and post-learning attribution statistics that summarize empirical success, harm, inertness, and support. At inference time, Tahoe retrieves relevant hints and guides the LLM through Logic Planning followed by SQL Synthesis. We implement and evaluate the development-phase workflow, leaving deployment-time human-feedback updates for future work. On Spider 2.0-Snow, Tahoe substantially improves Text-to-SQL without updating model parameters. On 113 supervised Spider 2.0-Snow-0212 examples using GPT-5.5, Tahoe raises pass rate from 61.95 percent to 79.42 percent and pass-at-4 from 72.57 percent to 87.61 percent, achieves 100 percent Snowflake syntax pass rate, and reduces average compiler-feedback critic rounds from 2.79 to 0.12 per sampled candidate. The same Hint Bank also transfers to weaker backbones, including a 19.7 percentage-point pass-rate gain on Doubao-2.0-lite.

Advancing scientific understanding through mechanistic modeling requires posing the right experimental questions to yield maximally informative data. To automate this pursuit within cognitive science, we introduce ATLAS (Active Theory Learning for Automated Science), an active learning framework for the data-driven discovery of interpretable behavioral models. ATLAS iterates between generating mechanistic hypotheses--instantiated as a diverse ensemble of sparse neural networks (Disentangled RNNs)--and designing experiments that optimally distinguish between them. We test this approach on the problem of recovering reinforcement learning agents from their behavior in bandit tasks. ATLAS designs varied sequences of qualitatively novel experiments with temporal structure tailored to underlying agent characteristics. The models trained on these experiments are evaluated against a comprehensive set of metrics for mechanistic modeling that capture behavioral, structural, and computational similarity. ATLAS achieves a 5-10x improvement in sample efficiency across all metrics compared to random experimentation, and its performance is further validated against expert-designed experiments derived from literature. These in silico results showcase ATLAS's potential to accelerate human-interpretable insights in cognitive science and other domains where scientific inquiry relies on discovering mechanistic models.

Modern LLM training pipelines increasingly rely on other models to generate data, filter corpora, judge outputs, and guide development decisions. These dependencies are recursive: a model may depend on an upstream artifact whose own dependencies are documented only in separate releases and artifacts. As a result, the full dependency structure is fragmented across heterogeneous public artifacts, with complexity and recursive depth far outpacing humans' ability to trace. We introduce ModSleuth, an agentic system that recursively reconstructs LLM dependency graphs from public artifacts with source-grounded evidence. We find that the primary challenge is no longer information extraction, but defining what constitutes a dependency and reconciling artifact references across inconsistent documentation. We address these challenges through a formalization that distinguishes direct and indirect dependencies, represents heterogeneous pipeline roles through operation-centered relationships, and resolves artifact identities across names, versions, and repositories. Applying ModSleuth to four public-artifact-rich LLM releases, we recover 1,060 source-verified dependencies and construct large-scale dependency graphs of modern LLM development. These graphs reveal multi-hop license obligations, train-evaluation coupling, discrepancies between released and training-time artifacts, and documentation inconsistencies that would otherwise be difficult to uncover. We release ModSleuth and the resulting dependency graphs to support transparent analysis of the increasingly complex ecosystems underlying modern LLMs.

Recent advances in agentic Reinforcement Learning (RL) have substantially improved the multi-turn tool-use capabilities of large language model agents. However, most existing methods assign credit over coarse heuristic units, such as tool-call boundaries or fixed workflows, making it difficult to identify which intermediate decisions influence downstream outcomes. In this work, we study agentic RL from two perspectives: \textit{where to branch and how to assign credit after branching}. Our pilot analysis shows that influential decision points are broadly distributed throughout the generated sequence rather than concentrated at tool calls, while token entropy alone does not reliably reflect their impact on final outcomes. Motivated by these observations, we propose \textbf{Agentic Procedural Policy Optimization (APPO)}, which shifts branching and credit assignment from coarse interaction units to fine-grained decision points in the sequence. APPO selects branching locations using a Branching Score that combines token uncertainty with policy-induced likelihood gains of subsequent continuations, enabling more targeted exploration while filtering out spurious high-entropy positions. It further introduces procedure-level advantage scaling to better distribute credit across branched rollouts. Experiments on 13 benchmarks show that APPO consistently improves strong agentic RL baselines by nearly 4 points, while keeping efficient tool-calls and maintaining behavior interpretability.

The Strength Pareto Evolutionary Algorithm 2 (SPEA2) is a popular and prominent evolutionary algorithm for solving multi-objective optimisation problems. Despite its popularity, theoretical analyses of SPEA2 have only appeared recently. Moreover, these analyses focus exclusively on how SPEA2 handles non-dominated solutions and disregard the algorithmic components responsible for handling dominated solutions. We conduct a first runtime analysis of SPEA2 for which these components are analysed. We prove that, unlike other prominent algorithms, including NSGA-II, NSGA-III and SMS-EMOA under the same setting of constant population size and duplicate elimination, SPEA2 is unable to cover the Pareto front of the OneTrapZeroTrap benchmark efficiently. Our results indicate that using k-th nearest-neighbour distance in the fitness assignment provides an insufficient signal to maintain diversity among dominated individuals. To address this issue, we propose an improved variant, SPEA2$^+$, that considers all pairwise distances. The new algorithm achieves the same performance guarantees as the other prominent algorithms on OneTrapZeroTrap, while matching the performance of the original SPEA2 on simpler problems. Experimental results complement our theoretical findings.

Physiological awareness is important for service, social, and assistive robots that interact with humans in everyday environments. Remote photoplethysmography (rPPG) enables non-contact heart-rate (HR) estimation from an RGB camera, making it a promising sensing modality for robot-mounted vision systems. However, illumination variation remains a major barrier to robust deployment. This paper presents an end-to-end spatial-temporal transformer framework for remote HR estimation on a new dataset with varied illumination. Our estimator integrates PRNet-based 3D face alignment, clip-level illumination augmentation, the Residual Temporal Standardization Module, and controlled hybrid temporal-frequency supervision. The training objective combines a Soft-Shifted Pearson waveform loss with a spectral Kullback-Leibler divergence loss, where a tuned weight ($\mathbfβ$) controls the contribution of frequency-domain heart-rate guidance. Experiments on a static all-level mix protocol covering three illumination levels show that $\mathbfβ=5$ provides the strongest result among the tested beta settings, achieving a best-run HR mean absolute error (MAE) of 0.79 bpm and an HR correlation of 0.982. Compared with the PhysFormer baseline evaluated on our dataset, our estimator reduces HR MAE by 93.6 %, while increasing HR correlation from 0.088 to 0.982, making it usable when illumination varies.

Reinforcement Learning (RL) with verifiable environments has emerged as a powerful approach for enhancing the reasoning capabilities of Large Language Models (LLMs). While prior research demonstrates that scaling environment quantity improves RL performance, existing manual or individual construction methods suffer from linear scaling limits, thereby hindering scalable reasoning generalization. This paper introduces RACES (\textbf{R}ecursive \textbf{A}utomated \textbf{C}omposition for \textbf{E}nvironment \textbf{S}caling), a framework that conceptualizes verifiable environments as composable building blocks that can be recursively assembled. The key insight is that when the codomain (output type) of one environment matches the domain (input type) of another, they can be automatically fused into a new verifiable environment, enabling recursive composition. RACES is implemented with 300 individual environments and defines a set of composition operators (\textsc{SEQUENTIAL}, \textsc{PARALLEL}, \textsc{SORT}, and \textsc{SELECT}) that induce diverse reasoning patterns. Extensive experiments show that RL training on these composite environments consistently enhances reasoning generalization. Specifically, RACES improves DeepSeek-R1-Distill-Qwen-14B by an average of 3.1 points (from 48.2 to 51.3) and boosts Qwen3-14B performance from 58.8 to 61.1 on six benchmarks, which are unseen during the construction of training environments. Moreover, RACES achieves performance comparable to training on 300 individual environments using only 50 base environments, demonstrating significant efficiency in environment utilization.

Human-in-the-loop reinforcement learning (HiL-RL) has emerged as an effective paradigm for real-world robotic manipulation, enabling online policy improvement with human guidance. However, current HiL-RL frameworks remain intervention-intensive, relying on frequent human corrections to redirect the policy out of unproductive exploration, which incurs high labor cost and limits real-world scalability. To address this, we propose UniIntervene, an agentic intervention model that detects unproductive exploration and autonomously recovers the policy toward high-value states, taking over the bulk of interventions from human operators. Specifically, UniIntervene first performs future-conditioned action-value estimation, predicting the latent consequence of the current action and evaluating its induced value, which provides a more stable progress signal. Building on this, a temporal value-risk critic aggregates recent value dynamics and triggers intervention when the estimated value exhibits sustained stagnation or degradation. When intervention is required, UniIntervene retrieves a high-value recovery target from a memory of past intervention episodes and produces executable corrective actions through a goal-conditioned recovery policy. In this way, UniIntervene turns intervention from passive human correction into a value-aware recovery process for efficient real-world RL. Extensive experiments on diverse real-world manipulation tasks demonstrate that UniIntervene improves the average success rate by 8.6% while reducing human interventions by 57% relative to state-of-the-art HiL-RL baselines.

Reinforcement learning (RL) has become a key component in modern large language models, yet the rollout stage remains the key bottleneck in RL training pipelines. Although Multi-Token Prediction (MTP) offers a natural solution to accelerate rollouts through speculative decoding, many studies have observed that MTP acceptance rates degrade significantly during RL training, leading to limited speedup performance. To address this bottleneck, we present Bebop, a systematic study of MTP in LLM post-training, and offer practical recipes to integrate MTP into large-scale RL pipelines. First, we reveal that the MTP acceptance rate is fundamentally bounded by the fluctuation of model entropy, which demonstrates a clear negative linear relationship with the rise of entropy in the RL stage. Second, we show that probabilistic rejection sampling largely alleviates the disturbance introduced by entropy in RL compared to greedy draft sampling. We further identify that the conventional MTP training objectives (cross-entropy or KL) are suboptimal in such settings, and therefore we propose a novel end-to-end TV loss that directly optimizes multi-step rejection sampling acceptance rate, yielding ~10% acceptance rate improvements, achieving up to 95% acceptance rates and up to 25% extra inference throughput gains across mathematical reasoning, code generation, and agentic tasks. Third, we test various online MTP training strategies during RL and show that pre-RL MTP training with e2e TV loss and rejection sampling achieves a consistent acceptance rate and speedup throughout the entire RL, eliminating the need for costly online MTP updating. We provide extensive experiments and analysis that validate our findings. Experimental results show our method achieves up to 1.8x end-to-end acceleration in async RL training of Qwen3.5, Qwen3.6, and Qwen3.7 models.

We propose Ambient Diffusion Policy, a simple and principled method for imitation learning from suboptimal data in robotics. High-quality, task-specific robot data is expensive and time-consuming to collect, while suboptimal datasets with lower-quality or out-of-distribution demonstrations are abundant. Existing methods that co-train on both data sources in robotics often fail to separate the meaningful and the harmful features in the suboptimal samples. In contrast, our method extracts only the useful features by introducing a new axis to co-training in robotics: noise-dependent data usage. Ambient Diffusion Policy restricts the contribution of suboptimal data during training to only the high and low diffusion times. To rigorously justify our approach, we first observe that robot action data exhibits a spectral power law. This induces two important properties on the optimal Diffusion Policy that we exploit: a global-to-local hierarchy and locality. We theoretically formalize this discussion using a simplified model. Our experiments validate Ambient Diffusion Policy on four types of suboptimal action data (noisy trajectories, sim-to-real gap, task mismatch, and large-scale data mixtures) across six tasks. The results show that it effectively learns from arbitrary sources of suboptimal data. Notably, it outperforms existing co-training baselines by up to 33% when scaled to Open X-Embodiment - a large dataset with heterogeneous data quality and unstructured distribution shifts. Overall, Ambient Diffusion Policy increases the utility of suboptimal demonstrations and expands the set of usable data sources in robotics.

Transformers dominate modern sequence modeling, but their quadratic attention incurs substantial computational cost. Subquadratic architectures offer a scalable alternative. However, it remains unclear which designs yield the most effective sequence models. We compare three leading approaches: xLSTM, Mamba-2, and Gated DeltaNet. We evaluate these models on tasks with complex dependencies: (1) code-model pre-training, (2) distillation of code models from large language models, and (3) pre-training of time-series foundation models. Across these settings, xLSTM delivers the strongest overall performance. To explain xLSTM's advantage, we present a unified formulation and analyze the underlying architectural mechanisms, focusing on state tracking and memory dynamics. Our results show that xLSTM enables more flexible and stable memory correction via its gating scheme. We corroborate these findings on controlled synthetic length-generalization tasks. Overall, our findings indicate that xLSTM's gains on complex tasks stem from robust state tracking and accumulation.

We study multimodal learning under missing modalities, with particular motivation from bioscience applications in which heterogeneous modalities are often only partially available when decisions need to be made. We propose Latent World Recovery (LWR), a framework built on two key ideas: (i) modality-specific embeddings from different modalities are aligned in a shared latent space, and (ii) a unified representation is constructed by fusing only the embeddings of the modalities that are actually available at both training and inference time. Rather than imputing missing modalities or requiring a fixed modality set, LWR treats each modality as a partial perception of an underlying latent state and performs availability-aware representation learning directly from the observed modalities. This combination of neighbor-based latent alignment and availability-aware modality fusion enables robust multimodal prediction under partial observation, while avoiding error propagation from explicit reconstruction of missing modalities. We evaluate the proposed framework on real-world incomplete multi-omics benchmarks and demonstrate that it provides an effective approach to downstream tasks such as cancer phenotype classification and survival prediction.

Language-model post-training is the main stage at which model behavior is shaped, yet it still largely involves optimization of scalar rewards that summarize diverse desiderata. This abstraction gives practitioners little visibility into what their data actually teaches models, allowing spurious correlations to be learned by a model and inducing undesirable behaviors such as over-stylization and sycophancy. To address this problem, we ask: can we inspect a preference dataset before optimization and decide, at the level of concepts, which behaviors a model should be allowed to learn? Motivated by this, we introduce a data-centric post-training pipeline that uses interpretability protocols to develop statistical hypotheses for the latent concepts separating preferred from dispreferred generations, making them explicit for fine-grained user feedback. Building on this view, we unify several interpretability-based training protocols as ways of shaping rewards via feature or data interventions. Empirically, we show that our pipeline diagnoses undesirable signals in existing preference data, mitigates off-target learning, and can also help amplify or shape desired properties such as safeguards and model personality. More broadly, our results suggest that interpretability can turn post-training from optimizing opaque proxy rewards into a process of auditing and sculpting the learning signal itself.

Multi-robot collaboration allows robots to efficiently take on a wide range of tasks, from moving a couch through a doorway to assembling structures on a construction site. However, achieving such coordination in mobile multi-robot settings remains challenging: centralized methods conditioned on the combined observations of a team scale poorly with team size, and decentralized methods that train one policy per robot often require explicit alignment procedures or information sharing at inference time to overcome partial observability. Our key insight is that the visuomotor priors of pretrained vision-language-action (VLA) models should enable reactive, decentralized collaboration from each robot's local observations alone, without these inference-time assumptions. We propose CHORUS, a framework that adapts a single VLA backbone to control diverse, multi-robot teams. At inference time, each robot runs an independent copy of CHORUS, conditioned only on its own observations and a robot-identifying prompt. In real-world experiments including mobile tape measurement, library book handovers, and laundry basket lifting, CHORUS achieves a 64% point improvement over decentralized, from-scratch models, improves reactivity to teammate behavior by 40% points, and outperforms centralized baselines. Together, these results show that a shared VLA backbone is capable of achieving decentralized multi-robot collaboration, without per-robot policies or inter-robot communication at inference.

Offline preference optimization has become a practical substitute for reinforcement learning from human feedback, but pairwise objectives such as Direct Preference Optimization (DPO) and its variants use only the chosen and rejected responses stored in a static dataset. This leaves a useful signal unused: the response that the reference model itself would generate for the same prompt. We propose Direct Preference Optimization with Penalization (DPOP), a simple extension of DPO that augments the base preference loss with a gated penalty on reference-greedy responses. DPOP activates this penalty only when the current policy still assigns a lower likelihood to the preferred response than to the rejected response. On AlpacaEval 2.0, DPOP improves length-controlled win rate over DPO, SimPO, and AlphaDPO on both Llama-3-8b-it and Gemma-2-9b-it, achieving relative gains of 5.3\% and 4.4\% over baselines on the two models, respectively. Ablations further show that a SimNPO-style length-normalized penalty is stronger than NPO and token-level unlikelihood in this setting.

The rapid proliferation of large language models (LLMs) raises critical questions about human creativity and individual expression in an era of AI-assisted creation. When do humans adopt AI suggestions, and what are the implications for individual voice? This study examines these questions through a gamified writing exercise where 74 participants (214 responses) replied to prompts while AI-generated word suggestions were available as they wrote. The game simulates a dystopian future in which an AI is attempting to learn from what remains of human individuality, and disincentivizes AI-like writing. In doing so, it attempts to create conditions that reveal authentic user preferences rather than default behaviors, such as accepting a readily available AI-generated suggestion. Note that this is a deliberate inversion of the "helpful assistant" design pattern; the system is explicitly forbidding you from accepting AI suggestions. We analyze user behavior patterns across different task types, user behaviors, and response characteristics to understand the factors influencing human-AI interaction in creative tasks. The study focuses on when users choose to maintain creative autonomy versus violating the rules of the game and accepting AI assistance. It also explores how these choices relate to response patterns, task characteristics, and user behavior. This gamified approach offers both a framework for studying authentic human-AI interaction and a provocative lens for understanding the tension between efficiency and authenticity in AI-augmented creativity.

Hematoxylin and eosin (H&E) staining is the cornerstone of histopathology, yet scalable, quantitative analysis of H&E whole-slide images (WSIs) remains a central challenge in computational pathology. We present Atlas H&E-TME, an AI-based system built on the Atlas family of pathology foundation models that predicts tissue quality, tissue region, and cell type labels across multiple cancer types, yielding over 4,500 quantitative readouts per slide at cell-level resolution. A key challenge to validating such systems is overcoming morphological ambiguity inherent to H&E-only ground truth and the limited scalability of more informed references drawing on modalities such as immunohistochemistry (IHC). We address this with a dual validation framework combining biologically grounded depth with technical and morphological breadth. For depth, we propose an IHC-informed multi-pathologist consensus protocol that substantially improves inter-rater agreement over conventional H&E-only annotation. This yields a molecularly grounded reference against which we compare Atlas H&E-TME and pathologists working from H&E alone. For breadth, we benchmark Atlas H&E-TME on over 200,000 high-confidence H&E-only pathologist annotations across 1,500+ cases spanning eight cancer types and their most common metastatic sites, with subtypes covering >90% of clinical cases per cancer type, drawn from 25+ sources and 8+ scanner models. Benchmarked against the IHC-informed consensus, Atlas H&E-TME matches or exceeds pathologist H&E-only performance and generalizes consistently and robustly across this broad morphological and technical scope. In doing so, Atlas H&E-TME turns the H&E slide -- the most ubiquitous data in pathology -- into a scalable, quantitative window into the tumor and its microenvironment, laying a foundation for the next generation of tissue-based biomarkers in translational and clinical research.

General-purpose agents such as OpenClaw are increasingly used as autonomous tool users, but their coding ability is difficult to measure under SWE-bench: a generic agent does not by itself satisfy the clean Docker workspace, patch, and prediction contract required for scoring. We introduce Claw-SWE-Bench, a multilingual SWE-bench-style benchmark and adapter protocol that makes heterogeneous agent harnesses, or claws, comparable under fair settings including a fixed prompt, runtime budget, workspace contract, patch extraction procedure, and evaluator. The full benchmark contains 350 GitHub issue-resolution instances across 8 languages and 43 repositories, drawn from SWE-bench-Multilingual and SWE-bench-Verified-Mini after future-commit cleanup. We also release Claw-SWE-Bench Lite for faster validation, which is an 80-instance subset selected by a cost-aware, rank-aware procedure over 17 calibration columns. On the full benchmark, OpenClaw with a minimal direct-diff adapter scores only $19.1\%$ Pass@1, whereas the full adapter reaches $73.4\%$ with the same GLM 5.1 backbone, showing that adapter design is essential for enabling OpenClaw-style harnesses to perform coding tasks effectively. Across an OpenClaw $\times$ nine-model sweep and a five-claw $\times$ two-model sweep, model choice changes Pass@1 by $29.4$ pp and harness choice by $27.4$ pp under fixed models; systems with similar accuracy can differ substantially in total API cost. Claw-SWE-Bench therefore treats harness and cost accounting as first-class axes of SWE-style coding-agent evaluation, providing both a full benchmark and a low-cost reference set for reproducible comparison. The data is available at https://github.com/opensquilla/claw-swe-bench and https://huggingface.co/datasets/TokenRhythm/Claw-SWE-Bench.

Modern Artificial Intelligence (AI) workloads deployed across the heterogeneous tiers of an edge--cloud continuum must satisfy multi-dimensional Service Level Objectives (SLOs) over latency, throughput, and output quality. For each incoming task, the scheduler picks both a target node and a processing mode (e.g., full or reduced inference precision). We call this class of problems \emph{Continuous Multi-Mode Scheduling} (CMMS). Comparing CMMS algorithms fairly is difficult because prior studies typically evaluate each controller in its own stack, under a single workload, and without reporting per-decision overhead. To close these gaps, we present an open source benchmark platform that features (i) a unified controller interface, (ii) a closed-loop workload driver covering multiple workload patterns, and (iii) dual-metric SLO scoring that reports raw SLO (overall compliance) and steady-state SLO (compliance during stable operation) separately. Running six controllers across five cluster configurations and two load regimes (424 episodes), we find that controller rankings are strongly configuration-dependent: a deep reinforcement-learning winner under light workloads loses to a rule-based heuristic by nearly 29 percentage points once load intensifies, at roughly 500$\times$ the per-decision operational overhead. We further show that separating raw from steady-state SLOs exposes switching costs that a single aggregate score would otherwise conflate.

Domain fine-tuning degrades the safety of large language models: fine-tuned specialists readily comply with harmful prompts framed in domain language. Existing inference-time defenses that mix logits from a safe anchor model require both models to share a vocabulary, which rules them out for the cross-family specialists where safety is most degraded. We present ALIGNBEAM, a training-free method that lifts this restriction by translating anchor logits into the target model's vocabulary token-by-token at each decoding step; a small LLM judge then selects the safest among K candidate continuations. No weights are changed, and the safety-utility trade-off can be tuned at deployment without retraining. Across both cross-vocabulary and same-vocabulary evaluation pairs, ALIGNBEAM substantially raises refusal on adversarial benchmarks while keeping task accuracy and inference overhead within practical bounds. The results show that safety alignment can be transferred between model families at inference time, without touching either model's weights.

Inverse problems governed by partial differential equations (PDEs) are central to computational mechanics and are commonly solved by adjoint-based optimization, while physics-informed neural networks (PINNs) have emerged as a flexible alternative. Their relative performance remains difficult to assess because the two approaches are often compared under different formulations, parameterizations, optimizers, and regularization choices. We present a fair comparison of adjoint optimization and PINNs for PDE-constrained inverse problems. From a common abstract formulation, we instantiate both methods on identical domains, governing equations, observation models, and regularization terms, while matching the optimizer, unknown parameterization, and arithmetic precision wherever applicable. The benchmarks include unsteady Burgers, noisy Darcy permeability inversion, three-dimensional Allen--Cahn reaction identification, and unsteady Navier--Stokes viscosity identification. The results show that the representation of the unknown largely determines the preferred method: grid-based fields favor the discrete adjoint, whereas neural representations are native to PINNs and relevant for closure and constitutive modeling. For time-dependent problems, adjoint inversion can be dominated by trajectory storage and differentiation, while PINNs provide satisfactory reconstructions at lower cost. A PINN-warm-started adjoint strategy then recovers adjoint-level accuracy at substantially reduced cost.

High-precision robotic manipulation requires fine-grained spatial reasoning that is often difficult to achieve with RGB-only policies due to depth ambiguity and perspective scale issues. Policies that leverage 3D information directly, such as those based on point clouds, offer a stronger geometric prior over purely image-based ones, yet their performance remains highly task-dependent. We hypothesize that this discrepancy may be due to the spectral bias of neural networks towards learning low frequency functions, which especially affects architectures conditioned on slow-moving Cartesian features. We thus propose to map point clouds from Cartesian space into high-dimensional Fourier space, effectively equipping the point cloud encoder with direct access to high-frequency features. We experimentally validate the use of Fourier features on challenging manipulation tasks from the RoboCasa and ManiSkill3 benchmarks and on a real robot setup. Despite their simplicity, we find that Fourier features provide significant benefits across diverse encoder architectures and benchmarks and are robust across hyperparameters. Our results indicate that Fourier features let policies leverage geometric details more effectively than Cartesian features, showing their potential as a general-purpose tool for point cloud-based imitation learning. We provide source code and videos on our project page: https://fourier-il.github.io/fourier-il

Evaluating multi-turn dialogue is challenging because quality emerges across turns rather than within individual responses. We focus on a key dimension of information-seeking dialogue: semantic progress, defined as the accumulation of new, question-relevant, and non-redundant information over the course of a conversation. We formalize semantic progress as question-conditioned uncertainty reduction and introduce an information-theoretic metric that approximates it in embedding space. Our main estimator uses a tractable Gaussian formulation with closed-form updates, while a complementary maximum-entropy argument shows why log-determinant structure arises more broadly when only second-order embedding information is retained. This formulation yields desirable theoretical properties, including monotonicity, additive decomposition of total information gain across turns, and diminishing returns for redundant evidence. Unlike LLM-as-a-judge approaches, our metric requires no autoregressive inference at evaluation time and is fully reproducible for a fixed embedding model. Experiments on MT-Bench, Chatbot Arena, and UltraFeedback show that the proposed metric achieves competitive agreement with human judgments despite targeting only semantic progress, with improved alignment on MT-Bench and UltraFeedback compared to several LLM-based judges. Notably, the method remains effective with lightweight embedding models under CPU-only execution, indicating that semantic progress can be captured without reliance on large model capacity.

AI coding assistants now support a growing share of software work, from quick scripts to production applications. Yet these agents remain largely stateless: each new session re-reads project files, re-derives prior decisions, and - most costly - may repeat debugging attempts that already failed. Reconstructing this context can consume an estimated 5,000-20,000 tokens per session; the bottleneck is often not model capability but missing project memory. We present projectmem, an open-source, local-first memory and judgment layer for AI coding agents. projectmem records development as an append-only, plain-text event log of typed events - issues, attempts, fixes, decisions, and notes - and deterministically projects that log into compact, AI-readable summaries served through the Model Context Protocol (MCP). Beyond storage, projectmem adds a deterministic pre-action gate that warns an agent before it repeats a previously failed fix or edits a known-fragile file. We frame this as Memory-as-Governance: memory that does not merely answer the agent but acts on its next action. The system runs fully offline with no telemetry; its immutable log also serves as a provenance trail for reproducible, auditable AI-assisted development. projectmem ships as a three-dependency Python package (14 MCP tools, 19 CLI commands, 37 automated tests) and is evaluated through a two-month self-study across 10 projects comprising 207 logged events. Source code: https://github.com/riponcm/projectmem.

Enterprise security was built to govern data boundaries: the protected surface was data at rest and in transit, and the controls -- access control, data-loss prevention, perimeter inspection -- governed crossings of that boundary. Production AI agents dissolve this assumption. An agent reads context, calls tools, invokes connectors, and modifies systems of record on an enterprise's behalf, so risk moves inside the workflow, into sequences of individually-permitted actions that may transform a business process no one authorized. Existing policy engines do not extend to this regime: they evaluate request-time decisions against atomic principals, where agentic systems require stateful evaluation against composite principals whose authority attenuates through delegation chains. We present a reference architecture for the runtime governance of production agents, built from four composable primitives: a five-plane decomposition (a reasoning plane that adjudicates intent, and four enforcement planes -- network, identity, endpoint, data -- that realize the decision), stop-anywhere mediation, composite principals with capability attenuation, and audit as a structured evidence substrate. We define a taxonomy of six interruption primitives that generalize allow and deny, state and argue for four correctness invariants, and demonstrate the foreclosure of seven production-agent threats across five concrete workflows. A reference implementation of the policy-engine core supplies measured evidence: attenuation correctness and evidence reconstructability hold on every trial, adjudication runs in single-digit microseconds, and the audit substrate's tamper-evidence behaves exactly as designed. We are explicit about scope: the architecture governs delegated action, not model behavior, and a full-system evaluation against a live agent benchmark is the invited next step.

Neural operators approximate mappings between function spaces, but often generalize poorly to other operators and usually require fine-tuning or retraining. In-Context Operator Networks (ICON) addresses this issue by prompting the model with numerical context so that the model learns specific operators from prompts and adapt to different operators without fine-tuning. However, ICON may still fail to generalize to out-of-distribution (OOD) operator tasks. Inpired by the success of harness engineering of Large Language models (LLMs), we introduce Chain of Operators (CHOP), a framework that harness a frozen ICON to OOD operator tasks without updating its parameters. Specifically, CHOP constructs a chain of operators consisting of explicit elementary transformations and the frozen ICON. Experiments on a scalar conservation law and a mean-field control problem show that CHOP reduces relative inference error over direct ICON evaluation, while each operator in the chain remains interpretable and in closed form. A chain constructed on one PDE family further generalizes to a different family, indicating shared mechanisms across harness systems.

Self-supervised learning (SSL) has opened new opportunities in bioacoustics by enabling scalable modeling of animal vocalizations without the need for expensive manual annotation. However, current SSL models in this domain prioritize broad generalization across species and are not optimized for uncovering the fine-grained structure of individual communication systems. In this work, we collect and release a novel dataset of over five years of longitudinal recordings, from five known dolphins in a semi-naturalistic marine environment, an unprecedented resource for studying dolphin communication. We adapt the Wav2Vec2.0 Baevski et al. (2020) architecture to this domain and introduce Dolph2Vec, the first large-scale, species-specific SSL model trained exclusively on this data. We benchmark our model on two biologically relevant tasks: signature whistle classification and whistle detection. Dolph2Vec significantly outperforms general-purpose baselines in both tasks. Beyond performance, we show that learned embeddings and codebook structure capture interpretable acoustic units aligned with dolphin whistle categories and possibly sub-whistle structure, enabling fine-grained analysis of communication patterns. Our findings demonstrate how SSL can serve as both a model and a scientific tool to explore hypotheses in animal communication research.

Temporal grounding--returning the interval $[t_s, t_e]$ for a natural-language query over a video--is the language interface to long-form video, yet has been studied on short videos; the dynamics of hour-scale natural-language grounding remain underexplored. We take the position that at hour-scale, the binding constraint is search, not recognition: Video-LLMs are bottlenecked not by localizing a nearby event, but--given a natural-language query--by searching for the relevant region of a long video. To test this, we release ExtremeWhenBench, the first open hour-scale grounding benchmark (2,273 queries over 194 videos, mean 75.7 min, max 9 hr) with an open-form query distribution. Every open Video-LLM collapses while a frame-level retrieval baseline outperforms them; a failure taxonomy attributes 85% of failures to search; and a retrieve-then-ground hybrid recovers 6.7x over the monolithic Video-LLM--mirroring retrieve-then-read in open-domain QA.

Vision-Language-Action (VLA) models provide a natural language interface to robot control, but the mapping from language to behavior is often brittle and unintuitive: semantically similar instructions can induce drastically different behaviors, while some capabilities may not be elicitable through prompting alone. As a result, both human instructions and zero-shot language models can fail to reliably steer VLAs toward successful task execution. In this work, we propose a framework that interactively searches for language sequences that improve closed-loop VLA task performance, distills these sequences into a test-time language feedback policy (LFP), and learns an improvement head that predicts when language steering will improve performance. We conformalize this improvement head to prevent harmful steering interventions, where the LFP decreases task performance relative to the original instruction on out-of-distribution scenarios. Crucially, our approach operates on arbitrary frozen pre-trained VLAs, requiring neither access to the original training distribution nor fine-tuning of the underlying model. On seen environments, our conformalized LFP improves base VLA performance by 24.7% in simulation and 65.0% in hardware. On visual and semantic perturbations, our conformalized LFP has strong harmlessness guarantees, and produces recovery behaviors not observed with open-loop prompting.

This overview paper presents the results of the shared task for the second workshop on Multimodal Augmented Generation via Multimodal Retrieval (MAGMaR). In this shared task participants submitted systems focused on either (i) video retrieval or (ii) grounded generation of articles given retrieved videos. Teams could submit to either task. For the retrieval task, we had 2 participating teams that submitted a total of 17 systems -- all of which beat a baseline derived from the winner of last year's shared task. On the generation side, we had 4 teams submit 16 systems. All teams had at least one generated report that was labeled the best by a human annotator.

Large language models (LLMs) now reach expert-level scores on medical licensing exams, encouraging the assumption that high scores imply safe medical judgment while patients increasingly use them for health advice. We show this assumption is fragile: when misleading context is injected into questions that LLMs originally answer correctly, they abandon the correct answer. We call the ability to maintain correct judgment under adversarial context epistemic resilience, and introduce MedMisBench to measure it. MedMisBench contains 10,932 medical question items and 48,889 misleading context-option pairs spanning medical reasoning, agentic capability, and patient-journey evaluation. Across 11 model configurations, mean accuracy falls from 71.1% on original questions to 38.0% under focused misleading context, with 51.5% attack success. The most damaging injections are formal, rule-like fabrications: authority-framed falsehoods reach 69.5% attack success and exception-poisoning claims reach 64.1%. A 14-member clinical panel from 7 countries identified serious potential harm in 38.2% of reviewed cases. MedMisBench exposes a structural blind spot in LLM evaluation in medical settings: existing benchmarks measure what models know, but not whether they preserve correct medical judgment under misleading context.

As Artificial Intelligence models grow in complexity, interpretability has become an indispensable tool for understanding, debugging, and controlling their computations. However, interpretability lacks general theories to deductively design interpretable methods. This gap between theories and methods results in a fragmented literature and inconsistent evaluation protocols. To fill this gap, we introduce the Standard Interpretable Model (SIM), a general theory grounded in Lagrangian mechanics that enables the deductive design of interpretable methods. Specifically, the SIM summarises, in a set of premises, what interpretability is for a target user. From these premises, the SIM systematically derives interpretability symmetries and corresponding constraints, which shape the landscape of a Lagrangian whose minima correspond to optimal interpretable models. To reach the minima, one can either update the parameter values of an opaque model to make it more interpretable or compile constraints into an interpretable architecture. We empirically show that the SIM identifies and solves limitations of existing methods (including traditional, concept-based, and mechanistic interpretability), highlights underexplored research directions, and informs the design of core programming interfaces. Beyond being a research method, the deductive nature of the SIM offers pedagogical grounding for interpretability curricula and may shift the scientific community's perspective of a discipline that has long been fragmented.

The Transformer architecture is widely regarded as the most powerful tool for natural language processing, but due to a high number of complex operations, it inherently faces the issue of high energy consumption. To address this issue, we consider Spiking Neural Networks (SNNs), which are an energy-efficient alternative to conventional Artificial Neural Networks (ANNs) due to their naturally event-driven approach to processing information. However, this inherently makes them difficult to train. Often, many SNN-based models circumvent this issue by converting pre-trained ANNs. More recently, attempts have been made to design directly trainable SNN-based adaptations of the Transformer model structure. Although the results showed great promise, the application field was computer vision. Moreover, the proposed model incorporates only encoder blocks. In this paper, we propose SpikeDecoder, a fully SNN-based implementation of the Transformer decoder block, for applications in natural language processing. In a series of experiments, we analyze the impact of exchanging different blocks of the ANN model with spike-based alternatives to identify trade-offs and significant sources of performance loss. We further investigate the role of residual connections and the selection of SNN-compatible normalization techniques. Besides the work on the model architecture, we formulate and compare different embedding methods to project text data into spikes. Finally, we demonstrate that our proposed SNN-based decoder block reduces the theoretical energy consumption by 87% to 93% compared to the ANN baseline.

Expressive performance rendering (EPR) aims to generate realistic performances constrained on sequences of notes. However, flow matching audio editing models manipulate only synchronized music samples of the same duration, limiting their understanding of expressive timing. We introduce PianoKontext, a flow matching rendering model for classical piano music that generates variable-length performances in the latent space of a pretrained Music2Latent model. We synthesize MIDI scores into deadpan audio and employ Dynamic Time Warping (DTW) in the latent space to construct paired data for training. The aligned embeddings are concatenated in DiT blocks, allowing for a simple and effective learning of the dependencies between the score and performances. Audio samples are available at our demo page: https://realfolkcode.github.io/pianokontext_demo/.

In Decentralized Training and Decentralized Execution (DTDE) for cooperative Multi-Agent Reinforcement Learning (MARL), action-advising-based knowledge sharing promotes interpretable and scalable cooperation among agents. However, current action advising approaches often adhere too much to the teacher's guidance without evaluating teacher-student compatibility, which causes excessive advising, suboptimal stability, and degraded performance. To overcome these challenges, this paper presents a Consensus-based Communication and Knowledge Sharing (CCKS) framework, which allows agents to adopt recommendations based on consensus-derived constraints and to follow the teacher's instructions more smartly. This mechanism enables agents to balance exploration and learning from experienced teachers, improving overall performance. The key is the consensus model construction, for which we propose to employ contrastive learning to construct consensus models based on local observations in the agents' training phase. In action selection, agents score and choose actions based on consensus and shared knowledge. Designed as a plug-and-play solution, CCKS integrates seamlessly with existing DTDE algorithms. Experiments conducted in the Google Research Football environment and the complex StarCraft II Multi-Agent Challenge demonstrate that the integration with CCKS significantly improves cooperation efficiency, learning speed, and overall performance compared with current DTDE baselines. The code is available at https://github.com/yuanxpy/CCKS.

Post-training quantization lets large text-to-image diffusion transformers run on consumer GPUs, yet the hardware-specific trade-offs are seldom measured directly. We quantize Ideogram 4.0 - a 9.3B flow-matching diffusion transformer (DiT), shipped as two separate-weight copies of a single-stream 34-layer backbone for classifier-free guidance and conditioned by a Qwen3-VL-8B encoder - for Ampere RTX 3090 GPUs, which lack FP8 tensor cores. Our INT8 W8A8 recipe (per-channel weights, per-token dynamic activations, SmoothQuant, and mixed-precision protection of a small high-fragility layer set) holds the FP8 quality ceiling: on a 200-prompt benchmark the paired same-seed bootstrap CI for INT8-FP8 includes zero on both Pick and CLIP, while INT8 improves on NF4 by $+1.9$ CLIP (95% CI $[+1.21,+2.64]$, excluding zero). A per-category OCR analysis, to our knowledge unreported for this model class, confirms text legibility is preserved, and an ablation isolates protection of the FFN down-projections as the dominant quality lever. Our GGUF Q4_K quantization beats NF4 at equal on-disk size and is the Pareto winner on the quality-memory frontier, with paired confidence intervals excluding zero (Q8_0 is quality neutral). Finally, we characterize where 8-bit quantization helps and where it does not: INT8's weights match FP8's footprint rather than shrink it, so a speed gain on Ampere awaits a fused INT8 kernel.

Recent work in ML applies genetic algorithms at inference time to iteratively improve solutions to optimization problems. The basic mutation and recombination operators involved are qualitatively different from those studied classically. Mutations are no longer random; an ML algorithm mutates a solution with the goal of improving an objective. Similarly, recombination is not based on random collages of parent solutions. Instead, it is an ML optimization-based operator whose goal is to synthesize improved solutions from its inputs. Thus, these mutation and recombination operators are more likely to improve the objective, but their computational cost is much higher. We introduce a general model of genetic algorithms and formulating optimization in this model as a query-complexity problem, using the language of reinforcement learning. We then study specialized models. We show that some optimization problems require generation, mutation, and recombination to be solved. We then obtain qualitatively tight algorithms for a family of problems within this framework that captures the nontrivial role of diversity in the solution pool, a key feature of practical ML genetic algorithms.

Neural network pruning reduces model size by removing less important parameters while aiming to preserve predictive performance. Although the Lottery Ticket Hypothesis (LTH) shows that sparse subnetworks can match dense networks when trained from suitable initializations, its iterative pruning procedure requires multiple complete training cycles. This work evaluates progressive magnitude-based pruning as a single-cycle alternative. The method gradually increases sparsity during training using a linear schedule and updates pruning masks based on active weight magnitudes. We conduct systematic experiments on CIFAR-10 and MNIST across ResNet, VGG-style, and LeNet architectures, comparing the proposed method with representative iterative and initialization-based pruning baselines, including LTH, SNIP, and GraSP. On CIFAR-10, the method achieves 95.12\% accuracy on ResNet-18 at 72.9\% sparsity, compared with 90.5\% reported for LTH. At extreme sparsity, it achieves 93.13\% accuracy on a VGG-like architecture at 97\% sparsity, compared with approximately 92.0\% for SNIP, and 93.44\% accuracy on VGG-19 at 97.97\% sparsity, compared with 92.19\% for GraSP at 98\% sparsity. A sparsity-accuracy analysis on ResNet-18 further shows that accuracy remains within 0.1 percentage points of the dense baseline across 70--85\% sparsity. These results indicate that progressive magnitude-based pruning provides an effective single-cycle approach for neural network sparsification under the evaluated settings.

In this paper, we propose an approach to finding sets of similar-performing models (in terms of loss/accuracy measurements) with highly different context-aware characteristics. Through experiments on the METABRIC dataset, we show that the proposed method finds multiple models with highly different gene expressions than those found by the control methodology without performance penalties. We argue that the proposed methodology is important whenever one aims to analyze any global characteristic of a model to extract insight into the underlying phenomenon being studied.

Diffusion large language models (dLLMs) offer an efficient alternative to autoregressive models through parallel decoding, yet existing post-training methods largely rely on random masking strategies that overlook intrinsic token dependencies. In this work, we present an empirical analysis of attention in dLLMs and show that tokens attending more strongly to unmasked context exhibit greater generation stability and play a critical role in reasoning. Motivated by these findings, we propose AGDO, an attention-guided denoising and optimization framework that aligns both training and optimization with attention-derived dependencies. AGDO determines the denoising order based on attention structure and emphasizes attention-critical tokens during supervised fine-tuning and reinforcement learning. Experiments on mathematical and coding benchmarks demonstrate that AGDO consistently improves reasoning performance, outperforming state-of-the-art post-training methods for dLLMs.

Advanced AI systems have extensive knowledge of their environments; in fact, their knowledge may (far) exceed that of their developers or users. Consequently, a desirable property for an AI system is that it is honest -- that it accurately reports its beliefs about the world. Designing an AI system to be honest may be difficult, especially if we want to ask it questions about latent variables in the environment -- variables which are hidden from the human interacting with it. This gives rise to the problem of eliciting latent knowledge (ELK): the problem of training an AI agent to honestly report its beliefs. In this paper, we make ELK formally precise using Causal Influence Diagrams (CIDs). CIDs can be used to describe the relationship between an agent's training environment and its subjective representation of the world. We use CIDs to formalise the distinction between observable and latent variables, to specify what exactly it means for an agent to be honest, and to formally define goal misgeneralisation. We show that, under certain circumstances, developers can incentivise an agent to honestly answer questions by providing correct feedback during training. However, a natural, but undesirable, way for an agent to generalise is to provide answers which humans would evaluate as true, rather than honest answers. We prove an impossibility theorem stating: There is no feedback-based training strategy that depends only on agent behaviour and with certainty produces an honest agent, even if feedback is perfect during training.

We propose that value -- the quantity goal-directed agents create, destroy, and exchange -- is a lawful structural quantity in the same category as information. Following Shannon's method, we make one ruthless abstraction: value is the rate at which an agent converts a resource into goal-progress, relative to a frame fixed by its goal. A scale-invariance axiom forces a logarithmic measure, $V=\sum_i k_i \ln e_i$; compounding of a reinvested resource forces the same form via the ergodicity argument of Peters (2019). The two routes are kin rather than independent; their agreement is a consistency check, not an over-determination. We derive a coding theorem of value: $ΔG \le I(X;Y)$, achieved by Bayes-proportional allocation; realized value decomposes as $G=D(q\|r)-D(q\|p)$, identifying misalignment with measurable waste. For populations, value is frame-relative while price is frame-independent; a fleet that pools its resource and fuses its perception inherits the ceiling $G_{\mathrm{fleet}} \le I(X;Y_{1:m}) \le H(X)$ (a corollary; an earlier sum-form claim was wrong and is corrected in v5). A dynamical layer yields an is/ought asymmetry from which alignment emerges as a control-stability condition with a closed-form residual. We test the single-frame laws on live language models in a pre-registered scale-up: perception mutual information tracks realized capability rather than parameter count (Spearman $ρ= 0.977$ pooled over 30 model$\times$domain points), out-of-sample $ΔG$ tracks $I(X;Y)$, and over-confidence is measurable dissipation; a further pre-registered test shows the bridge is shape-invariant across four task shapes ($n=42$, slope 0.953). None of the mechanisms is individually new -- generalized Kelly, Armstrong & Mindermann (2018), classical control; the contribution is their unification and the governance mapping (incentive design over oversight) that follows.

We propose Conformal QoT, a policy-driven framework that combines statistically guaranteed QoT estimation with operational decision policies, enabling reliable lightpath-feasibility predictions under domain shift and improving accuracy from 92\% to 99.6\% on open datasets.

How can we design a market of human-generated content for use in training AI models that both enables technological progress and preserves individual incentives for high-quality content creation? Existing approaches take polar positions: a "free-for-all" model based on fair use and a "strong intellectual property rights" model. We show that both fail: Free-for-all does not compensate creators, and -- by modeling as a static Stackelberg game -- strong intellectual property rights also underpower creative incentives. We find this especially true for more innovative creators, a phenomenon we term the "originality penalty." Extending this insight to a dynamic model, we find another market failure undermining AI model performance, even for an initially good model: Such a model induces greater reliance by humans on AI-assisted creation, resulting in homogenized content feeding back into training, which degrades the model performance -- a "curse of precision." We further propose a market design with a data intermediary internalizing cross-creator externalities and subsidizing innovative contributions, thereby restoring efficiency.

Cache partitioning is among the strongest structural defenses against eviction-based cache side channels, yet a decade-old design issue has blocked its widespread deployment in secure shared-OS settings. The issue is that write-shared coherence collapses under strict partitioning. We present SCP (Secure and Coherent Partitioning), which combines strict eviction isolation with write-shared coherence by partitioning only the tags, sharing a single data pool, and sizing the data pool so capacity-driven cross-partition eviction cannot occur. Timing obfuscation extends protections to the inter-partition lookup path. Coherence-based leakage on shared-writeable lines is mitigated by routing those writes through to the LLC once a leakage threshold is crossed, which makes attacker write probe latency independent of victim activity. Using gem5 for implementation, SCP mitigates Prime+Probe and Flush+Reload, which are the basis for more sophisticated cache attacks. We also demonstrate that a shared-writeable-line attack is mitigated. All these attacks yield results no better than random guessing. SCP's hardware cost is a modest +2.8% LLC SRAM. Performance matches DAWG within 0.3% IPC on the SPEC CPU2017 benchmarks that we evaluated. Sharing-intensive microbenchmarks demonstrate a tunable security-performance tradeoff based on a system-specified leakage threshold.

Training deep neural networks for clinical time-series analysis is computationally demanding, yet many healthcare settings lack the resources required for repeated model development and deployment. This challenge is particularly evident in electrocardiogram classification, where large datasets and long training schedules make efficiency practically important. Progressive Data Dropout reduces training cost by excluding samples from gradient updates once they are learned, but it relies on model confidence and may retain samples that are difficult due to noise or ambiguity rather than useful signal. In this work, we introduce ERTS, an explainability-based reliability training signal for efficient ECG classification. ERTS uses explanation quality during training to distinguish between informative and unreliable uncertainty. Building on progressive data selection, we compute Grad-CAM attention maps for candidate samples and derive a focus score that measures whether model predictions are supported by coherent and localised patterns. Samples with low focus are filtered out, while those with meaningful attention are prioritised for gradient updates. We evaluate ERTS across three ECG datasets and multiple backbone architectures, showing consistent improvements in macro-F1 alongside reduced effective training cost. These results suggest that explanation quality can serve as a practical signal for improving both efficiency and reliability in clinical time-series learning. Code will be released.

Gradient-based adversarial attacks remain a dominant threat to deep neural networks (DNNs), as they exploit gradient information to efficiently optimize adversarial perturbations. To address this, we investigate whether reinforcement learning (RL) training can disrupt the gradient structure used by attackers by training image classifiers with policy-gradient objectives and epsilon-greedy exploration. Through systematic experiments across CIFAR-10, CIFAR-100, and ImageNet-100 with multiple architectures, we find that RL-trained classifiers significantly disrupt gradient-based adversarial optimization. To explain this, we conduct a comprehensive mechanism analysis using loss landscape visualization, static and dynamic gradient indicators, and predictive entropy. Our analysis reveals that RL acts as an implicit regularizer, producing models with highly unstable gradient directions and smaller gradient magnitudes. This combination makes each PGD step both unreliable in direction and limited in magnitude, causing gradient-based attacks to fail within practical iteration budgets. We further show that combining RL with adversarial training (RL-adv) provides a dual-layer defense operating at two complementary levels: RL degrades gradient information available to attackers (gradient-level defense), while adversarial training strengthens decision boundaries (boundary-level defense). RL-adv achieves the highest robustness across all major attack types evaluated, including gradient-based (PGD, AutoAttack), transfer-based, and query-based attacks, outperforming SL-adv by a significant margin. These findings identify RL-induced gradient disruption as a complementary robustness mechanism and motivate future research on hybrid SL-RL training schedules that combine SL's efficiency with RL's gradient-regularization properties.

Large language models (LLMs) in medicine are mainly evaluated using multiple-choice question answering (MCQA), which can overestimate real clinical ability due to guessing strategies and answer biases. To address these limitations, we introduce an expanded and more challenging benchmark based on Polish medical exams, adding over 15,000 questions, two new domains, and four structural modifications that reduce MCQA-specific artifacts and better test reasoning. We evaluate 21 LLMs and show that evaluation design strongly affects results. Under our harder setup, the best model (Qwen3.5-122B) drops by 28.4 and 31 pp on English and Polish exams, respectively. Despite low evidence of data contamination, standard MCQA scores do not reliably reflect true medical competence. To facilitate further research, we make our benchmark publicly available.

Research on bias in large language models (LLMs) has predominantly focused on third-person audits, which study how models represent or evaluate demographic groups as external subjects. However, this paradigm overlooks a structural blind spot because the user is absent from the audit. In practice, LLMs are used in open-ended, personal interactions, during which the model implicitly represents the user and adjusts its responses accordingly. When identical requests yield different responses depending on who is asking, bias manifests not in how the model describes others but in how it treats its interlocutor. We propose Situated Interaction Auditing (SIA), a user-centered framework for studying how user profile signals -- implicit sociodemographic markers, writing style, and stated identity -- systematically shape LLM response quality, content, and tone. We demonstrate the framework through a case study that intersects gender and socioeconomic status signals across multiple task domains and outline a research agenda for SIA as a new mission for natural language processing.

In Online Learning to Rank (OLTR), ranking models are trained directly from live user interactions, but existing systems rely on a trusted central server to collect and process these interactions. This leaves operators free to introduce biases that conflict with user interests. Decentralized learning offers an attractive alternative, allowing users to collaboratively train a shared ranking model by exchanging model updates directly with one another, without any central authority. In such settings, however, malicious nodes can send poisoned model updates that degrade the ranking quality of honest nodes. We introduce RankGuard, a decentralized OLTR framework in which users collaboratively train ranking models and exchange model updates directly with other nodes. RankGuard defends against poisoning attacks by carefully evaluating incoming models against the user's own private click history, corrected for position bias. An incoming model is only aggregated if it better explains the user's past interactions than the current local model, making it fundamentally hard for malicious nodes to craft updates that pass this test without also genuinely helping the user. We derive a theoretical convergence guarantee of RankGuard. To the best of our knowledge, this is the first formal convergence analysis of a decentralized OLTR algorithm. We evaluate RankGuard against four poisoning attacks, including a powerful adaptive attack, using four standard benchmarks and three click models. RankGuard outperforms all baselines in most settings while being up to 62x more efficient than its closest competitors.

Cold-start item recommendation remains a persistent challenge in real-world systems due to the absence of interaction histories. While prior models attempt to bridge this gap using item content features, they universally suffer from the \textbf{seesaw dilemma}: enhancing performance for cold items inevitably degrades performance for warm items, and vice versa. We identify that this dilemma stems from a fundamental \textbf{distributional disparity}: warm item embeddings occupy a complex ``behavioral manifold" shaped by rich interaction signals, whereas cold item embeddings are constrained to a ``semantic manifold" derived solely from auxiliary content. Existing methods often force a rigid mapping between these inconsistent spaces, causing the model to sacrifice the precision of warm representations to accommodate cold ones. To address this, we propose \textbf{DiffCold}, a diffusion-based generative model that unifies warm and cold representations. Unlike GANs or VAEs, DiffCold leverages conditional diffusion to reconstruct warm item embeddings from content, preserving the underlying manifold structure without degradation. We further tailor this paradigm with two specific designs: a \textbf{Retrieval-enhanced Aggregator} that initializes generation using semantically similar warm items to bypass inefficient noise, and a \textbf{Simulation-based Representation Alignment} module that enforces distribution consistency between generated and real embeddings via contrastive learning. Experiments on three benchmarks confirm that DiffCold resolves the seesaw dilemma, consistently outperforming state-of-the-art methods across all metrics.

Speculative decoding (SD) addresses the high inference costs of LLMs by having lightweight drafters generate candidates for large verifiers to validate in parallel. Existing draft-verify methods use binary decisions: accept or fully recompute. Yet we find that many rejected tokens can be verified correctly by a slim submodel derived from the full verifier via intra-model routing, instead of the full verifier. This motivates our slim-verifier to handle tokens requiring moderate verification resources, reducing expensive large-model calls. We propose Verification via Intra-Model Routing for Speculative Decoding (VIA-SD), a multi-tier framework using a routed slim-verifier. Draft tokens are processed hierarchically: direct acceptance for high-confidence cases, slim-verifier regeneration for medium-confidence cases, and full-model verification for uncertain cases. Across four representative tasks and multiple model families, VIA-SD reduces rejection rates by 0.10-0.22 and delivers 10-20% speedups over strong SD baselines, while achieving 2.5-3x acceleration over non-drafting decoding. Moreover, VIA-SD is compatible with existing SD frameworks without modifying their training procedures. Our results suggest multi-tier SD as a general paradigm for scalable and efficient LLM inference. Project page: https://zju-xyc.github.io/VIA-SD-Project-Page/

Multivariate time-series data often exhibit complex temporal dependencies, irregular sampling, and heterogeneous dynamics across multiple time scales, making accurate sequence modeling particularly challenging. Traditional recurrent neural networks (RNNs), such as Long Short-Term Memory (LSTM) networks, operate in discrete time and may struggle to effectively capture continuous and irregular temporal behaviors. Liquid Neural Networks (LNNs) address some of these limitations through continuous-time dynamics, but standard LNN architectures typically rely on a single dynamical system, limiting their ability to model heterogeneous temporal patterns. To address these challenges, we propose a Multi-Rate Mixture-of-Experts (MR-MoE) framework built on top of Liquid Neural Networks. In the proposed architecture, multiple LNN-based experts operate at distinct time scales, enabling the model to explicitly separate fast-changing dynamics from slow-evolving temporal trends. A gating network further enables adaptive expert specialization based on input conditions. In addition, we incorporate both feature-level and temporal attention mechanisms to improve robustness, interpretability, and long-range dependency modeling. Feature-level attention suppresses noisy or irrelevant variables, while temporal attention selectively focuses on informative historical states. We evaluate the proposed framework on a complex multivariate time-series prediction task and compare it against strong baselines, including LSTM, monolithic LNN, and standard MoE models. Experimental results demonstrate that the proposed MR-MoE framework consistently achieves improved AUROC and AUPRC performance while maintaining favorable computational efficiency. These results highlight the effectiveness of combining continuous-time dynamics, multi-scale expert decomposition, and adaptive attention mechanisms for time-series modeling.

Traffic microsimulation combined with surrogate safety measures has increasingly been used as a proactive alternative to historical crash data for predicting crash frequency for current or planned road infrastructure designs. However, existing microsimulation-based safety studies have adopted simplified rule-based behaviour models, which reproduce traffic flow reasonably well but often fail to generate realistic conflict dynamics, limiting crash prediction accuracy. Recent advances in machine learning (ML)-based behaviour models offer a promising opportunity to potentially improve microsimulation realism and crash frequency predictions by learning human driving behaviour directly from large-scale trajectory datasets. To investigate this possibility, traffic microsimulation was conducted for five real-world signalised intersections in Leeds, UK, using both a standard rule-based model and a state-of-the-art ML model. Simulated vehicle trajectories were analysed using a two-dimensional Time-to-Collision metric to identify simulated conflicts, which were then modelled using Extreme Value Theory to predict crash frequency. Results show that conflicts from the ML model yielded crash predictions in line with the real-world crash data, whereas the rule-based model did not permit meaningful predictions, presumably due to a lack of model calibration to the specific simulated intersections. Directly using ML-generated simulated crashes to predict real-world crash frequency also yielded poor results, suggesting that while current ML models can realistically reproduce conflicts, they are not yet able to generate realistic crashes. Overall, the findings demonstrate that ML-based behaviour models are promising for improving crash prediction from simulated conflicts, without a need for location-specific model calibration, and suggest clear future directions for ML-based traffic microsimulation.

Continuous long-term monitoring and diagnosis of biomedical signals, such as electrocardiograms (ECGs), can help mitigate an increasing threat to public health. Artificial Intelligence (AI) models, such as Convolutional Neural Networks (CNNs), provide accurate monitoring and classification for relevant diseases; however, they require more computational resources than conventional AI hardware can typically afford, especially for a resource-constrained environment on the edge. In this work, we present BenDi, an energy-efficient quasi-stochastic systolic architecture for bioelectronic systems on the edge. BenDi leverages multiple levels of energy and power optimization, ranging from circuits to software quantization, including low supply voltage, the \underline{Ben}t-Pyramid data format for quasi-stochastic multiplication, the \underline{Di}P systolic dataflow, and hardware-aware quantization, to handle CNNs with high accuracy on the edge within limited hardware budgets. The hardware implementation results, using a commercial 22nm technology, show that BenDi architecture, at 0.5 Voltage and 100 MHz, offers 3.35x smaller area and 5x higher energy efficiency, compared to state-of-the-art binary-based weight-stationary systolic architectures. Regarding Bioelectronic edge systems, BenDi achieves an order-of-magnitude improvement in energy efficiency and another order-of-magnitude improvement in area efficiency, compared to its counterparts. This significant improvement comes at the cost of 1\% to 3.3\% accuracy loss on the MIT-BIH and Apnea-ECG benchmarks, respectively, compared with conventional computing using the 32-bit floating-point format.

Controlling the output of Large Language Models (LLMs) is a central challenge for their reliable deployment, yet a clear understanding of the involved trade-offs remains elusive. Current approaches to conditioning are often evaluated with a narrow focus on their effectiveness at injecting or removing a target concept, neglecting generation quality. We systematically investigate a range of conditioning methods in both injection and removal scenarios. We find that efficient steering methods frequently achieve conditioning at a steep cost to fluency. Furthermore, we identify a critical yet previously overlooked interaction with the training paradigm: activation steering methods are far less effective on instruction-tuned models than on their base counterparts. Simple prompting and full-fledged supervised fine-tuning, on the other hand, are viable options for concept injection, but are not as good at concept removal. Finally, cheaply computed textual metrics highly correlate to costly LLM-as-judge scores, and provide insights on the behavior of conditioning methods.

Masked diffusion language models (dLLMs) have recently emerged as a competitive alternative to autoregressive language models, with the promise of faster inference via parallel token generation. A notable limitation of the masked formulation, however, is that once a token has been unmasked it can no longer be revised, leaving dLLMs vulnerable to early sampling mistakes. To address this, a growing body of work has sought to extend masked dLLMs with self-correcting (remasking) capabilities. One appealing subset of these methods does so in a training-free, post-hoc manner based on token confidences, with encouraging early reported results. In this work, we revisit the empirical evaluation of a representative post-hoc remasking method, WINO [Hong et al., 2026], and find that under standard decoding settings (shorter block lengths) it brings little-to-no benefit over confidence-based unmasking alone [Wu et al., 2025]. Extending the evaluation to non-greedy decoding, we find that while confidence-based remasking can mitigate errors introduced by increased stochasticity to some extent, it also exacerbates the diversity collapse previously reported for confidence-based unmasking. Overall, our results show that the benefits of post-hoc confidence-based remasking are highly setting-dependent, underscoring the need for a more comprehensive evaluation framework.

The adoption of AI-powered Integrated Development Environments (AI IDEs) has introduced "Rules" as a novel software artifact, allowing developers to persistently inject project-specific constraints and architectural guidelines into the context of Large Language Models (LLMs). Despite their role in aligning AI behavior with developer intent, the taxonomy, evolution, and practical impact of these rules remain largely unexplored. To bridge this gap, we conducted a mixed-methods empirical study on AI IDE rules. By mining 83 open-source projects and extracting 7,310 rules, we established a comprehensive taxonomy comprising 5 primary and 25 secondary categories. We then triangulated these artifacts with survey responses from 99 practitioners. Our analysis identified a contrast between developer priorities and actual configurations: while practitioners rate architectural constraints as highly important, rule files in repositories primarily consist of low-level workflow and code formatting constraints. Furthermore, our analysis of 1,540 rule evolution events revealed that rules are updated frequently. Repository data further indicate that rule evolution is primarily driven by constructive context expansions (29.17%) and enrichments (26.59%). In contrast, surveyed developers reported modifying rules primarily to correct AI errors (77.78%), typically by adding new negative constraints rather than editing existing ones. Finally, an artifact compliance assessment of 160 rule evolution events revealed that updating rules significantly improves the adherence of software artifacts, with the average artifact compliance rate increasing by 22.99% (from 49.14% to 72.13%) following an update. Our study provides empirical insights that can help developers optimize prompting strategies and guide tool builders in designing automated conflict-detection and context-management mechanisms for AI IDEs.

Understanding spatial distribution of fallow land is important for optimizing the food-water (FW) nexus, given fallowing's role in crop rotation and water conservation. Fallow is a low accuracy class in USDA Cropland Data Layer (CDL). Geospatial foundation model (GFM), Prithvi-EO has shown strong transferability across computer vision tasks. However, its Vision Transformer (ViT) backbone produces features at a single spatial scale that are ill-suited for the multi-scale features required by object detection heads. Existing approaches synthesise multi-scale pyramids through scaling of single stride tokens, sacrificing spatial heterogeneity, and full backbone fine-tuning is computationally prohibitive for GFMs. We evaluate a fallow detection pipeline combining two parameter-efficient fine tuning (PEFT) schemes: Low-Rank Adaptation (LoRA) and a hybrid PEFT, with three neck designs: pseudo multi-scale, Lite ViT-Adapter, and Full ViT-Adapter. Our best configuration, Lite ViT-Adapter with a one-stage head, achieves a mAP@50 of 0.9479 with the Diou loss, suggesting the effectiveness of center-aware localization for irregular fallow field detection. ViT-Adapter free one-stage detection under LoRA improves the adapter-free anchor-based approach by 6.42%, and the best configuration improves baseline adapter-free anchor-based approach by 25.70%. These results demonstrate that lightweight spatial prior fusion and selective backbone unfreezing enable Prithvi-EO to capture local fallow patterns more effectively, outperforming approaches that rely on reshaped single-stride ViT tokens.

World Action Models (WAMs) offer a promising route for robot manipulation by using video generation models to model future scene evolution before producing control actions. However, our empirical observations reveal a phenomenon: generating plausible visual futures does not always guarantee the extraction of accurate actions. To diagnose this failure, we conduct action-head attention analysis and causal interventions. We find that the action decoder fails to focus on task-relevant interaction regions and remains sensitive to perturbations in task-irrelevant areas. This reveals a representation mismatch: hidden states optimized for visual reconstruction are not inherently organized in a form useful for low-level action control. In this paper, we propose AGRA, an Action-Grounded Representation Alignment objective that regularizes the world-action interface by aligning intermediate video diffusion features with spatially coherent semantic representations from a foundation visual encoder. We evaluate AGRA on real-world manipulation tasks. Experiments show that AGRA makes world model representations more action-grounded: by focusing the action decoder on the correct interaction regions, it improves object localization accuracy and affordance understanding, and makes the policy more robust to perturbations in task-irrelevant regions. As a result, AGRA consistently improves both in-distribution performance and out-of-distribution generalization over the baseline world action model.

The explosive growth of user-generated video content on online platforms is accompanied by the emergence of numerous near-duplicate videos--videos that are identical or highly similar but differ by partial edits. These duplicates degrade user experience and increase storage and bandwidth costs, making large-scale video deduplication a critical task. Existing video deduplication frameworks face a fundamental challenge in retrieving sufficient high-quality candidates under a limited index budget, as well as trade-offs between efficiency and precision. To address these issues, we propose MLT-Dedup, an efficient large-scale online video deduplication framework with Multi-Level representations and spatial-Temporal matching. Our approach employs a Multi-Level Video Encoder (ML-VE) to extract both fine-grained frame-level and sparse clip-level embeddings: sparse embeddings support efficient candidate retrieval, while fine-grained embeddings are loaded for precise pairwise matching. During matching, we introduce DiF-SiM, a Differential Feature-enhanced Similarity Module capable of locating duplicated temporal segments and providing reliable similarity evidence to support policy-driven deduplication decisions. Extensive experiments on a real-world large-scale platform demonstrate that MLT-Dedup reduces online repetition rates by 91% at 90% precision. Furthermore, our sparse retrieval design achieves a 5x increase in indexing capacity, enabling broader candidate coverage in real-world deployment.

The rapid integration of large language models (LLMs) into mobile applications has introduced a new class of credential security risk: leaked credentials that grant unauthorized access to LLM inference services, causing financial damage to developers. Prior work on credential leakage has focused primarily on Android apps; to date, no empirical study has systematically investigated LLM API key leakage in iOS applications. We present the first in-depth empirical study of API key leakage in LLM-integrated apps. We construct a high-quality dataset of 444 iOS applications, filtered from 1092 candidates through a standardized process, and develop LLMKeyLens, a dynamic analysis framework that detects LLM API key leakage via traffic interception, provider-specific key extraction, and active validity confirmation, requiring neither source code access nor binary decryption. Our analysis reveals that 282 applications expose exploitable LLM API credentials in network traffic, spanning at least ten providers. We identify three leakage patterns: JWT-based token leakage (48%), unauthenticated backend proxy access (33%), and plaintext API key transmission (19%). To assess remediation, we re-analyzed the same 282 vulnerable applications three months after responsible disclosure; only 28% had remediated the reported vulnerability, while 72% remained exploitable, with persistent issues stemming from unauthenticated backends and broken JWT implementations. Our findings show that LLM API key leakage is both prevalent and persistent in the iOS ecosystem, exposing a systemic gap between developer practice and secure integration principles, and suggest that secure LLM integration requires not only developer awareness but also explicit security guidance from providers and platform-level enforcement.

Can financial news reliably predict short-term stock movements? Despite advances in large language models, this question remains unresolved. We revisit this problem using a zero-shot natural language processing framework, investigating whether models can extract actionable signals from financial news without domain-specific training. We design a structured pipeline that combines zero-shot natural language inference with temporal aggregation, explicitly modelling recency and event-dependent impact horizons when integrating information across articles. To address the need for transparency in high-stakes settings, we introduce a multi-layered explainability framework that links predictions to token-level, article-level, and aggregate evidence, and produces grounded natural language rationales. Across multiple models and prediction horizons, we find that zero-shot approaches consistently fail to outperform simple baselines, with particularly weak performance on negative movements, suggesting deeper structural limitations in mapping news sentiment to short-term price dynamics. However, explainability signals reliably distinguish between trustworthy and unreliable predictions, offering practical value even when accuracy is limited. These findings highlight the limits of zero-shot financial NLP and motivate a shift toward decision-support systems that prioritise transparency and uncertainty awareness. Code: https://github.com/alimert05/zero-shot-stock-xai

Embodied intelligence now spans navigation, household assistance, manipulation, autonomous driving, aerial agents, and multimodal large-model control. This expansion has made benchmark construction a central bottleneck for reliable evaluation. Unlike static datasets, embodied benchmarks combine task specifications, environments, robot data, demonstrations, annotations, metrics, evaluation scripts, and release policies into a single evaluation system. This survey reviews the literature through a five-stage construction pipeline: requirement and task construction, data acquisition, data cleaning and annotation, benchmark suite generation and metric definition, and evaluation execution with diagnostic feedback. For each stage, the survey analyzes the transition from manual curation to traditional automation, foundation-model assistance, and agentic closed-loop workflows. It also compares qualitative construction costs across human labor, data and asset acquisition, compute and simulation, validation and debugging, governance and maintenance, and rework risk. The main conclusion is that automation does not simply reduce benchmark cost. Instead, it often shifts cost toward validation, auditability, version control, and long-term governance. Progress in embodied evaluation will therefore depend not only on larger benchmark suites, but also on construction pipelines that are diagnosable, auditable, and responsibly refreshable.

Large language models (LLMs) are widely used to tackle complex tasks with autonomous workflows. Recently, reusable natural language skills have emerged as a popular paradigm to inject procedural knowledge into LLM applications. Since popular skills are often invoked repeatedly, placing their full text in every context significantly increases prefill cost and latency. While text compression techniques have the potential to solve this problem, most existing methods are designed to compress factual knowledge in documents instead of procedural knowledge, making them insufficient for skill compression. In this paper, we argue that an effective skill compression method should: 1) preserve logical dependencies among workflows and tool protocols, 2) enable lightweight, offline compression for frequently updated community skills, and 3) be adaptable to varying complexities across skills. To address this, we present SKIM (SKIll coMpression), an adaptive multi-resolution soft token compression framework for procedural skills. Depending on the complexity of each skill, SKIM creates different numbers of soft tokens that not only improve the efficiency of LLM inference, but also preserve the effectiveness of skill usage. Experiments indicate that SKIM compresses skills to 30 to 60 percent of their original token length while preserving task performance better than existing compression methods.We have released our code at https://github.com/bebr2/SKIM .

We study policy representation learning from unlabeled multi-policy behavioral data. Each episode is generated by a fixed policy, but policy labels are unavailable. This setting appears in robotics play, demonstrations, games, racing, and other datasets where heterogeneous behaviors are mixed without annotations. We introduce \emph{Behavioral INR}, a self-supervised generative model that adapts implicit neural representations (INRs) from vision to behavior. Instead of mapping coordinates to RGB values, Behavioral INR represents a policy as a state-action function mapping states to subsequent actions. An episode-level latent modulates this function through FiLM layers, yielding a generative prior over policies and allowing policy identity to be inferred without supervision. Because INRs treat each datapoint as samples from an underlying function, the same model naturally accommodates variable episode lengths and different sampling granularities, as in vision INRs with different image resolutions. We also define policy-level out-of-distribution (OOD) shifts along state-distribution and action-distribution axes, which arise when policies overlap in states or actions but are not captured by standard behavioral OOD settings based only on new agents or environments. We evaluate on synthetic Gaussian random field data, MuJoCo demonstrations with controlled OOD splits, and real-world chess, Formula 1 racing, robotics, and Seek-Avoid datasets. Behavioral INR most consistently improves policy identifiability in the hardest continuous state-action settings, especially when longer episodes, more policies, and OOD splits reduce the usefulness of marginal shortcuts; amortized history encoders remain competitive when policy identity can be recovered from symbolic repetition or low-dimensional action statistics. We release code and checkpoints.

Spoken dialogue models typically start from text LLM backbones, yet reasoning often degrades when conditioning on speech instead of text. We attribute part of this modality gap to a temporal-granularity mismatch: speech tokens are temporally redundant and far longer than text under matched semantics, diluting per-token semantic density and weakening text-native reasoning dynamics. We study speech token design as a representation selection problem and sweep frame rates under a frozen LLM backbone with a fixed information rate. To make low frame rates feasible, we introduce factorized FSQ and a lightweight non-autoregressive audio LM head, scaling capacity to nearly 300\,bits/frame without sacrificing efficient prediction. With the bottleneck removed, we sweep frame rates (50$\rightarrow$2.08\,Hz) and alignment depth, and observe a consistent best regime for speech QA at 4.17\,Hz with intermediate-layer representation alignment.

Environments serve as interactive systems for large language model (LLM) based agents across diverse scenarios and play a crucial role in driving the continual evolution of model capabilities. Despite this importance, existing work lacks a systematic categorization and deep analysis. This paper systematically studies current researches on agentic environments from the perspective of the environment engineering lifecycle, covering their modeling, synthesis, evaluation and application. Specifically, the paper first introduces representative environments from the perspectives of eight attributes and eight domains, providing detailed analyses of their development paths and highlighting their core capabilities. Second, for automated environment synthesis, two paradigms are introduced, such as symbolic synthesis and neural synthesis. This paper also shows different environment evaluation methods in each paradigm. Thirdly, the corresponding environment applications from the perspective of agent-environment co-evolution are discussed. In specific, the paper characterizes the primary pathways for agent evolution in dynamic environments from four complementary perspectives: memory-centric experience evolution, orchestration-centric workflow evolution, trajectory-centric offline evolution, and exploration-centric online evolution. And three paradigms of environment evolution are identified, namely neural-driven, difficulty-driven, and scaling-driven approaches. At last, several promising future directions are discussed, including Environment-as-a-Service, Multi-agent Environments, and Neural-Symbolic Environments.

Enthymemes, arguments with unstated premises or conclusions, are pervasive in persuasive discourse, yet their annotation remains notoriously subjective. We present a resource of 1,482 tweets from politically controversial discourse, annotated by five annotators for the presence of enthymemes and their argument structure, designed to study label variation. We first revisit the definition of enthymemes and propose annotation guidelines anchored in Walton's argumentation schemes, offering a structured and constrained approach that nonetheless preserves room for the interpretive nature of the task. This contrasts with past resources, which tend to eliminate disagreement, obscuring its sources and preventing investigation of its potential benefits for model performance. We further propose a complexity analysis of the task, identifying where annotation imposes high cognitive load and may give rise to inconsistent annotation. Our preliminary experiments show that models trained on annotator disagreement outperform models trained on hard majority-vote labels. We close by reflecting on how structural openness in enthymeme definitions and guidelines enables the study of variation in subjective inferential processes for future resources and downstream NLP applications concerned with human inference.

Identifying the governing equations of complex dynamical systems remains a fundamental challenge across science and engineering. While early approaches relied on empirical data and heuristics, modern data-driven methods offer greater flexibility and fewer assumptions. However, data acquisition in real-world settings is often expensive. This work addresses this challenge by introducing an active learning strategy for dynamics discovery in the ultra-low data limit. Rather than sampling randomly, our method iteratively prioritizes regions that are most informative for model identification. This approach builds on Sparse Identification of Nonlinear Dynamics (SINDy), and utilizes an ensemble extension, E-SINDy, to estimate epistemic uncertainty and guide the sampling for both ordinary and partial differential equations (ODEs/PDEs). For ODEs, an exhaustive analysis is conducted on the Lorenz system across varying data budgets and noise levels. For PDEs, two systems with contrasting dynamical characteristics are examined: the Burgers' equation, where a sharp shock front creates a distinction between informative and uninformative regions, and the Kuramoto-Sivashinsky equation, which presents a more spatially complex sampling landscape. Across all scenarios, the proposed method accurately identifies the governing dynamics with significantly fewer data samples than random sampling.

Knowledge Distillation (KD) and mixup have proven effective at inducing smoothness in class boundaries; KD captures inherent class relationships in probability distributions, and mixup enforces them through convex combinations of inputs. Their interaction, however, remains poorly understood, particularly when mixup is applied only during student training. In this setting, the teacher is queried on inputs drawn from a vicinal distribution it never saw during training, a controlled mismatch whose effect on knowledge transfer has not been characterised. We show that this mismatch causes the teacher's supervisory signal to be dominated by distributional confusion rather than inter-class structure. Despite it, the student does not merely imitate the teacher: it independently acquires greater linearity in the vicinal region, a structural property that the teacher lacks, and goes beyond dark-knowledge transfer. KD with mixup consistently improves student accuracy and reduces overconfidence by an order of magnitude relative to the baseline, across CIFAR and ImageNet with varying-capacity teachers. Crucially, calibration propagates from teacher to student independently of accuracy transfer, and temperature scaling governs a measurable accuracy-calibration trade-off that becomes more pronounced under vicinal training. These results reframe mixup distillation not as a degraded version of standard KD, but as a richer transfer channel that simultaneously shapes discriminative performance, uncertainty estimation, and representational geometry.

High-stakes clinical use of large vision-language models (LVLMs) requires reasoning that is grounded in visual evidence and clinical knowledge, not just correct final answers. We introduce OpenMedReason, a large-scale, open multimodal medical reasoning corpus comprising approximately 450K image-question-answer instances whose reasoning traces are primarily derived from curated biomedical, human-authored scientific articles. OpenMedReason provides high-fidelity supervision beyond synthetic chains of thought, covering diverse medical domain vision modalities such as radiological scans, microscopic images, visible light photographs, charts, and others. We complement it with OpenMedReason-Bench, a held-out benchmark that allows fine-grained evaluation of LVLMs along three complementary axes of capability, including perception, medical knowledge, and rationale, enabling diagnostic evaluation beyond final-answer accuracy. OpenMedReason is a rich training resource that exhibits its effectiveness in both supervised fine-tuning (SFT) and reinforcement-based alignment. Training with OpenMedReason yields a 20% average improvement in VQA accuracy over the base model and achieves performance within 4.2% of the strongest comparable-scale medical LVLMs. Fine-grained performance analysis confirms that the gains are not concentrated in any single axis: OpenMedReason improves perception, medical knowledge, and rationale jointly, and its reasoning traces are preferred over those of the base model in 86.1% of pairwise comparisons. We release the code and dataset at huggingface.co/datasets/neginb/OpenMedReason.

Decoding-time truthfulness methods -- layer-contrast decoding, inference-time intervention, and learned logit adapters -- have demonstrated 10-30 point gains on TruthfulQA when applied to base language models. However, modern instruction-tuned LLMs already achieve substantially higher baselines (61-76%), raising the question of whether these methods remain effective in practice. We design a six-control evaluation framework -- out-of-distribution training, multi-judge validation, simple decoding baselines, confound controls, bootstrap confidence intervals, and seed variance -- and apply it across 5 models (1B-70B), 3 benchmarks, and 15 methods. We find that previously reported gains shrink substantially under strict controls: on the full TruthfulQA benchmark (N=817), no token-level method achieves statistically significant improvement, and the best learned adapter scores -2.0 points below greedy (p=.23). We identify five evaluation sensitivities -- contamination, judge choice, missing baselines, confounds, and statistical noise -- that individually or jointly account for these discrepancies. Cross-benchmark validation on HaluEval QA and TriviaQA confirms that these patterns extend beyond TruthfulQA. Deliberative prompting methods (chain-of-thought, self-critique) appear more robust in the evaluated regime, with CoT achieving +5.6-19pp across benchmarks as a training-free, single-pass method. We release a seven-point evaluation checklist and discuss implications for future truthfulness research.

Model merging (MM) has gained significant attention as a cost-effective approach to integrate multiple task-specific models into a unified model. However, recent work reveals that MM is highly susceptible to backdoor attacks. Existing defenses based on task arithmetic often fail to eliminate backdoors without substantially degrading clean-task performance, owing to their reliance on direct parameter-space editing. To address this gap, we propose Linear Feature Path Minimization (LFPM), a backdoor mitigation framework for model merging, which introduces an anti-backdoor task vector into the backdoored merged model. Unlike prior approaches, LFPM formulates the backdoor robustness of the merged model from a unified feature-space perspective under the Cross-Task Linearity (CTL) framework, which leverages the approximate linearity of features across tasks. This perspective guides the optimization of the anti-backdoor task to suppress backdoors while preserving clean-task performance. Furthermore, we introduce an effective optimization mechanism based on gradient accumulation and loss path-integral, ensuring robust backdoor suppression along the interpolation path. Extensive experiments demonstrate that LFPM consistently exhibits strong robustness against backdoor attacks in both full fine-tuning and Parameter-Efficient Fine-Tuning (PEFT) settings.

Vision-language-action (VLA) models predict chunks of future actions from the current observation, an assumption that fails under partial observability, where decisions depend on information no longer visible. Existing memory-augmented VLAs simultaneously introduce recurrence, retrieval, compression modules, auxiliary objectives, hierarchical memory, or task-specific architectural changes, so the contribution of recurrence itself remains entangled with surrounding machinery. We present a controlled isolation study of recurrence in a strong pretrained VLA backbone. Our formulation augments the transformer with a small set of learnable memory tokens carried across timesteps and updated through self-attention, trained end to end with truncated backpropagation through time, with no auxiliary losses and no architectural changes. We instantiate this as $μ$VLA, a family of OpenVLA-OFT variants parameterized by memory width m, TBPTT length K, and the memory update rule (cross-step gradients or a detached EMA), so that recurrence is the only varying factor. On MIKASA-Robo, $μ$VLA improves average success rate on five training tasks from 0.42 to 0.84 at the strongest setting and reaches 0.23 on held-out tasks with the same memory structure versus 0.07 for the memoryless baseline. On tasks requiring different memory structure, performance remains near baseline. On LIBERO, the strongest recurrent variant achieves 96.2% average success, indicating no regression under full observability. We interpret these results as a calibration of the capability envelope of minimal in-backbone recurrence, identifying the regime in which it is sufficient and the regime where additional memory structure is required. Demos and videos can be found in https://avanturist322.github.io/mu-vla/.

The design of reinforced concrete highway barriers is a safety-critical process that requires strict compliance with regulatory provisions such as the AASHTO-LRFD bridge design guidelines. Current engineering practice relies heavily on manual, iterative, and heuristic calculations to satisfy complex nonlinear material and mechanics constraints. Although Large Language Models (LLMs) demonstrate strong generative capabilities, their direct application to structural engineering remains limited by hallucination risks and insufficient physical grounding. To address these challenges, this study proposes a novel "generation-evaluation-optimization" closed-loop framework for automated concrete barrier design using the multi-agent orchestration capabilities of AutoGen. Experimental results demonstrate that the proposed agentic framework achieves over 98% design accuracy, significantly outperforming standalone general-purpose LLMs. More importantly, the study reveals that design performance is not necessarily correlated with model scale, where an 8B-parameter lightweight model could outperform unconstrained 631B-parameter flagship models. This finding highlights the potential to substantially reduce computational costs while improving the accessibility of AI-assisted engineering tools for industry applications. The source code for the proposed multi-agent design framework is available at the project GitHub repository: https://github.com/MXY820/barrier-design. Keywords: Structural Engineering; Multi-Agent Systems; Large Language Models; Concrete Barrier Design; AutoGen; Design Automation.

Finite element (FE) modeling of safety-critical infrastructure such as bridge barriers requires high-fidelity nonlinear dynamic analysis, yet the current FE modeling process remains labor-intensive and lacks automation. This paper presents the Human-Enhanced Loop Modeling (HELM) framework, a collaborative human-agent protocol that decomposes long-sequence finite element modeling into discrete, visually verifiable checkpoints across geometry generation, boundary condition definition, and material assignment. The framework is demonstrated through a 20-case matrix of reinforced concrete bridge barriers under MASH TL-4 and TL-5 lateral loading conditions, interfacing specialized agents with two widely used commercial FE softwares, i.e., ANSYS and LS-PrePost. Experimental results show that HELM improves the baseline autonomous modeling success rate from 20% to 75%, with agent-level pass rates for geometry and boundary condition tasks approximately doubling. Error analysis reveals that spatial reasoning and algebraic logic limitations constitute the primary failure modes, underscoring the value of structured human-in-the-loop intervention for modeling automation. The complete agent design code and prompts are open-sourced and can be accessed at: https://github.com/SimAgentDev/Ansys-LSPP-AgentKit.

Reducing real-world trial and error has long been a central goal of decision making, and generative simulators advance this goal by modeling the evolution of future states. An even more challenging yet meaningful task is simulating how disturbance events (e.g., accidents) propagate their impacts across real-world networks. The existing approaches fall short of modeling both structured attributes and unstructured semantics of events, and capturing topological structures in simulating network event evolution. Therefore, we are motivated to propose Net-Ev$^2$ ($\underline{\textbf{Net}}$work $\underline{\textbf{Ev}}$ent $\underline{\textbf{Ev}}$olution), a novel generative simulator that jointly leverages event cues while preserving network topology in simulations. Specifically, the framework consists of two stages, namely structure-guided masked pre-training and topology-aware diffusion process, which is achieved by U-Net-like graph downsampling and upsampling during denoising. At inference time, Net-Ev$^2$ can generate simulations using natural-language event input only, with greater flexibility for practical usage. Furthermore, we introduce Net-Ev$^2$-6.5M, a multimodal benchmark of aligned event and network traffic data across four large-scale road networks, as well as a new topology-aware metric, namely JL-MMD, to evaluate topological fidelity in generated network dynamics. Extensive experiments demonstrate the state-of-the-art performance and strong generalization ability of Net-Ev$^2$. Code is made available at https://github.com/Guangyu4/Net-Ev-2.

Certified local robustness verification for recurrent neural networks (RNNs) is challenging because approximation errors introduced by nonlinear relaxations can propagate through recurrent connections and accumulate over time. As a result, scalable linear bound propagation methods often become overly conservative and fail to certify inputs that are in fact robust, especially when many pre-activation intervals cross zero. We propose an abstraction-refinement framework for RNN verification that partitions such intervals to remove the dominant relaxation error: on each refined branch, ReLU becomes exact, and smooth activations such as tanh and sigmoid admit substantially tighter linear envelopes. To control the combinatorial cost of splitting in long sequences, we introduce a SHAP-guided timestep selection strategy that ranks hidden states by their contribution to the verification objective and refines only the most critical timesteps in temporal order. Experiments on CIFAR10 and MNIST stroke benchmarks demonstrate consistent improvements in verification success and robustness-margin tightness over abstraction-only baselines, while exposing clear runtime trade-offs between ReLU and tanh models.

Molecular communication (MC) suffers from severe diffusion memory because molecules released for one symbol may arrive during later symbols. Neural sequence detectors, especially sliding bidirectional recurrent neural networks (SBRNNs), can substantially outperform threshold detectors in such channels. This raises a central question for MC channel coding: does a code whose advantage was established under threshold detection retain it when both coded and uncoded transmission are evaluated with neural detection? This letter answers this question for run-length-limited ISI-mitigation (RLIM) codes, a class of constrained codes previously shown to provide large BER gains in MC. Across the tested operating points, the best RLIM-SBRNN receiver beats the best uncoded receiver, chosen between threshold and SBRNN detection, in $46$ of $59$ cases, with a mean gain of $10.36\times$ over those wins. We also propose an RLIM-tailored training mask for compact SBRNN detectors, improving the unmasked RLIM-SBRNN in $227$ of $236$ comparisons with $3.267\times$ mean gain when masking is beneficial. Finally, the compact masked RLIM-SBRNN is competitive with channel-state-aware MLSE despite using no channel knowledge.

Predicting psychological traits from asynchronous video interviews (AVIs) is a challenging problem in AI-assisted interview assessment because labeled datasets are limited while each response contains high-dimensional visual, acoustic, and verbal signals. This paper presents our solution for the ACM Multimedia AVI Challenge 2026, which evaluates two tasks: Track~1 predicts self-reported HEXACO personality traits from personality-related interview responses, and Track~2 classifies cognitive ability levels from structured AVI responses. We treat the problem as a small-sample representation learning task. Instead of fine-tuning large pretrained models, we use frozen multimodal encoders, including CLIP for visual features, Whisper for acoustic features and transcripts, and RoBERTa, E5, and DeBERTaV3 for textual representations, followed by low-capacity downstream models. For Track~1, our trait-specific regression and late-fusion system achieves an average validation MSE of 0.2696, improving over the official baseline of 0.3334. Ablation results show a three-step improvement from a global model (0.3189), to per-trait modeling (0.2871), to per-trait late fusion (0.2696), corresponding to a 19.1% relative MSE reduction over the official baseline. For Track~2, a compact subject-attribute baseline reaches 0.5781 accuracy, while our multimodal ensemble reaches 0.5313, both above the official baseline of 0.4062. We interpret this result as evidence of possible subject-attribute shortcuts in the validation split rather than robust cognitive inference from AVI content. Overall, our findings suggest that AVI-based psychological assessment benefits from trait-specific multimodal modeling, but cognitive ability prediction requires careful control of dataset shortcuts.

Research on Text-Attributed Graphs (TAGs) has gained significant attention recently due to its broad applications across various real-world data scenarios, such as citation networks, e-commerce platforms, social media, and web pages. Inspired by the remarkable semantic understanding ability of Large Language Models (LLMs), there have been numerous attempts to integrate LLMs into TAGs. However, existing methods still struggle to generalize across diverse graphs and tasks, and their ability to capture transferable graph structural patterns remains limited. To address this, we introduce the GraspLLM, a framework that combines Graph structural comprehension with semantic understanding prowess of LLMs to enhance the cross-dataset and cross-task generalizability. Specifically, we represent node texts from different graphs in a unified semantic space with a frozen general embedding model, on top of which we perform motif-aware contrastive learning across multiple motif-induced adjacency matrices to extract dataset-agnostic structural information. Then, with our proposed optimal contextual subgraph, we extract the most contextually relevant subgraph for each target node and align these subgraphs to the token space of LLM via an alignment projector. Extensive experiments on TAG benchmark datasets spanning diverse domains reveal that GraspLLM consistently outperforms previous LLM-based methods for TAGs, especially in zero-shot scenarios, highlighting its strong generalizability across different datasets and tasks. Our code is available at https://github.com/Heinz217/GraspLLM.

Learning compatible representations aims to learn feature representations that can be used interchangeably over time whenever a model undergoes updates. In this paper, we demonstrate that stationary representations learned by d-Simplex fixed classifiers imply compatibility as in its formal definition. This result establishes a foundation for future works and can be directly exploited in practical learning scenarios. We address the challenge of learning compatibility using $d$-Simplex fixed classifiers when the model is sequentially fine-tuned. Learning according to a d-Simplex fixed classifier with the cross-entropy loss aligns feature distributions at the first-order statistics. Consequently, it may not fully capture higher-order dependencies in the representation between model updates. To address this issue, we demonstrate that training the model using a $d$-Simplex fixed classifier through a convex combination of the cross-entropy loss and a contrastive loss not only captures higher-order dependencies, but is also equivalent to learning with the cross-entropy under the compatibility constraints. We confirm our findings with extensive experiments also considering a new scenario where a pre-trained model is sequentially fine-tuned and occasionally replaced with an improved model. We show that stationary representations enable uninterrupted retrieval services (without reprocessing gallery images) while improving performance during model updates and replacements, achieving state-of-the-art. Code at https://github.com/miccunifi/iamcl2r.

Post-training quantization (PTQ) is essential for efficient large language model inference, but reliably quantizing activations remains challenging when weights, activations, and KV caches are all quantized to 4-bit precision. A key difficulty lies in massive activations, whose extreme values dominate the activation range and amplify quantization errors. State-of-the-art methods mainly mitigate massive activations through transformation-based smoothing, such as orthogonal rotations and affine scaling, but overlook the cross-layer dynamics of the residual stream. In this paper, we show that massive activations emerge and disappear in a phase-wise pattern across network depth, triggering large residual changes. These changes cause newly injected layer-wise updates to dominate the 4-bit quantization scale and weaken historical residual information. To characterize this behavior, we introduce Jump Ratio and Historical Feature SNR. This suggests that static transformation-based smoothing cannot fully resolve dynamic quantization instability caused by cross-layer residual changes. Based on this analysis, we propose DynamicPTQ, a Dynamic Post-Training Quantization policy for phase-aware mixed-precision activation quantization. DynamicPTQ identifies quantization-sensitive layers from residual-stream dynamics and assigns 8-bit activation precision only to these layers, while keeping weights, KV caches, and other activations in 4-bit precision. It can be directly integrated with strong PTQ baselines such as QuaRot, SpinQuant, and FlatQuant. Experiments on LLaMA-2 and LLaMA-3 show that DynamicPTQ consistently improves perplexity and zero-shot QA performance under W4A4KV4 quantization, while achieving 1.05 to 1.07 times throughput improvement with modest memory overhead. These results demonstrate a practical path toward robust low-bit LLM inference.

Condition-based Predictive Maintenance (PdM) for truck fleets has gained momentum in recent years. This maintenance strategy aims to minimize unplanned downtimes and reduce costs by monitoring the health status of vehicles and taking proactive action based on their condition. However, the implementation of condition-based PdM systems is challenging due to the large volume of data generated by the trucks, the inherent complexity of detecting failures through sensor data and the difficulties in finding cost-effective trade-offs in the solution's implementation. In this paper, we define and validate a condition-based PdM methodology built on the assumption that the wear-and-tear state of the monitored component can be represented as a monotonically non-decreasing time series. It involves selecting only the most recent observations from the time series and transforming them into a tabular format for classification using machine learning (ML) models designed for tabular data. Our results indicate that the proposed methodology reduces costs on the Scania Component X dataset compared to current state-of-the-art (SOTA) approaches, while also simplifying the modeling process through AutoML.

Training interactive web agents through imitation learning from expert trajectories has emerged as a highly effective approach. However, determining the optimal timing for expert intervention presents a critical challenge in this context. Delayed intervention often leads to the accumulation of early-stage errors, pushing the page state into an irrecoverable regime. Conversely, premature or excessive intervention causes the agent to become overly reliant on expert policies, trapping the model in local optima characterized by a single, rigid trajectory. We propose Speculative Rollback Correction (SRC), a branch-level imitation framework for resettable agent environments. Instead of requesting teacher labels at every visited state or correcting only after a completed trajectory, SRC uses fixed-horizon branch review: the student executes a short speculative segment before teacher review, and the teacher localizes the first harmful deviation only when local progress breaks. Rollback preserves useful prefixes, while successful rollouts are filtered by a hard verifier and retained in a lightweight quality-diversity archive. The resulting data supports next-action supervised fine-tuning on both localized corrections and verifier-passing trajectories. On WebArena-Infinity, SRC collects 977 verifier-passing trajectories and 9,183 next-action examples; fixed-horizon review improves the recovery-versus-query tradeoff over step-level review while retaining verifier-passing solution variants. Code is available at https://github.com/LongkunHao/SRC_gui_agent.

Uncertainty in the terrestrial carbon cycle remains a major constraint in climate projections, partly driven by the uncertainties affecting the land surface representation and variability in Earth system models. To address this limitation, we present a data-driven framework AI4Land, for generating high-resolution historical reconstructions and future projections of key land surface variables. The framework follows a two-phase approach using a U-Net architecture. In the first phase, which is the focus of this work, it reconstructs annual land use and land cover by integrating coarse-resolution scenario data with static geophysical features. In a planned second phase, the resulting high-resolution maps will be used to predict dynamic biophysical variables, particularly leaf area index, at finer temporal scales. Trained on Earth observation data, the models learn to reproduce spatially explicit and physically consistent land surface patterns, extending temporal coverage to periods lacking direct observations. AI4Land was developed and trained on MareNostrum5, demonstrating how GPU-accelerated HPC infrastructure enables global-scale climate AI pipelines. The final product is a suite of open-source emulators designed for real-time coupling with digital twin platforms, such as those developed under the Destination Earth initiative. By delivering realistic and evolving land surface conditions on demand, this work aims to reduce critical uncertainties and improve the predictive power of next-generation climate simulations.

Video Large Multimodal Models have achieved remarkable progress in video understanding, yet they remain prone to hallucinations, where generated responses are not faithfully supported by the input video. In this paper, we propose MultiToP, a multimodal-context-aware visual token patching framework that mitigates hallucinations by refining unreliable visual tokens before language generation. MultiToP introduces a lightweight Visual Token Patcher to predict token-level replacement distributions and selectively substitute unreliable visual tokens with a dynamic global patch token. To train the patcher effectively, we further propose information-guided rank calibration, which uses answer-conditioned frame-level information cues derived from the backbone to guide token replacement. Combined with ground-truth answer supervision and sparsity regularization, MultiToP enables localized visual evidence refinement without modifying the original model. Extensive experiments demonstrate that MultiToP effectively reduces hallucinations on Vript-HAL with negligible inference overhead, improving the F1 scores of Qwen3-VL-4B-Instruct by 50.60% over the vanilla model. Meanwhile, MultiToP preserves general video understanding ability, yielding an 18.58% relative accuracy gain on ActivityNet-QA for Video-LLaVA-7B.

Anomaly detection plays a critical role in identifying unusual patterns across domains such as fraud detection, network intrusion, and system fault diagnosis. Recently, Christoffel function-based methods, rooted in polynomial optimization, have emerged as promising alternatives to deep learning due to their strong mathematical foundations and computational frugality. However, their practical applicability is hindered by the need to invert a matrix whose size grows exponentially with the data dimension, rendering the method intractable even for moderate-dimensional datasets. This paper addresses the dimensionality limitations of Christoffel function-based anomaly detection while preserving its key theoretical properties, i.e., the on-off support dichotomy behavior and the accurate support shape capture. We introduce UCF, a univariate Christoffel function which is based on the squared distance between the query point and the support points. Extensive experiments on the ADBench benchmark demonstrate that UCF consistently outperforms 14 state-of-the-art baselines in terms of Average Precision. By resolving the scalability bottleneck of the Christoffel Function, this work expands the toolkit of anomaly detection methods with a robust, theoretically grounded, and universally applicable approach.

Blind grasping with a dexterous hand is a crucial manipulation capability. Nevertheless, learning such tactile-only policies for real robots remains challenging due to the tactile sim-to-real gap and the limited expressiveness of sparse tactile signals. To bridge this gap, we propose a framework for tactile-only blind grasping that is deployable on a physical multi-fingered robotic hand. Our approach combines three key components. First, we introduce a Real2Sim tactile calibration pipeline that constructs a contact-calibrated digital-twin simulator capable of reproducing real tactile signals. Second, we improve the expressiveness of sparse tactile observations using a layout-aware tactile encoder, which incorporates sensor-geometry priors through self-supervised pretraining. Third, to improve generalization to unseen objects, we train object-specific reinforcement-learning experts in the calibrated simulator and aggregate their successful grasp trajectories into a tactile-conditioned Diffusion Policy. We evaluate our method on a physical LEAP Hand equipped with distributed tactile sensing across 10 seen and 10 unseen objects. The deployed policy achieves a 27\% real-world grasp success rate across all 20 objects, without real-world grasping demonstrations or visual input. Simulation ablations show that layout-aware tactile pretraining improves grasping performance, while sensing-level evaluations confirm that Real2Sim calibration increases the consistency of tactile contact events between simulation and hardware. Together, these results suggest that contact-event calibration, geometry-aware tactile representation learning, and diffusion-based policy aggregation provide an effective path toward tactile-only blind grasping on real dexterous robotic hands. Project page:Dex-Blind-Grasp.github.io.

Large language models (LLMs) have demonstrated strong reasoning and instruction-following capabilities, making them potentially powerful tools for time-series analysis. However, time series lie outside their native textual modality, raising a fundamental question: how should time series be represented so that LLMs can reason about them effectively? Existing work typically serializes raw numerical sequences or fine-tunes pre-trained LLMs on time-series data. These approaches place the burden of extracting temporal structure directly on the LLM, creating a modality mismatch that often degrades performance on long sequences and introduces substantial computational overhead. In this work, we introduce Time-Series-to-Structured-Program representation (T2SP), a deterministic, training-free method that represents a time series as a structured symbolic program. T2SP decomposes time series into trends, periods, and salient events, expressing them in a program-friendly format aligned with the textual and code-like modalities on which LLMs are natively trained. By shifting temporal-structure extraction from the model to the representation itself, T2SP enables off-the-shelf LLMs to leverage their existing reasoning capabilities for time-series understanding. We evaluate T2SP on three reasoning tasks -- editing, captioning, and question answering -- where it consistently improves performance, reduces reasoning time, and lowers failure rates compared with raw-string representations. Our results demonstrate that T2SP provides an effective interface between time series and LLMs.

Large language model (LLM) routing has emerged as an effective paradigm for leveraging the complementary strengths of multiple LLMs through dynamic model and reasoning-strategy selection. Recent reinforcement learning (RL)-based routing methods further improve routing quality by optimizing routing policies from interaction feedback. However, they still struggle to provide informative and comparable learning signals under heterogeneous tasks with varying difficulty. In practice, multiple objectives (e.g., correctness, format behavior) are aggregated into a single scalar reward, leading to ambiguous credit assignment and conflicting optimization signals. Moreover, reward signals exhibit significant variability across instances, where some instances produce higher or more variable rewards, introducing optimization bias that favors trivial samples over informative ones. To address these issues, we propose \textbf{ReCal}, a \textbf{\underline{Re}}ward \textbf{\underline{Cal}}ibration framework for RL-based LLM routing. We first introduce a hierarchical reward decomposition mechanism with component-wise advantage estimation. We further propose a distribution-aware optimization strategy that calibrates optimization variability through variance-aware reweighting and per-dataset normalization. Experiments on seven datasets demonstrate that ReCal consistently improves routing performance, and training stability over baselines. Code is available at https://anonymous.4open.science/r/ReCal.

Attention mechanisms are central to modern sequence models, yet standard attention computes relevance primarily through individual query--key similarities. Although softmax normalization introduces competition among positions, a standard attention layer does not explicitly parameterize learnable interactions between attention decisions. This limits its ability to directly model cooperative or antagonistic co-attention structure within the attention mechanism itself. We propose Boltzmann attention, an energy-based generalization in which attention patterns are governed by an interacting Ising model. The method augments the usual data-dependent local fields with learnable pairwise couplings, allowing the model to represent inter-position correlations beyond those captured by softmax or sigmoid attention. Experiments on character-level language modeling and synthetic bracket matching show that Boltzmann attention consistently improves over standard softmax attention within a standard Transformer architecture, with the advantage becoming more pronounced as sequence length increases. A four-way ablation confirms that the improvement arises from the learnable pairwise couplings. These results suggest that explicit inter-position interactions provide a principled enhancement for attention-based sequence modeling. Moreover, the Ising formulation opens a natural path toward quantum-computing-based sampling strategies: we demonstrate that diabatic quantum annealing provides a practical training method while maintaining competitive performance with exact Boltzmann computation.

Token-level hallucination detectors are evaluated as classifiers, by AUC over all tokens, yet a streaming monitor is judged by its reaction time: the number of tokens that pass between the onset of a hallucination and the alarm. We formulate hallucination onset detection as a quickest change detection problem. A first-order Markov model of the latent faithful/hallucinated state, validated on RAGTruth, places the task inside classical change-point theory and yields Lorden's lower bound on detection delay: about 1.3 tokens at a false-alarm rate of 0.01. We then show that a causal recurrent labeler acts as a CUSUM with a learned increment; at a matched false-alarm rate it detects in 11-13 tokens, against 31 for a linear per-token baseline, and a controlled decomposition attributes most of this advantage to a better per-token score rather than to temporal accumulation. An information-rate optimality theorem of Donsker-Varadhan type explains the remaining order-of-magnitude gap: the learned score realizes only 1/4.5 of the divergence the features carry, a deficit that recalibration cannot remove, with the remainder a finite-horizon effect. Classification metrics conceal this delay structure; sequential analysis makes it measurable

We propose UR-BERT, a Romanized transcription-based text-to-speech (TTS) encoder for massively multilingual TTS systems. Conventional grapheme-to-phoneme (G2P)-based approaches are limited to around 100 languages due to the availability of reliable G2P resources. In contrast, UR-BERT scales to 495 languages by unifying diverse writing systems into a shared Romanization representation. To further enhance phonetic fidelity and text-speech alignment, we introduce a speech token prediction objective during training, which encourages the encoder to learn speech-aware phonetic representations in a data-efficient manner. Experiments show that TTS systems built on UR-BERT consistently outperform recent text encoder baselines across a wide range of languages and resource conditions, and demonstrate strong generalization to unseen languages.

LLM-based multi-agent systems (MAS) solve complex tasks through inter-agent collaboration, but their communication-driven nature also allows security risks to spread across agents and trigger system-wide failures. Existing MAS defenses mainly follow a reactive paradigm after execution by detecting and isolating harmful agents, which may cause irreversible damage and degrade collaborative utility. To address this, we propose a proactive defense framework for MAS security, namely a Simulation-aware Interception Guard (SAIGuard). SAIGuard performs communication-state simulation over the MAS interaction graph, estimates the impact of incoming messages on local agent states and the global MAS state, and detects risky messages via reconstruction deviations from benign communication patterns. Instead of isolating agents, SAIGuard sanitizes or regenerates suspicious messages before it propagation into system. Experiments across diverse topologies and attack scenarios show that SAIGuard reduces attack success rates while maintaining MAS utility, outperforming reactive defenses.

A social highlighter's most useful signal -- which passages a crowd of readers marks -- exists only for documents people have already read. Can the aggregate crowd salience of a document be predicted from its text before its marks accumulate? Prior work on this data found that zero-shot language models recover highlight locations worse than a trivial lead (position) baseline, so we ask whether a model trained on the highlight corpus can beat that baseline. Using a pre-registered ladder of models and a by-document cluster bootstrap, we find a small but robust edge: a logistic ranker over sentence embeddings and positional/contextual features beats the lead baseline by +0.044 average precision (95% CI [+0.029, +0.058]; clears a pre-registered margin delta=0.03 in 97% of resamples, and stable across pipeline re-runs). Two unsupervised extractive baselines (centroid, LexRank-style centrality) lose to lead, and the trained model beats them by +0.108, so the edge is not recovered by generic unsupervised proxies -- it reflects learning from real reader marks. In product terms, precision@3 rises from 0.25 to 0.39 (+55% relative) and the model beats lead on 69% of documents. An ablation attributes the edge to the raw embedding (+0.014) and training augmentation (+0.010), each with a positive CI. The edge is not a temporal-generalization failure, and we find no evidence that content drift or near-duplicate leakage explains it. A standardized regression shows the advantage is governed mainly by document popularity (lower popularity, larger edge) and by label reliability. It nearly vanishes only on the most popular content; there it is the lead baseline that strengthens, not the model that weakens. Because our evaluation conditions on documents that eventually accumulated readers, these results are a retrospective cold-start simulation.

Klindt, LeCun, and Balestriero (arXiv:2605.26379) proved that Joint-Embedding Predictive Architectures (JEPAs) achieve linear identifiability, the linear recovery of the world's true latent variables, if and only if the world's latent dynamics follow a Gaussian, stationary process. This Gaussian boundary implies a fundamental limit on temporal consistency: for any non-Gaussian physical system, the representation error of a statistical World Model grows monotonically with time. We prove that this limit is an artifact of the statistical alignment mechanism, not a property of World Models in general. We introduce the Physics-Grounded Symbolic Architecture (PGSA) and prove three results: (1) a PGSA achieves exact linear identifiability for all physical regimes, regardless of the latent distribution; (2) the per-step error of a PGSA is bounded by numerical precision alone; and (3) as a direct consequence, a PGSA maintains temporal consistency for an unbounded number of transitions, a property we term near-infinite temporal consistency. We further prove that statistical World Models cannot achieve this property for any non-Gaussian system, regardless of model capacity or the volume of training data. The algebraic cores of four of the theorems are formalized in Lean 4 with Mathlib4 v4.31.0 (zero sorry placeholders); the Klindt et al. converse is taken as an external premise. The contrast establishes that symbolic grounding in the causal generator of the world's dynamics is the sufficient condition and, in non-Gaussian regimes, the only condition for near-infinite temporal consistency.

Cross-modal alignment (CA) and cross-modal prediction (CP) are the dominant paradigms for multimodal representation learning, yet there is no systematic understanding of when each succeeds, when each fails, and when cross-modal training helps at all -- a gap that leaves practitioners, especially in scientific domains like biomedicine or astrophysics, with heterogeneous instruments and multiple levels of organization and measurement, unable to diagnose why standard methods underperform the best single modality. We develop a unified linear framework that addresses both questions. Under a spiked signal-plus-noise model with structured cross-modal nuisance correlation, we derive separation ratios for both objectives that expose complementary failure modes: alignment whitens each modality and fails when nuisance is strongly correlated across views; prediction encodes whatever is cross-predictable through a one-sided whitening, with recovery governed by source-modality quality. The resulting phase diagram partitions multimodal problems into four regimes: Both, CA only, CP only, and Neither. We present a data-driven procedure to locate real-world datasets in this diagram using a small labeled subsample, identifying the preferred objective and prediction direction before any cross-modal training. Experiments on synthetic data, stereo-vision benchmarks, image-caption pairs, and real astrophysical data validate the predictions in the nonlinear regime, including the Neither regime where cross-modal training is actively harmful. Our framework lets practitioners diagnose their multimodal problem and choose the right objective before committing to training. Code to reproduce the results is available at https://github.com/IlayMalinyak/mm_align_vs_pred.

Recent years have witnessed the rapid evolution of AI agents toward handling increasingly complex, real-world tasks. However, existing benchmarks rarely evaluate whether agents can operate graphical user interfaces to complete long-horizon, high-value professional workflows across diverse domains. Current GUI benchmarks still predominantly focus on general-purpose software, relatively simple applications, and short-horizon tasks, leaving it largely unknown whether modern agents can follow user instructions to autonomously operate domain-specific professional software and accomplish economically valuable work in an end-to-end manner. To bridge this gap, we introduce Workflow-GYM, a benchmark for long-horizon GUI tasks centered on professional domains and specialized software environments. Through extensive experiments on state-of-the-art models, we find that even the strongest models achieve only slightly above 30% success rates, highlighting that professional long-horizon GUI workflows remain highly challenging for current GUI agents. Further analysis reveals that current agents struggle to maintain long-horizon workflow consistency, frequently exhibiting workflow stage omission, error propagation, objective drift, and insufficient understanding of professional software environments. Our findings provide important insights into the limitations of current agent systems and suggest key directions for the next generation of GUI-agent research.

Reinforcement learning with verifiable rewards (RLVR) has become standard for improving LLM reasoning. However, existing PPO-style trust-region mechanisms remain position-agnostic by enforcing uniform thresholds across all tokens independently. This pointwise treatment conflicts with autoregressive generation in two critical ways. First, uniform thresholds ignore autoregressive asymmetry. Early-stage deviations produce compounding sequence-level drift, causing static thresholds to under-regulate early divergence and excessively constrain late-stage exploration. Second, evaluating token-level divergence in isolation overlooks cumulative prefix drift, granting the same divergence allowance regardless of how far the conditioning history has already deviated from the rollout policy. To address this limitation, we propose CPPO (Cumulative Prefix-divergence Policy Optimization), a token-level masking rule that aligns updates with a finite-horizon policy-improvement bound via two coupled mechanisms. First, a position-weighted threshold imposes stricter limits at early positions whose effects persist longer, relaxing constraints for late-stage tokens. Second, a cumulative prefix budget tracks historical deviations, dynamically restricting further token-level deviation to prevent compounding errors along the prefix. Empirically, CPPO enhances training stability and significantly improves reasoning accuracy across various model scales.

Warning: This paper contains several toxic and offensive statements. Modern large language models (LLMs) are typically aligned through large-scale post-training to ensure fair and reliable behavior. In this work, we investigate how easily such guardrails can be broken by Group Relative Policy Optimization (GRPO). We show that one-shot GRPO training on a single biased example is sufficient to induce systematic bias, with stereotype-driven reasoning generalizing across attributes, categories, and benchmarks. We further find that models differ in their susceptibility based on the initial likelihood of producing biased outputs. Our results reveal a critical vulnerability in post-training: alignment can be overridden by a single example.

Pre-trained language models (PLMs) have achieved strong performance in keyphrase extraction (KPE), largely due to their ability to generate rich contextualized representations. However, long-document KPE remains challenging because salient keyphrase evidence may be scattered across distant document sections that cannot be jointly captured within the limited context window of most PLMs. Although long-context large language models (LLMs) can process broader textual contexts, their computational cost limits their practicality for efficient and high-throughput KPE. To overcome this limitation, we propose an attention expansion mechanism that augments PLM token representations with information from surrounding out-of-context chunks using pre-trained word embeddings. The proposed mechanism expands the effective contextual scope of PLM-based KPE models without requiring full-document attention or expensive LLM-based inference. We evaluate our approach across five PLM backbones, including general-purpose, scientific, task-specific, and long-context encoders, using two training regimes and five benchmark corpora from scientific and news domains. Experimental results demonstrate that attention expansion consistently enhances KPE performance across all evaluation settings, outperforming state-of-the-art models and yielding notable improvements in F1 score. The improvements extend to domain-specific, task-specialized, and native long-context models, showing that the proposed mechanism provides complementary information rather than merely compensating for limited input length. These results establish attention expansion as an efficient and effective strategy for long-document KPE.

Dexterous hands are essential for fine-grained manipulation, but their hardware designs vary substantially across embodiments. Differences in kinematics, joint definitions, and degrees of freedom make it difficult to define a shared state representation compared with parallel grippers. As a result, dexterous-hand data remains fragmented and difficult to use for joint training. In this work, we propose the Unified Dexterous Hand Model (UDHM), which maps human and robot hand states into a shared 22-DoF semantic interface. Based on UDHM, we introduce UniDexTok, a retargeting-free state tokenizer that learns embodiment-conditioned discrete tokens from standardized real joint states. UniDexTok provides a unified representation for heterogeneous dexterous hands without relying on retargeting or simulation data. Compared with the recent baseline UniHM, UniDexTok reduces MPJAE from 15.63 degrees to 0.16 degrees and MPJPE from 18.51 mm to 0.18 mm, corresponding to error reductions of 98.98% and 99.03%, respectively. These results improve reconstruction from centimeter-scale to sub-millimeter accuracy. Experiments further show that data from other embodiments improves target-embodiment reconstruction accuracy, demonstrating the benefit of cross-embodiment tokenization. UniDexTok also shows strong zero-shot and few-shot reconstruction ability when new dexterous hands are introduced.

Transformer-based models have emerged as leading paradigms in time-series forecasting in recent years, employing self-attention mechanisms to capture long-range dependencies. Despite their success, these single-stage forecasting architectures exhibit persistent systematic residual biases arising from structural discrepancies, unmodeled stochastic components, or inadequate multi-scale temporal representations. This limitation persists when residuals are treated as irreducible noise, precluding adaptive correction of structured error patterns. To address this limitation, we introduce a two-stage, model-agnostic framework that explicitly decouples forecasting and residual learning into distinct stages of representation learning. A base transformer first generates the initial predictions. Subsequently, a dedicated meta-corrector dynamically models structured error patterns across multivariate channels, preserves cross-variable dependencies, and iteratively refines the residual bias of the base transformer. By formalizing this pipeline as a hypothesis space expansion, our framework addresses approximation limitations inherent in single-stage architectures, removes reliance on restrictive assumptions, and enables end-to-end learning of complex error dynamics. Evaluated on eight popular benchmark datasets using established protocols, our approach achieves state-of-the-art performance, with significant improvements in standard metrics (MSE, MAE). The results demonstrate the framework's ability to mitigate systematic biases and enhance robustness to complex temporal dynamics, advancing the practical applicability of transformer-based forecasting models.

Machine learning weather prediction (MLWP) models have achieved impressive forecasting performance at a small fraction of the computational costs required for traditional physics-based methods. However, they are primarily (1) data-driven and (2) evaluated using pixel-wide error metrics (e.g., RMSE), so there are no guarantees that their forecasts are consistent with known physical laws. We introduce PhysMetrics$.$Weather, an evaluation framework that assesses the physical realism of MLWP models across three types of metrics: conservation, spectral, and dynamical. By quantifying physical realism, this tool guides the development of physics-informed architectures and helps evaluate whether MLWP models are reliable for operational use. Our framework is available on Github at https://github.com/Emmakast/PhysMetrics.Weather.

Long-horizon language agents accumulate observations, reasoning traces, and retrieved facts that exceed their finite context windows, making memory retention a fundamental resource-allocation problem. Existing memory systems improve management through heuristic scoring, retrieval optimization, or learned compression, but largely treat retention as a local decision problem and do not explicitly model its long-term consequences under realistic observability constraints. To fill this gap, we formulate memory retention as a constrained stochastic optimization problem with explicit budget feasibility, evidence utility, and delayed costs including miss penalties, reacquisition delays, and stale-information risk. We then propose OSL-MR (Observability-Safe Learning for Memory Retention), a novel framework that enforces a strict separation between online-observable features and offline-available supervision (OAS). OSL-MR combines an evidence learner trained from realized evidence supervision with a Mixed-Score heuristic that serves both as a deployable online-safe baseline and as a structured inductive prior for learning. The resulting policy learns query-conditioned evidence value directly from interaction data while remaining deployable under the same observability constraints. Experiments on LOCOMO and LongMemEval show that OSL-MR consistently outperforms recency-based methods, Generative Agents-style scoring, and other heuristic baselines, particularly under tight memory budgets. The Mixed-Score prior further improves precision while preserving recall, and sensitivity analysis demonstrates robustness across a wide range of cost configurations.

Math reasoning benchmarks have proliferated, yet most lack a per-item difficulty signal grounded in actual human performance. We introduce KCSAT-ML, a decade (2014-2025) of Korean College Scholastic Ability Test (KCSAT; Suneung) mathematics: 664 problems with a 339-item core set carrying official per-item error rates from nationwide cohorts of hundreds of thousands of examinees. We pair the benchmark with Difficulty-aligned Reasoning Gain (DRG): a score-orthogonal metric that asks whether a model's mistakes concentrate on the items humans found hard, or on items humans found easy. Together they expose, across a wide range of VLMs (and LLMs via OCR), three patterns: (i) low-budget accuracy collapses on the high-human-error tail at every model size; (ii) test-time scaling (TTS) raises token use roughly linearly with cohort error rate, while accuracy gains follow a non-monotonic curve; (iii) within a single family, TTS flips between anti-scaling on the hardest items and overthinking on easier ones -- two faces of the same alignment failure. On DRG, models with near-identical accuracy can sit at near-opposite values: one model gets wrong what humans also find hard, while another solves the hardest items yet fails on items humans find easy -- a contrast that aggregate accuracy hides. Our code and dataset builder will be open-sourced at https://github.com/naver-ai/KCSAT-ML.

An improved GAN-based imaging logging image restoration method is presented in this paper for solving the problem of partially missing micro-resistivity imaging logging images. The method uses FCN as the generative network infrastructure and adds a depth-separable convolutional residual block to learn and retain more effective pixel and semantic information; an Inception module is added to increase the multi-scale perceptual field of the network and reduce the number of parameters in the network; and a multi-scale feature extraction module and a spatial attention residual block are added to combine the channel attention. The multi-scale module adds a multi-scale feature extraction module and a spatial attention residual block, which combine the channel attention mechanism and the residual block to achieve multi-scale feature extraction. The global discriminative network and the local discriminative network are designed to gradually improve the content and semantic structure coherence between the restored parts and the whole image by playing off each other and the generative network. According to the experimental results, the average structural similarity measure of the five sets of imaged logging images with different sizes of missing regions in the test set is 0.903, which is an improvement of about 0.3 compared with other similar methods. It is shown that the method in this study can be used for the restoration of micro-resistivity imaging log images with good improvement in semantic structural coherence and texture details, thus providing a new deep learning method to ensure the smooth advancement of the subsequent interpretation of micro-resistivity imaging log images.

Flow-matching transformers achieve strong audio separation, yet their attention dynamics are opaque. We adapt established causal-intervention principles into a deterministic, inference-time probing protocol for SAM Audio. Orthogonal probing uncovers a dual-pathway text-conditioning mechanism: additive injections control semantic identity, while cross-attention refines acoustic structure. We observe an asynchronous layerwise convergence: stable layers build temporal scaffolds early, whereas fast layers continue resolving artifacts during sampling. The model also attenuates temporal segmentation cues to maintain continuous-flow stability. Using these insights, we propose Layer-Selective Attention Caching (LSAC), a training-free acceleration method that caches attention in stable layers. Across acoustic complexities, LSAC cuts self-attention computation by about ~25% with negligible quality loss and yields up to 6.7x higher quality retention than naive step reduction.

We study feed-forward ReLU networks with fixed readout and quadratic loss. The aim is to rewrite gradient descent not primarily as a dynamics in weight space, but as a collective dynamics closed in terms of fields defined on the training-set space. For a single hidden layer, the weight variables can be eliminated from the activation dynamics, yielding a closed equation for the residuals governed by a collective kernel that factorizes into an input-geometric matrix and a dynamical co-activation matrix. For deeper networks, the residual dynamics retains a clean layer-wise kernel structure. However, from depth three onward, closure requires a hierarchy of weight-induced Gram operators that mediate information transport across layers. Moreover, the conjugate-field dynamics is governed by operators satisfying a backward pullback recursion, of which the weight-induced Gram operators are the first nontrivial instances.

As autonomous research agents and AI co-scientist systems push large language models (LLMs) from drafting toward end-to-end manuscript production, the bottleneck shifts from generation to verification. Fluent LLM output can hide fabricated citations, numbers that drift from source tables, and unmet reporting-guideline items; existing tools generate without verifying, and self-critique inherits the blind spots that produce confident fabrication. We describe an architecture pairing generation with verification, resting on three principles: decompose the workflow into self-contained skills, gate every stage transition with halt-on-failure, and resolve each integrity question with the cheapest sufficient mechanism, a deterministic, re-executable check where one suffices and a prose-level probe only where interpretation is unavoidable. This determinism-where-possible split, organized as an integrity-gate taxonomy, is the core contribution. It is realized as MedSci Skills, an open-source toolkit of 43 skills with a 21-detector deterministic tier, evaluated on three public-dataset pipelines (STARD, PRISMA, STROBE) and a seeded-defect ablation. Across the three pipelines every content-hash manifest verified clean and the gates surfaced real defects; on 27 identical injected defects the deterministic gates detected all 27 with no false positives on the matched clean fixtures, whereas a single-prompt LLM reviewer detected 11, its misses in code, bibliography, and style defects the prose hides. Determinism-where-possible verification yields an auditable, re-executable trail that exposes the evidence a human needs to check an LLM-assisted manuscript: feasibility and reproducibility evidence, not a claim of human-competitive quality, which a separate blinded study addresses. MedSci Skills is MIT-licensed and archived (v3.8.0).

Computer-use agents (CUAs) increasingly operate in runtimes that combine visual desktop control, command-line execution, code editing, browsers, and external tools. Existing benchmarks, however, often evaluate these interfaces as separable capabilities, leaving long-horizon cross-interface orchestration under-tested. Thus, we introduce WeaveBench, a long-horizon hybrid-interface benchmark with 114 tasks across 8 real-world work domains, grounded in real user requests and publicly verifiable artifacts. Each task requires agents to combine GUI observations/actions with CLI/code operations within a single trajectory. We evaluate these tasks on a real Ubuntu desktop inside deployed CLI-agent runtimes, augmented with a minimal desktop-control plugin. We also propose a companion trajectory-aware judge that inspects deliverables, files, screenshots, logs, and action traces, while detecting shortcut behaviors such as fabricated visual evidence or hard-coded metrics. Across frontier model-runtime pairings, the best PassRate reaches only 41.2%, showing the benchmark remains far from saturated. The trajectory-aware judge further reveals that outcome-only grading substantially overestimates agent performance. Overall, WeaveBench exposes a critical gap in CUA evaluation and provides an effective testbed to measure whether agents can orchestrate GUI, CLI, and code operations across long-horizon real-world tasks.

Medical agent systems are increasingly expected to support interactive clinical decision making rather than only static question answering. In such settings, effective agents must reuse prior experience across evolving cases, yet existing memory mechanisms often retain raw historical traces that are redundant, noisy, and difficult to govern. More importantly, they rarely distinguish which memories are truly useful for future reasoning. This limits their ability to accumulate compact and reliable experience for long-horizon clinical reasoning. To close this gap, we propose SkeMex, a post-deployment self-evolution framework that improves medical agents through a skill-based memory without updating model weights. SkeMex distills informative interaction trajectories into structured skills that encode reusable procedural knowledge, and organizes them into a multi-branch repository spanning general, task-specific, and action-level experience. To determine which memories should be reused and retained, SkeMex estimates context-dependent utility from environment feedback and uses it to guide value-aware retrieval and repository governance. A closed-loop ``Read--Write--Assess--Govern" lifecycle further supports continual evolution by writing new skills, updating utilities, promoting useful memories, and removing harmful entries. Experiments across diverse clinical tasks show that SkeMex consistently outperforms representative memory-based agents in both offline and online settings. It also generalizes across model backbones and supports transferable skill memory. All data and code will be released publicly.

Trusted Execution Environments (TEEs) have enabled confidential Byzantine Fault-Tolerant (BFT) consensus systems with confidentiality and improved scalability. However, TEEs do not provide state continuity: during recovery, a compromised host can roll back a crashed enclave to a stale persistent state, significantly threatening both safety and availability. Existing defenses face a fundamental tradeoff: they either impose substantial overhead on critical consensus paths, reducing throughput and increasing latency, or incur prolonged recovery delays, hurting availability. We present the first systematic taxonomy of rollback-resilient recovery for confidential BFT consensus, distilling prior approaches into four categories. We further expose their inherent limitations. Guided by this detailed analysis, we design CHIMERA, a protocol-aware recovery framework that breaks this tradeoff. Our key insight is that rollback protection in consensus systems should not be uniform. Different types of persistent states differ fundamentally in their state distribution, update behavior, and representation form. CHIMERA separates persistent state into metadata and logs according to these protocol-level properties and applies distinct recovery mechanisms to each type. We formally model CHIMERA in Maude and verify its safety and liveness properties. We implement it on Braft and ZooKeeper using Intel TDX, and evaluate it in both LAN and WAN settings. Results show that CHIMERA achieves higher throughput, lower recovery latency, and better availability than state-of-the-art rollback-resilient baselines.

Reinforcement learning from human feedback (RLHF) is bottlenecked by reward hacking, where the policy exploits errors in a proxy reward model (RM) and produces high RM scores without genuine quality gains. A natural mitigation is pessimism: lowering rewards in regions where the RM is uncertain. However, standard scalar RMs provide no principled notion of uncertainty. We argue that the right object is a distributional reward model $p(r\mid x,y)$. Under either a Bayesian inference or a KL-distributionally robust optimization (KL-DRO) lens, the KL-regularized RLHF objective admits a closed-form effective reward $\tilde r(x,y) = \pmβ\log\mathbb{E}_p[e^{\pm r/β}]$. The pessimistic branch unifies the prior heuristics for RM ensemble aggregation: mean aggregation, worst-case optimization (WCO), and uncertainty-weighted optimization (UWO) all emerge as limits or truncations of this single expression. This also clarifies the implicit assumptions of each existing rule.

Understanding how microscopic structural dynamics relate to macroscopic mechanical response during yielding remains a central challenge in soft matter physics. Here, we introduce rheo-echo X-ray photon correlation spectroscopy (rheo-echo-XPCS) with nonlinear acquisition synchronized to oscillatory shear, enabling direct measurement of irreversible nanoscale dynamics under strain amplitude control. Applying this to a carbon black colloidal fractal gel, we resolve time-periodic echoes in the vorticity-direction intensity autocorrelation function whose decay encodes non-affine structural rearrangements. We find: (i)~ballistic-like decorrelation with $τ\propto q^{-1}$ at all strains, where the decorrelation velocity $v_τ= 1/\langle qτ\rangle$ scales linearly with the loss tangent $\tanδ= G''/G'$, establishing $\tanδ$ as a direct macroscopic signature of the rate of irreversible structural decorrelation; (ii)~functional form continuous evolution from compressed exponential ($α\simeq 1.5$) at low strain, consistent with three-dimensional dipolar strain fields in the intact network-to stretched exponential ($α\simeq 0.5$) at high strain, reflecting a dimensional reduction from $d_f = 3$ to $d_f = 1$ as stress transmission shifts from bulk to quasi-one-dimensional filamentary backbones during network fragmentation.

Phonon polaritons, quasiparticles formed by the resonant hybridization of light and lattice vibrations, exhibit unique properties like the possibility of hyperbolic dispersions. In this context, exfoliated hexagonal boron nitride (hBN) has emerged as a promising material for phonon polariton-based research. However, to advance toward practical applications, it is essential to demonstrate efficient phonon-polariton propagation and confinement in large-area epitaxial BN. To address this topic, we use metalorganic vapor phase epitaxy (MOVPE)-grown BN and investigate the phonon polariton properties using scattering-type scanning near-field optical microscopy (s-SNOM) and nanoscale Fourier transform infrared spectroscopy (nano-FTIR). We report remarkably long phonon polariton propagation lengths, indicating the high crystalline quality of the BN layer. By using epitaxially grown isotopically pure triangular islands, we further demonstrate efficient phonon-polariton confinement with mode patterns tunable by the incident light wavelength. Our results pave the way for implementing all-epitaxial, microscale, high-quality polariton resonators for nanophotonics and quantum optics.

The exploration of novel multiferroic materials with strong coupling between ferroelectric polarization and photovoltaic effects is crucial for next-generation optoelectronic devices. In this study, we characterized highly oriented 0.5Pb(Zr0.52Ti0.48)O3-0.5Pb(Fe0.5Nb0.5)O3 multiferroic thin films grown by pulsed laser deposition on SrTiO3 (001) substrates with a SrRuO3 bottom electrode. The films exhibited excellent crystalline quality, with a single perovskite phase and (001) orientation. They displayed good ferroelectric properties (remanent polarization $\sim$17 $μ$C/cm$^2$, PUND; coercive field $\sim$150 kV/cm), alongside weak ferromagnetic behavior at room temperature (remanence 1.30 emu/cm$^3$; coercive field 90 Oe). Photovoltaic measurements demonstrated a robust, polarization-dependent photoresponse under 403 nm monochromatic laser illumination, achieving Jsc values between $\pm$20 $μ$A/cm$^2$. This compelling observation confirms the intrinsic coupling between ferroelectric polarization and photovoltaic effects, highlighting the considerable promise of these single-phase multiferroic thin films for advanced photovoltaic and optoelectronic memory applications.

Exceptional points (EPs) are non-Hermitian degeneracies at which two or more eigenvalues and their corresponding eigenvectors coalesce. In open quantum systems, exceptional points can arise in the Lindbladian governing the dissipative dynamics. Their signatures have so far been mainly identified in finite-time observables, such as transient currents, while steady-state average currents generally provide no direct evidence of the underlying exceptional-point structure. In this work, we demonstrate that signatures of Lindbladian EPs can nevertheless be accessed in the steady-state regime through current noise. We derive general expressions for current correlation functions within a Lindblad master-equation framework and show, in particular, how exceptional points affect their behaviour as a function of the time delay. We illustrate these results with the paradigmatic example of two interacting qubits coupled to two reservoirs, where the steady-state noise clearly distinguishes overdamped, underdamped, and critical regimes. Our results establish current correlation functions as a steady-state probe of Lindbladian EPs in open quantum systems.

Quantum mechanically, an electron moving in a uniform magnetic field forms Landau levels. A curious feature is that for states with a negative angular quantum number, the total probability current vanishes, which appears to contradict the classical picture of cyclotron motion. While a geometric interpretation based on classical orbits exists, alternative interpretations remain of interest. In this paper, we examine the probability current density and identify a critical radius that naturally partitions the plane into an inner clockwise-flow region and an outer counterclockwise-flow region. We show that the vanishing total current results from an exact cancellation between these two regions. Furthermore, by defining a partitioned kinetic angular momentum with respect to the critical radius, we reveal an intrinsic competitive structure: the electron simultaneously carries two opposing rotational components. The negative quantum number manifests in the strength of the inner counter-rotation, while the net kinetic angular momentum remains positive. This bidirectional flow picture also provides a dynamical interpretation of the infinite degeneracy of Landau levels.

The parent compound FeTe hosts a complex magnetic landscape that is highly susceptible to lattice distortions. Although theoretical models have predicted a bicollinear to dimer antiferromagnetic (AFM) phase transition under tensile strain, its experimental realization and deterministic control has remained elusive owing to severe magnetic frustration. Here, combining high-resolution scanning tunneling microscopy (STM) and density functional theory (DFT) calculations, we demonstrate the selective stabilization of bicollinear and dimer AFM orders in few-layer FeTe films via local uniaxial strain engineering. By mapping the strain fields near dislocation areas in FeTe films and FeTe/FeSe heterostructures, we establish a direct correspondence between specific strain components and the resulting magnetic ground states. We find that uniaxial compression along the Fe-Fe next-nearest-neighbor direction stabilizes the bicollinear AFM order, with the stripe orientation aligning parallel to the compression axis. Crucially, we report the experimental realization of the long-range dimer AFM order, which emerges under anisotropic strain along the Fe-Fe nearest-neighbor direction. This phase manifests as a distinct $\sqrt{2} \times \sqrt{2}$ electronic reconstruction and shares a common Neel temperature with the bicollinear phase. Our findings reveal that anisotropic strain effectively lifts the magnetic degeneracy among competing states. This work provides a robust strategy for the manipulation of elusive magnetic orders and offers insights into the interplay between lattice, spin, and electronic degrees of freedom in iron-based superconductors.

Directional solidification of bubbly liquids plays a critical role in shaping the microstructure and properties of many materials, yet the fluid dynamics governing bubble behavior during solidification remain poorly understood. Using cryo-confocal microscopy and particle image velocimetry, we quantify fluid flows around bubbles during solidification of water containing surfactants and tracers. Our results reveal that volumetric expansion dominates fluid motion, with velocities scaling linearly with the solidification rate (1-20$~μm/s$), while Marangoni flows-hypothesized to play a key role-are negligible ($< 5~μm/s$) under our experimental conditions. Diffusiophoresis and thermophoresis also contribute minimally. These findings challenge existing theoretical models and provide a framework for controlling bubble distribution in solidified materials

We study the interference of critical dynamics associated with zero modes (ICDZM) in the generalized Creutz ladders using closed quench paths that pass through two critical points successively. By reading out the final zero-mode transfer probability, we find rich ICDZM interference patterns dependent on the quench path. In particular, when the closed path links two topologically nontrivial phases, the ICDZM pattern may either vanish or exhibit period doubling. Within the framework of WKB analysis, this phenomenon is well clarified by the interference phase accumulated in the quench procedure. We also demonstrate that the zero-mode transfer probability can be detected by the deviation of the boundary particle number from its initial fractional value, which arises from the blending of bulk modes in the critical dynamics. As an edge defect, the zero-mode transfer probability captures both the ICDZM oscillation and the known anomalous defect production in a non-closed quench path. These results identify ICDZM and the corresponding edge defect as probes for critical dynamics associated with topological zero modes.

We experimentally investigate the magnetic gyrotropic mode in a system of vortex ferromagnetic (FM) nanooscillator exchange coupled to an antiferromagnetic (AFM) layer. The micron-sized disks formed from the Ni80Fe20(12 nm) / Ir80Mn20 (5 nm) FM/AFM heterostructure are prepared so that the vortex magnetic state is imprinted into the AFM layer. We apply a magnetic resonance force microscopy (MRFM) method to locally study magnetic oscillations in single FM/AFM disks. We show that the gyrotropic mode frequency is significantly (approximately 4 times) shifted to the high frequency compared to similar structure consisting of a single ferromagnetic disk. Upon applying an in-plane magnetic field, we observe a strong peak at a double frequency which was previously predicted in theory. This is governed by the fact that vortex dynamics is strongly nonlinear in a noncentrosymmetric system under investigation.

Intrinsically chiral metal surfaces, where handedness arises from the asymmetric step-kink-terrace topology of high-Miller-index planes, are model systems for enantiospecific catalysis, sensing, and spintronics. Yet, no consistent method exists to classify their handedness directly from experimental observables. We report a dual-domain machine learning framework that decodes crystallographic surface chirality from two independent image representations: atomic structure models in real space and simulated momentum-resolved photoemission maps of the Fermi surface projections in reciprocal space. ResNet18, a deep convolutional neural network, fine-tuned on a database of labeled images achieves ~73% classification accuracy on atomic models and ~99% on Fermi surface projections. We show that the latter transfers directly to synchrotron-acquired experimental images after fine-tuning on just two labeled frames. We identify a working correspondence between the two representations: just as the kink site geometry fixes the orientation of crystallographic planes in real space, the surface normal position in a momentum-resolved photoemission map anchors the orientation of the Fermi surface polygons in reciprocal space. It is precisely this relative orientation that encodes handedness into the map topology with high accuracy. The pronounced difference in accuracy shows that handedness is more readily recovered from the momentum-space electronic pattern than from the local atomic geometry of the kinked surface. This finding has direct implications for the disorder resilience of geometric chiral-induced spin selectivity (CISS) at realistic metal surfaces.

We prove the existence of Long-Range Order in a class of two-dimensional Euclidean lattice Gross-Neveu models with an even number of fermion flavors, covering three standard lattice discretizations, including naive and staggered fermions widely used in numerical studies. By performing a Hubbard-Stratonovich transformation, we map the fermionic systems to bosonic ones and establish Reflection Positivity for the resulting measures. Exploiting this structure, we combine Chessboard Estimates with a Peierls-type contour argument to prove Long-Range Order for the chirally charged fermion-mass bilinear $\overlineψψ$ at sufficiently small coupling and sufficiently large flavor number. Our analysis is robust with respect to the choice of lattice discretization and applies uniformly across different realizations of the same underlying continuum model. Moreover, we obtain uniform pointwise bounds on the bosonic two-point function, equivalently on the fermionic mass-mass correlator, showing that it is quantitatively controlled by the minimizers of the effective potential. This provides a fully rigorous and non-perturbative demonstration of Long-Range Order in lattice Gross-Neveu models and establishes a direct connection between the rigorous theory and its large-$N$ (mean-field) predictions.

Ultracold atoms in multimode optical cavities provide a rich testbed for many-body phenomena enabled by light-mediated interactions. Recent experiments include realizations of spin glasses and associative memories, as described by multimode Dicke models with disordered couplings. However, the properties of multimode Dicke models with ordered coupling geometries remain largely unexplored. In this work, we investigate the stable steady-states of the multimode Dicke model with an ordered nearest-neighbor coupling geometry, where $n_c$ atomic clusters are coupled via $n_c-1$ cavity modes. We show that the number of mean-field stable steady-states in the superradiant phase exhibits Fibonacci scaling with the number of atomic clusters, and that a subset of these steady-states exhibit persistent oscillations. Using both the truncated Wigner approximation and the numerically-exact hierarchy of pure states, we further demonstrate that these features of the stable steady-state solutions persist for finite cluster sizes. Ordered multimode Dicke models, such as the nearest-neighbor coupling geometry considered here, are accessible with current experimental technologies and point toward a broader class of strongly interacting dissipative systems with similarly rich behavior.

Dense granular flows made of elongated particles exhibit a strong inhibition of particle rotation compared to spherical grains, but the mechanisms responsible for this effect remain unclear. Using three-dimensional discrete element simulations, we investigate the angular dynamics of elongated particles in dense, confined shear flows. We systematically vary particle aspect ratio, interparticle friction, and boundary conditions to elucidate their respective roles. We show that the reduction of the average angular velocity cannot be attributed to particle shape, friction, or solid fraction alone. Instead, it is controlled by the degree of collective alignment developed under shear, quantified by a nematic order parameter. Based on this observation, we propose a simple scaling law linking the average angular velocity to the local shear rate through a hampering parameter that depends solely on the orientational order via the nematic order parameter. This scaling successfully collapses data obtained for different particle properties (shape, friction), different flow patterns, and, remarkably, remains valid for two additional flow configurations.

Most conventional micromagnetic frameworks in spin caloritronics rely on a unidirectional coupling approximation, wherein thermal fluctuations drive magnetization dynamics while the feedback of magnetic dissipation onto the thermal reservoir is neglected. Here, we establish a rigorously self-consistent bidirectional magneto-thermal coupling model by integrating the stochastic Landau-Lifshitz-Gilbert (sLLG) equation with a generalized heat transfer equation. In this closed-loop framework, the local temperature acts as a dynamical variable, and the damping-induced dissipation alongside stochastic work dynamically feeds back into the thermal bath as localized heat sources. Utilizing Ito stochastic calculus, we analytically prove that this coupled system strictly obeys the first law of thermodynamics and spontaneously recovers the correct Boltzmann statistics at equilibrium. Spatially resolved micromagnetic simulations further validate the energy exchange mechanism, capturing the finite-bath temperature reduction induced by spatial variation of magnetic moments and the modified density of states under exchange interactions. This bidirectional framework provides a robust microscopic foundation for investigating complex nonequilibrium magneto-thermal dynamics, such as the unidirectional spin-wave heat conveyer effect, paving the way for advanced spin-caloritronic applications.

We design the first neural quantum state for continuum particles that, for any chosen allowed momentum $\mathbf{k}$, is by construction an exact eigenstate of total momentum with eigenvalue $\mathbf{k}$. Our architecture, EVE, enables off-the-shelf VMC to solve for momentum-sector ground states. We test EVE on 2D bosons with mutual $1/r$ interactions, finding that a single unified ansatz is capable of describing four qualitatively different states: superfluid, roton, crystal, and phonon. At different densities, we extract the underlying phase of matter from the dispersion's shape. At $r_s = 20.0$, we see the roton minimum at finite $k$ expected of a superfluid. At $r_s = 100.0$, we see striking zone folding indicative of crystalline order, with periodically spaced minima representing floating crystals connected by phonon arcs in between. Using density-density correlation functions, we confirm the phase diagnoses and probe the excitations' correlation structures. Finally, we analyze the roton's phase texture and find unexpected multi-particle phase strings, formed when several vortex dipoles merge, leaving two vortices connected by a phase slip.

Using the Keldysh non-equilibrium Green's function approach, we investigate charge pumping through a single-stranded helical structure described by a tight-binding model that includes either short-range hopping (SRH) or long-range hopping (LRH). While quantum pumping has been studied in various low-dimensional systems, the detailed behavior of the spectral current and the pumped dc current in helical geometries in the presence of higher-order electron hopping (beyond nearest neighbors) has not yet been systematically explored. Here, we focus on the interplay between helicity and extended hopping ranges, analyzing how they jointly control the energy-resolved and dc pumped currents under time-periodic end potentials. For LRH, the pumped dc current exhibits pronounced plateau-like regions as a function of chemical potential when energy levels are sparsely spaced -- consistent with adiabatic transport -- whereas SRH yields more parameter-sensitive currents without clear plateaus. The plateau stability is controlled by the drive frequency: at higher frequencies, Floquet side-band mixing destroys the plateaus, leading to oscillatory currents. The phase dependence remains nearly sinusoidal, and the current vanishes at zero phase lag, confirming the necessity of out-of-phase potentials. Crucially, in helical systems, the decay exponent $(\ell_c)$ acts as an effective structural parameter that can tune both the magnitude and sign of the pumped current, offering a geometric knob for controlling quantum pumping. Our findings not only fill a gap in the understanding of spectral and pumped currents in helical systems with extended hopping but also provide tools that can be applied to analyze similar phenomena in other chiral or quasi-one-dimensional systems.

We study the controllability of a system of qubits under global control, where control pulses act identically on all qubits. Specifically, we consider a collection of qubits identically coupled to a single bosonic mode, or harmonic oscillator, via the Jaynes-Cummings interaction. This collective coupling, known as the Tavis-Cummings (TC) interaction, has been realized in several quantum computing platforms, including superconducting and atomic qubit systems. Although the qubits do not interact directly with one another, they can become entangled through their common coupling to the bosonic mode. We characterize the group of unitaries that can be implemented on the joint Hilbert space of the qubits and bosonic mode using the TC interaction together with a global $z$ field $J_z$, corresponding to identical z rotations on all qubits. We show that for n>2 qubits the set of realizable unitaries is restricted by an "accidental" symmetry of the TC Hamiltonian, distinct from its "standard" U(1) and permutational symmetries. On the other hand, we find that the Hamiltonian $J_z^2$ breaks this accidental symmetry and, together with the TC interaction and $J_z$, achieves semi-universality: it allows the implementation of arbitrary unitaries that respect permutational and U(1) symmetry, up to certain constraints on the center of the group. In a companion paper, we further analyze this remarkable accidental symmetry and show that it can be understood through Schwinger's bosonic model of angular momentum.

The Jaynes-Cummings (JC) Hamiltonian is a paradigmatic model of light-matter interaction and, more generally, qubit-boson interactions, widely used across atomic, optical, and superconducting qubit platforms. In the multi-qubit setting, where n qubits are identically coupled to a single boson mode, this interaction is known as the Tavis-Cummings (TC) Hamiltonian. The structure of the TC model is usually understood in terms of two standard symmetries: permutation invariance of the qubits and a U(1) symmetry associated with conservation of the total excitation number. Here we identify an additional, independent "accidental" symmetry of the TC Hamiltonian and construct the corresponding conserved observable. We show that, for n>2 qubits, this symmetry imposes strong constraints on the realizable unitary transformations. These constraints persist in the presence of the global $J_z$ Hamiltonian, but are removed by adding $J_z^2$, even though $J_z^2$ preserves both permutation invariance and the U(1) symmetry. Finally, we explain the origin of this previously unnoticed symmetry using Schwinger's boson representation of angular momentum. These restrictions have important implications for controllability of the TC system and for its applications to quantum computing, which are investigated further in a companion paper.

Proofreading is generally thought to improve accuracy at the expense of speed. We show that this trade-off can be reversed in stochastic processes with long-lived stalled states. Using a non-Markovian renewal framework, we derive exact expressions for the error rate and completion time under proofreading for arbitrary stall-time distributions. Our analysis reveals that fluctuations in stall durations, rather than their mean alone, determine whether proofreading can simultaneously increase speed and accuracy. In the limit of strong stalling, this regime emerges when the coefficient of variation of the stall time exceeds a threshold set by the intrinsic error rate. These results provide a general criterion for proofreading in systems ranging from self-assembly and polymer replication to immune recognition and other nonequilibrium information-processing systems.

Thermal transport in long polymer molecules is commonly attributed to ballistic propagation of long-wavelength acoustic phonons, which act as Goldstone modes arising from translational symmetry, while the transport of other phonons is suppressed by Anderson localization. This mechanism leads to thermal conductivity that increases with molecular length. Consistent with this picture, strongly aligned polymers exhibit exceptionally high thermal conductivity, whereas poorly aligned polymers are orders of magnitude less conductive and function as thermal insulators. Here we show that this strong sensitivity to molecular alignment originates from phonon scattering by molecular kinks. Even in the long-wavelength limit, the kink scattering remains strong because kinks break translational symmetry both for longitudinal and transverse phonons. As a result, randomly oriented kinks cause a rapid decrease in thermal conductivity with increasing molecular length. These findings identify alignment control by means of kink engineering as a route for tuning thermal transport in polymers.

We report measurements of the thermopower of atomic-size gold contacts realized by the mechanically controllable break junction (MCBJ) technique over a temperature range from 18 K to 295 K. A thermometer included in the lithographic structure close to the constriction provides a direct measurement of the temperature increase generated by heating one side of the contact with a focused laser beam. While the conductance histograms confirm the quantum nature of the transport, we observe a nonmonotonic temperature dependence of the ensemble-averaged thermopower with a minimum of $-2\,μ$VK$^{-1}$ at about 150 K. The values for the thermopower obtained at the lowest and the high temperature are compatible with values reported in the literature, but the nonmonotonic behavior in between disagrees with the expected linear dependence for quantum coherent conductors described by the Landauer formula. We develop a theoretical model based on an energy dependent transmission function that qualitatively reproduces the nonmonotonic behavior, but fails quantitatively. We therefore interpret our data as a result of phonon contributions to the thermopower beyond the Landauer model and with opposite sign than the classical phonon drag known from bulk systems. Our findings show that, firstly, the thermopower gives important insight into the transport properties of atomic-size structures and second that the linear approximation of the Landauer model has to be used with caution when studying more complex transport properties even for atomic contacts from free-electron metals.

This work is motivated by the experimentally observed coherence of the supercurrent in semiconductor nanowires covered by a half-shell metallic superconductor, which leads to flux dependent supercurrent oscillations with period h/2e, as expected for a tubular superconductor, i.e. Little-Parks oscillations. We perform microscopic model calculations and compare the results for full and half metallic shells. We use an effective Hamiltonian derived from the Green's function of the proximitized semiconductor nanowire, where the presence of the superconductor is represented by a self energy. Furthermore, we incorporate the electrostatic band-bending at the metal-semiconductor interface as a rectangular narrow quantum well on the semiconductor side. The properties of the eigenstates of the effective Hamiltonian are determined by the spatial profile of the corresponding transverse modes in the normal state. For half-shell wires, transverse modes with high-enough energy expand outside the interface quantum well and generate eigenstates with mixed electron-hole character that surround the entire circumference of the nanowire, similar to eigenstates of the full-shell system. We identify these states as being responsible for the observed Little-Parks effect.

We show that, under a short optical pulse, the quantum metric of Bloch states in the momentum-time (kx, ky , t) of graphene becomes dynamic and exhibits a wave-like behavior near Dirac points. This quantum metric wave reflects the Floquet-band structure caused by the pulse, as revealed by solving the time-dependent Schrödinger equation assuming that correlations and out-of-equilibrium effects can be ignored. The momentum and temporal components of the metric have very distinct time dependence that persists even after the pulse has passed. In addition, the pulse also generates a Berry curvature wave that is otherwise absent in static graphene. The time-dependent electron densities in conduction and valence bands also give arise to a Fisher information wave that constitutes part of the quantum metric wave, and is readily measurable by pump-probe experiments.

Wave turbulence describes the statistical dynamics of dispersive waves with weakly nonlinear interactions. While the classical kinetic equation captures the mean evolution of the wave spectrum, the study of its fluctuations due to finite-size effects and intermittency requires a probabilistic framework for space-time trajectories of the spectrum dynamics. Following the previous large deviation theories for wave turbulence, we develop a simplification meant for qualitative and numerical predictions of measurable quantities. We derive a new large deviation principle in the case of local wave interactions. It fully characterizes typical and rare fluctuations of the spectrum. In a joint article, we obtain a theory which is a generalised form of Macroscopic Fluctuation Theory, but with 2 conserved quantities (mass and energy). In this paper, we use it to analyse the structure of the equation for Gaussian fluctuations around out-of-equilibrium spectra. In addition to the usual equilibrium contribution, we obtain long-range correlations, which can be decomposed into 3 contributions: one is driven by the flux in the bulk and another is driven by the forcing and its possible fluctuations. In addition, these contributions are computed for the first time with a method adapted to boundary conditions where only the fluxes are fixed. The results provide a general, replicable method for analyzing wave turbulence in more complex settings. Finally, the generalization of this theory to the inhomogeneous wave turbulence provides a possible explanation to the instability of the Kolmogorov-Zakharov spectra in some 1D inhomogeneous models with 4-wave interactions such as the Majda-McLaughlin-Tabak. This work opens the discussion regarding universal and non universal properties in two-point correlation functions. This opens new range of study on the phenomena of intermittency which is partially developed here.

In our work, based on an exact solution of the Ising model on a decorated square lattice with an arbitrary number of decorating spins, we have demonstrated the fundamental possibility of describing the phenomenon of multiple magnetic phase transitions (magnetic reentrancy) in the regime of competing exchange interactions. We analyzed the magnetic behavior of the system and established the conditions for the occurrence of magnetic reentrancy. We have determined the relationships of the model parameter under which the formation of one, three, and even five magnetic phase transitions is possible, which is confirmed by a complicated magnetic phase diagram. We have also proposed a unique method for finding critical temperatures that allows for the precise determination of their number and values, including those at extremely low temperatures.

We develop a unified dynamical-spectral framework for equilibration in isolated quantum systems based on a subspace coarse-graining approach. Central to our formulation is the Leakage Fidelity Function (LFF), defined as the probability that a unitarily evolving state escapes the support of its initial subspace. This quantity provides a direct, operational measure of information flow and memory loss without invoking ensemble assumptions or perturbative arguments. We derive universal bounds on temporal fluctuations of the LFF, in terms of the spectral gap structure and the square of the effective dimension, evincing that large spectral delocalization suppresses fluctuations and guarantees equilibration on average. By introducing spectral power distributions and associated entropic measures, we establish a quantitative link between phase mixing, gap participation, and dynamical stability. We further investigate the equilibration timescale by connecting the LFF to quantum speed limits, thereby revealing the average time required for equilibration. Our results provide a state-dependent, geometrically transparent perspective on how spectral complexity and subspace information leakage jointly govern irreversibility in closed quantum many-body systems.

We study a simplified model of a ferromagnetic metal proximitized by a fully-gapped $s-$wave superconductor and integrated with a microwave resonator. The low-energy excitations in the combined system consist of ferromagnetic magnons, Bogoliubov excitations of the superconductor, and cavity photons. We show here that when the magnons and photons have comparable frequencies and are subject to an external drive, the hybridized driven magnon-polaritons induce a non-equilibrium crossover from the expected proximitized nodal $p-$wave superconductor to a fully gapped $(p_x+ip_y)-$superconductor. Moreover, the characteristic crossover temperature is inversely related to the magnon-photon detuning. We compute the temperature-dependent renormalization of the cavity photon frequencies across this nodal to nodeless evolution, which modifies the kinetic inductance of the resonator, and find a number of non-trivial features tied to the non-equilibrium (i.e., driven) nature of the problem. We compare and contrast these results with a recent circuit quantum electrodynamics (cQED) based experiment studying a permalloy-niobium bilayer, where a non-trivial dependence of the low-temperature cavity response on the magnon-photon detuning was observed. Our results pave the way for a principled exploration of engineering novel states of matter by coupling cavity photons to electronic collective modes in correlated two-dimensional materials and interfaces.

`Solvable' circuits, such as dual unitaries and its generalisations, have arisen as paradigmatic examples of tractable chaotic non-equilibrium dynamics, both in classical and quantum systems. However, while increasingly more complicated sufficient conditions have been proposed, a systematic theory classifying and understanding general features of solvable circuits is missing. We develop such a theory by introducing influence-solvable circuits, a class of $(1+1)D$ circuits whose influence matrix, which represents the `bath' generated by its own evolution, is given by a uniform MPS with finite bond-dimension $χ$. This property allows for efficient computation of subsystem dynamics and essentially contains all known examples of solvable circuits. We derive a set of necessary and sufficient local conditions by using a version of the fundamental theorem of MPS for open boundary conditions. Next we apply our theory to brickwork circuits with $χ=1$ influence-solvability and perform a systematic classification of classical brickwork circuits with local dimension up to $d=3$ and quantum brickwork circuits with $d=2$. Our search reveals new solvable circuits that are not captured by known solvability conditions.

Topological quantum computation relies on braiding non-Abelian anyons, but requires the underlying topological order to survive imperfect state preparation and environmental noise. We show that the instability of topological order to wavefunction deformations and to decoherence, with the latter probed by syndrome distributions, are generically captured by stat-mech models whose symmetries naturally expose the corrupting anyonic excitations. As an example, we combine this framework with Monte-Carlo simulations to resolve the stability of $D_4$ topological order under deformations and quantum channels that proliferate multiple non-Abelian anyon species that individually are unable to condense. We show that beyond a finite threshold, proliferation of two non-Abelian anyon species parasitically condenses a shared Abelian-anyon fusion outcome, destroying the topological order. Our symmetry-based approach sharply differentiates the resulting trivial phase from that obtained by condensing all Abelian charges; in other words, the trivial phase "remembers" which anyons condensed. This framework provides a first step into identifying the relevant symmetry for optimal decoders, conditioned on syndrome measurements, of non-Abelian topological order.

A new moiré material platform was recently proposed based on twisting two-dimensional atomic monolayers whose low-energy states lie at the three M-points of the Brillouin Zone. Continuum and ab initio modeling suggest that electrons in the conduction bands of these materials realize three-valley Hubbard models with valley-selective, quasi-one-dimensional hopping. Remarkably, the onsite Hubbard repulsion is almost $U(6)$-symmetric without fine-tuning. Here, we show that this class of systems naturally admits sign-free determinantal Quantum Monte Carlo simulations at a filling of three electrons per moiré unit cell. We use these to explore the phase diagram for interactions of various strengths and $U(6)$-breaking anisotropies. We show that for near-isotropic interactions as relevant to, e.g., AA-stacked twisted SnSe$_2$, the system exhibits an extended intermediate-coupling regime in which local-moment formation and itinerancy compete, and the crossover to a putative low-temperature ordered state can be understood in terms of fluctuating $U(6)$ local moments. We argue that many of these features persist beyond the idealized sign-problem-free limit.

We derive semiclassical equations of motion for general composite bound states in insulators and semiconductors, covering excitations such as excitons and trions. For neutral composites we find that a uniform external electric field does not couple to a Berry curvature term, contrary to the naive expectation from single-electron dynamics. Instead, a distinct quantum geometric quantity appears generically in the equations of motion. This quantity is the difference between inequivalent Berry connections that can be defined for the composite, generalising the concept of the quantum geometric dipole previously studied for excitons. In the case of charged composites such as trions, we find an additional Berry curvature contribution to the equations of motion. As we demonstrate, however, there is an infinite family of inequivalent composite Berry curvatures, and so care must be taken to make the correct choice that describes the physical dynamics. We explain how this choice should be made dependent on the definition of a spatial centre for the composite. We end by discussing composite dynamics that have no single-electron counterpart. We find that trions in magic-angle twisted bilayer graphene undergo a transverse drift under an applied electric field and that this is driven not only by the Berry curvature contribution but also by the quantum geometric dipole. The interplay of these two geometric contributions further imprints itself on the trion's internal dynamics, causing its dipole moment to oscillate in time.

We study the time evolution of tripartite entanglement in heavy local quenches in two-dimensional holographic conformal field theories. Our diagnostic is the genuine multi-entropy of adjacent intervals, computed from both bulk and boundary perspectives. A perturbative bulk analysis shows that the first-order small-mass perturbation around the vacuum geodesic network cancels identically at any time after the quench. In the fully back-reacted geometry, a vacuum-subtracted genuine multi-entropy arises from a mismatch between the winding selected by the trivalent geodesic network and the windings selected independently by the pairwise geodesics. In the sharp quench limit, the time dependence of genuine multi-entropy is kinematically fixed to logarithms of rational functions of time and is independent of the heavy operator dimension. The CFT calculation reproduces the same formula within the heavy-light vacuum block approximation, where the branch choice in the heavy-background uniformization map corresponds to the winding selection in the bulk. These results indicate that, in this setup, the genuine multi-entropy is controlled by global saddle selection, rather than by a local energy response or quasiparticle propagation.

The periodic table of elements transformed chemistry by revealing simple organizing principles underlying atomic behavior. Despite decades of effort, no analogous framework has emerged for crystalline materials -- their microscopic complexity and vast configurational space have defied reduction to fundamental organizing principles. Current databases catalog thousands of synthesized materials, but extracting predictive, interpretable models from this wealth of data remains a formidable challenge. Here we demonstrate that the materials landscape possesses a hidden geometric organization that can be unveiled through unsupervised nonlinear dimensionality reduction. Applying differential geometry techniques to the Inorganic Crystal Structure Database (ICSD), we reveal that just a few combinations of microscopic descriptors capture the vast majority of variance in material properties. This low-dimensional embedding autonomously segregates superconductors from ordinary materials and further distinguishes superconducting families in ways that transcend chemical similarity alone. Remarkably, the discovered geometric organization directly governs critical temperatures ($T_c$) across diverse superconducting families, enabling accurate $T_c$ predictions without any knowledge of the pairing mechanism. Our approach uncovers emergent organizing principles that control macroscopic quantum behavior, offering a new paradigm in how we build models of complex quantum materials directly from experimental data.

We analyze Maxwell theory coupled to an axion domain wall as a spectral boundary value problem. We find that a finite-width axion domain wall generically supports a localized normalizable chiral electromagnetic mode with linear, gapless dispersion. This mode arises from helicity-dependent coupling sourced by the axion gradient: one polarization experiences an effective attractive potential and forms a bound state, while the opposite polarization is repelled. The existence of this chiral surface photon is robust over a wide regime of wall structures and axion masses. Our result shows that axion domain walls generically support a localized chiral photon that has been missed in previous analyses.

Neutron scattering is central to identifying quantum states of magnetic materials. In the search for quantum spin liquids, broad spectral features of inelastic spectra have been cited as evidence for spinon excitations, but can also arise from magnon excitations excitations in the presence of quenched disorder and strong magnon interactions. We develop a new approach to this problem, based on the adiabatic continuity in the $XXZ$ Heisenberg model on geometrically frustrating (GF) lattices as a function of the model's anisotropy. Using this approach, we identify universal features and energies of finite-temperature spin correlators. Focusing on the kagome lattice, we show that the low-energy spin spectral function contains robust, momentum-independent peaks with frequencies: $ω_1 \approx 3.4 T^*$ and $ω_2 \approx 6.3 T^*$, where the ``hidden energy scale'' $T^*$ is the characteristic scale of a low-temperature peak in the heat capacity, at which many GF magnets also display spin-glass freezing. We show that the spectral features at low energies $ω\lesssim T^*$ arise from single-magnon scattering and identify the magnetizations of the respective excitations. We explore the evolution of the spectral features with temperature and discuss extensions to other GF lattices. Our results provide a sharp spectroscopic criterion for interpreting neutron scattering in kagome and other GF quantum magnets.

A new moiré-material platform has recently been proposed based on twisting two-dimensional triangular-lattice monolayers whose low-energy states lie at the three M points of the Brillouin zone. Continuum models derived from extensive ab initio simulations suggest that electrons in the conduction bands of one such M-point moiré material, twisted AA-stacked SnSe$_2$, realize a three-orbital Hubbard model with orbitally-selective, quasi-one-dimensional (quasi-1D) hopping, protected by a projective mirror symmetry. Here, we show that the resulting "mixed-dimensional" limit -- in which the hopping is exactly quasi-1D in each valley, while the valleys are coupled by interactions into a fully two-dimensional network -- can be sampled with Stochastic Series Expansion (SSE) quantum Monte Carlo (QMC) without a sign problem at any filling. We develop an efficient new SSE QMC algorithm that combines custom global updates with parallel tempering to overcome the equilibration challenges posed by the mixed-dimensional setting. We then use this algorithm to explore the phase diagram of M-point twisted AA-stacked SnSe$_2$. Over extended and realistic ranges of twist angles and interaction strengths, we find that at integer fillings the system supports correlated insulators whose nature and strength depend strongly on angle. At certain commensurate fractional fillings, we further find evidence for Wigner-Mott insulators. We analytically account for the main features observed numerically using a strong-coupling description. Finally, we discuss perturbations away from the mixed-dimensional limit and the possibility of applying our method to other realizations of mixed-dimensional Hubbard models.

The realization of on-chip photonic circuits requires scalable and deterministic single-photon emitters (SPEs) at room temperature, which remain a challenge in van der Waals materials. In this work, we report a novel three-step fabrication process for the generation of spatially controlled SPE arrays in hexagonal boron nitride (hBN). The process comprises site-selective gallium (Ga) focused ion beam milling, nanoscale conformal carbon deposition over the patterned regions, and subsequent thermal annealing. The synergistic combination of these steps resulted in a site-correlated emitter yield of ($\sim 89\%$) across 100 fabrication sites. Second-order autocorrelation measurements revealed pronounced three-level emitter dynamics where the best emitters exhibited high purity ($g^{(2)}(0)=0.15 \pm 0.09$).To the best of our knowledge, this is the first lithography-free, direct-write approach combining Ga-ion milling, selective carbon engineering, and thermal annealing to deterministically generate \hBN{} \SPE{}s. The reproducibility of the method is validated across multiple independently fabricated samples. These results establish a scalable, lithography-free pathway toward on-demand SPE arrays relevant to integrated quantum photonics.

In the infinite dilution limit, we show that the solvent-mediated potential of mean force (PMF) between solutes, extracted from the hypernetted-chain (HNC) closure of the Ornstein-Zernike equations, can expressed as a convolution between solute-specific generalised excluded volume functions. In the limit of a structureless solvent of point particles and hard core solutes, this recovers the exact Asakura-Oosawa depletion potential as the overlap between excluded volume spheres. The methodology can be deployed for ultrasoft particle systems such as those encountered in dissipative particle dynamics (DPD), where the solvent-mediated PMF can be recovered with considerable accuracy. These results confirm that in coarse-grained molecular DPD simulations the parametrisation of the non-bonded repulsions is sensitive to the assumed intramolecular bond lengths if they are smaller than the range of the DPD potential, due to the overlap of the soft excluded volume functions.

Renormalization group (RG) fixed points are commonly associated with scale invariance and a divergent correlation length. We show that this connection can fail in the tricritical $O(N)$ model in three dimensions in the limit $N\to\infty$. Revisiting the line of fixed points identified by Bardeen, Moshe, and Bander, we use the functional renormalization group to clarify the mechanism leading to mass generation at its singular endpoint (the BMB fixed point). We demonstrate that the generated mass is nonuniversal and originates from the nonanalytic structure of the effective potential. We show that the critical exponent $ν$ which takes the value $ν= 1/2$ along the regular part of the BMB line, that is, for $0 \leq λ< λ_{\rm BMB}$, jumps to $ν= 1/3$ on the singular part of this line with the BMB FP, corresponding to $λ= λ_{\rm BMB}$, being the pivotal point between these two regimes. We also show how its singular potential emerges dynamically along the renormalization flow.

We investigate the electronic structure, spin textures, and charge to spin/orbital transport in graphene encapsulated by 1T-TaS$_{2}$ monolayers in the charge density wave phase. Using first-principles calculations, tight-binding modeling, and the Kubo formalism, we show that the encapsulation stacking dictates fundamentally distinct transport regimes. In the asymmetrical (AA) stacking, proximity fields from both interfaces constructively interfere, yielding a cumulative Rashba phase of nearly $π/2$. This pure radial Rashba spin pattern leads to the unconventional Rashba-Edelstein effect, which robustly dominates over the conventional response by a factor of 35 across a wide energy range. Conversely, the symmetrical (AA') stacking preserves a horizontal mirror symmetry, establishing a stable, purely out-of-plane persistent spin texture. Furthermore, the computed orbital Hall effect is exceptionally efficient, surpassing the spin Hall effect by three orders of magnitude. Within the proximity-induced spectral gaps, the orbital Hall conductivity exhibits a finite plateau, whereas the spin Hall conductivity vanishes. Our findings establish graphene encapsulated heterostructures as a promising system for realizing distinct charge to spin and charge to orbital interconversion regimes determined by the choice of stacking order.

Origami-inspired metamaterials exploit the interplay between geometry and elasticity to achieve programmable mechanical responses. Yet the origin and tunability of snap-through instabilities in non-rigidly foldable patterns remain poorly understood. Here we show that the Mars tessellation, a degree-4 vertex origami pattern composed of alternating square and rhombic faces, is not rigidly foldable because the folding-speed ratios required for vertex compatibility cannot be propagated consistently across neighboring units. This geometric incompatibility forces the facets to bend during folding, giving rise to a reproducible snap-through discontinuity in the force-displacement curve with a mean force drop of about 92.6 +/- 5.5 %, marking a transition between metastable states. Laser scoring of additional diagonal creases, guided by strain-field simulations, enables continuous tuning of the snap magnitude. These results reveal a general mechanism by which geometric frustration can be harnessed to program multistability in thin-sheet metamaterials.

We investigate the localization properties of cavity-coupled electronic states in disordered systems, motivated by recent proposals of cavity-mediated hopping in quantum Hall systems. We first introduce a minimal two-band model in which localized states in a disordered one-dimensional band are coupled, through a homogeneous cavity mode, to an excited band of delocalized states. Combining perturbation theory with a transfer-matrix approach, we show that cavity-assisted hopping between localized states decays exponentially with distance, implying that the eigenstates remain localized even beyond the perturbative regime. Nevertheless, the corresponding localization length increases with the light-matter coupling strength and can extend over several lattice sites in the single-electron ultrastrong-coupling regime. We then study a disordered Landau band coupled to a cavity mode within the framework developed in Refs.[1,2]. We find that the effective cavity-mediated coupling between edge states also decays exponentially with distance, but with a localization length that can reach micrometer scales for experimentally realistic parameters. By analyzing the inverse participation ratio, we show that this enhanced coupling is predominantly mediated by the most extended states of the upper Landau band. Our results demonstrate that, while cavity-induced hopping in disordered quantum Hall systems remains exponentially localized, the associated localization length can become sufficiently large for the corresponding states to exhibit effectively delocalized behavior on mesoscopic length scales.

We study the roughening of an interface with nonlinear elasticity driven by temporally correlated noise, which breaks statistical tilt symmetry. Using scaling arguments and a self-consistent Hartree approximation, we derive the crossover diagram and the steady-state structure factor. We identify three scaling regimes associated with the Larkin, anharmonic Larkin, and Edwards--Wilkinson universality classes, and obtain the crossover lengths separating them. Numerical simulations of large systems confirm the analytical predictions over the full parameter range. Our results provide a unified description of finite-size and crossover effects in a minimal nonlinear-elastic Ornstein--Uhlenbeck active interface.

We discuss diffusion-model paths interpolating between a target distribution known only through p patterns and a reference distribution that can be sampled. These interpolating paths can be constructed symmetrically or else in forward direction (often referred to as a "noising") from the target patterns to the reference distribution or in backward direction (as a "denoising") from the reference distribution to the patterns. For backward paths with identical diffusion noise, we consider the path convergence in number of patterns p towards the path for infinitely many patterns. In a one-dimensional test case, we show that this convergence is on a scale 1/sqrt(p), but with infinite mean square deviation. We demonstrate that the path convergence allows for extrapolation towards the p=infinity path which samples the target distribution. We provide a proof-of-concept extrapolation algorithm and propose the convergence and extrapolation of paths as a possible strategy for density estimation and generalization. We illustrate all our algorithms through pseudo-codes and provide Python implementations.

Unconventional magnets combine the vanishing stray fields of antiferromagnets with the strong spin-splitting of ferromagnets, offering a unique material platform for spintronics. However, a critical challenge in realizing functional spin-logic devices lies in preserving long-range spin coherence against momentum-degrading scattering and gate-induced dephasing. Here, we demonstrate that the intrinsic momentum-dependent exchange field of a three-dimensional $p$-wave magnet can be precisely tuned against gate-induced Rashba spin-orbit coupling to establish a \textit{generalized} $\mathrm{SU}(2)$ spin-rotation symmetry. This emergent conservation law generates a symmetry-protected Persistent Spin Helix (PSH), effectively integrating the high energy scales of 3D bulk magnetic exchange with the macroscopic coherence of symmetry protection. By modeling a synergistic $p$-wave magnetic spin field-effect transistor (spin-FET), we reveal high-visibility Datta-Das conductance oscillations controlled purely by electrical gating. Crucially, our quantum transport simulations confirm that this symmetry-engineered transport regime exhibits exceptional resilience against strong non-magnetic Anderson disorder and geometric variations. These results establish a synergistic paradigm for non-magnetized spintronics, demonstrating how the active integration of spin-orbit coupling and unconventional magnetism can yield disorder-resilient spintronic logic.

We study an electron-phonon interaction driven thermoelectric diode. The nonreciprocity in this diode arises from the asymmetry between the probabilities of phonon emission and absorption in the electron-phonon interaction, as well as the structural reflection asymmetry. We reveal the intrinsic nature of this nonreciprocity, as the forward and backward electron transport remains asymmetric even when the applied temperature difference is not reversed. This intrinsic nonreciprocity gives rise to two novel transport phenomena. One is a novel thermoelectric effect which is driven by the temperature difference between the leads and the central device region, rather than the conventional temperature difference between the two leads. The second, and more significant, phenomenon is the suppression of electronic backscattering in the load resistor. This suppression decreases the resistance of the load resistor, which leads to the breakdown of Ohm's addition law. Under suitable conditions, the presence of electron-phonon interaction can yield a larger thermoelectric current compared to the case without it. This intrinsic nonreciprocity opens up a new pathway for low-power electronics besides topology and superconductivity, and for nonreciprocal thermoelectric devices.

We show that the strain-driven semi-Dirac transition in Kekulé-O graphene gives rise to a sequence of anisotropic optical fingerprints associated with band structure reconstruction. Across the transition, optical spectral weight is continuously redistributed among the dominant interband transitions, leading to a pronounced enhancement of the optical anisotropy. Combining numerical four-band calculations with analytical low-energy results, we identify three low-energy fingerprints that emerge with increasing strain: gapped absorption peaks, semi-Dirac critical scaling, and a pronounced van Hove optical resonance. At the semi-Dirac critical point, where the Kekulé gap closes at the $Γ$ point, the low-energy optical conductivity is characterized by $σ_{xx}(Ω)\proptoΩ^{1/2}$ and $σ_{yy}(Ω)\proptoΩ^{-1/2}$. Beyond the transition, the semi-Dirac point splits into two anisotropic Dirac cones, accompanied by the emergence of saddle points near the $Γ$ point. The resulting saddle-point excitations produce a pronounced van Hove optical resonance at energies well below those of graphene, while the split Dirac cones give rise to an anisotropic constant optical conductivity. We further show that the low-energy optical fingerprints can be traced to the continuous evolution of a dominant optical transition channel driven by strain-induced band reconstruction. Moreover, the fingerprints remain identifiable in the presence of moderate disorder broadening and finite-temperature effects, indicating their potential observability under experimentally realistic conditions.

The high mobility of current-driven skyrmions in synthetic antiferromagnets (SAFs) is widely explained by the macroscopic suppression of the skyrmion Hall effect through gyrotropic force compensation. This established view, however, overlooks a concurrent and significant reduction in the Gilbert damping parameter α, a key factor in the Thiele equation governing skyrmion velocity. Here, we show that this damping attenuation originates from a reconfigured magnon-electron scattering landscape. Using a microscopic s-d model, we demonstrate that the strong antiferromagnetic interlayer Ruderman-Kittel-Kasuya-Yosida (RKKY) exchange coupling in SAFs increases the magnonic gap of skyrmion collective modes, thereby suppressing the thermal magnon population and, consequently, the magnon-electron scattering rate that dominates damping in metallic ferromagnets. Our work establishes a dual-mechanism framework to fully explain the superior kinetics of SAF skyrmions: the macroscopic topological effect rectifies the motion direction, while the microscopic dissipation mechanism reduces the drag. This synergy enables high-speed and efficient motion, providing a fundamental elucidation of the enhanced mobility reported in recent studies such as the work by Pham et al. [Science 384, 307-312 (2024)].

Three-dimensional (3D) optical solitons bear far stronger relevance to multi-dimensional nonlinear dynamics prevalent in complex physical, chemical and biological systems than conventional 1D solitons, and they support far richer and more intricate phenomena arising from their spatiotemporal degrees of freedom. However, real-time recording of 3D soliton evolution with simultaneous spatiotemporal resolution remains a critical challenging. Here, we demonstrate long-term real-time visualization of 3D soliton dynamics with spatiotemporal resolution using high-speed photodetectors combined with joint space- and time-division multiplexing. We visually capture complex transient behaviors of 3D solitons in a multimode fiber laser and, by integrating our method with the time-stretch technique, simultaneously record pulse-resolved beam and spectral evolutions. We observe that during the birth of 3D solitons, the highly multimode beam stabilizes for a substantial interval prior to spectral broadening, indicating that a large number of transverse modes have already locked before longitudinal mode proliferation. These findings highlight the critical importance of real-time spatiotemporal visualization for advancing ultrafast multimode laser design and delivering new insights into high-dimensional nonlinear dynamics.

We develop a framework that uses the lattice structure of the $k$-th Volterra--Bogoyavlensky equations ($k\in\mathbb N$, $k>1$) to generate rational solutions of higher symmetric Painlevé equations. For $k=2$, we show that the Volterra lattice, equipped with suitable initial conditions, exactly models the one- and two-dimensional orbits generated by half-translation operators of the $A_2^{(1)}$ symmetric Painlevé IV equations. This correspondence yields explicit closed-form expressions for all solution components in terms of generalized Okamoto polynomials and leads to new algebraic recurrence relations among these polynomials. We then extend the construction to higher values of $k$. Rational solutions are generated from seed solutions invariant under the dihedral group $D_{2k-1}$ of the $A_{2(k-1)}^{(1)}$ symmetric Painlevé equations and are organized by Volterra--Bogoyavlensky shifts. The construction is carried out explicitly for the $k=3$ Bogoyavlensky lattice, producing a new class of rational solutions of the $A_4^{(1)}$ symmetric Painlevé equations. These results establish a direct connection between Volterra--Bogoyavlensky lattices, dihedral symmetries, and rational solutions of higher Painlevé systems.

The article is devoted to the problem of decay of initial stationary state for the Volterra lattice with zero boundary condition. We show that this process is asymptotically self-similar and calculate the propagation velocity of the decay wave, the leading terms of the asymptotics and corrections, in the main and transition sectors of the wave.

Can a classically relevant stress-tensor deformation be exactly solvable, duality preserving, and physically causal? We construct an exact power-law family of nonlinear electrodynamics preserving electromagnetic duality, together with a parallel two-dimensional Lax-integrable realization. Its auxiliary geometry yields the full characteristic-cone phase diagram and a universal finite-energy fold. For the Maxwell seed, every nonzero relevant branch is acausal, whereas every causal branch is caustic-free; undeformed Maxwell theory is the only causal point in the relevant regime.

We associate each quiver with soliton solutions of nonlinear integrable systems containing the KP and Toda hierarchies. We give an explicit and computable formula of these soliton solutions which are regarded as universal ones obtained by degenerating quasi-periodic solutions.

We study operator growth in quantized non-KAM systems using out-of-time-ordered correlators (OTOCs), focusing on the kicked harmonic oscillator as a representative example. Since the classical harmonic oscillator is degenerate, the dynamics fall outside the usual Kolmogorov-Arnold-Moser (KAM) framework, and resonances play a central role in shaping the phase space. We examine the system near resonances, where the ratio between the oscillator and driving frequencies takes integer values. Even though the classical Lyapunov exponent remains small at these points, and hence no conventional chaos, the phase space still undergoes strong structural changes. The OTOCs are particularly sensitive to these resonances, with a quadratic-in-time growth at resonance compared to linear growth away from it. Within a perturbative treatment, we derive closed-form expressions for the OTOCs and uncover a number-theoretic structure emerging in the behavior of OTOCs, governed by the Euler totient function of the frequency ratio. Overall, the results we present in this short note imply that resonant structures can play an important role in controlling information spreading.

The concept of multiple, long-delayed feedback systems is introduced and discussed with reference to a paradigmatic model. We analyse how the resulting chaotic dynamics is affected by the delay distribution. Via a mean-field approach, we show that a spatio-temporal representation equivalent to the one developed for the single-delay can be extended to this wider class of dynamical systems. Numerical simulations are complemented by a theoretical study based on a multiple-scale analysis, which, in the vicinity of a Hopf bifurcation, allows mapping the initial model onto a complex Ginzburg Landau equation. As a result, we find that the only relevant feature influenced by the multiple delays is the size of the coherent spatio-temporal structures which, in turn, depends exclusively on a generalized {\it variance} of the delay distribution.

Adaptive networks couple the evolution of node states to the evolution of the interactions between them. In fast-adapting phase oscillator networks, a slow-manifold reduction of a pairwise microscopic model can generate effective higher-order terms in the phase dynamics. We ask whether this higher-order structure survives the dense-graph continuum limit, and whether it matters if one first reduces and then passes to the continuum, or first passes to the continuum and then reduces. We prove well-posedness and discrete-to-continuum convergence for the unreduced and first-order reduced models, and we construct the continuum slow manifold directly in a Banach-space setting. Along admissible equal-cell step approximations, the two routes give the same first-order continuum vector field, including the same pairwise correction and triplet operator, up to controlled $O(\varepsilon^2)$ remainders. A continuum mixed-derivative criterion then shows that, for suitable coupling functions, the resulting triplet operator is genuinely nonpairwise in the smooth bounded-kernel class. Thus the higher-order term is not a finite-network artefact, but persists in the macroscopic continuum description considered here.

This study investigates the terminal breakdown of human neurophysiological function through the lens of non-linear dynamics by analyzing the multifractal spectrum. Using Multifractal Detrended Fluctuation Analysis (MF-DFA), we quantify the temporal evolution of complexity in synchronized electroencephalogram (EEG) and electrocardiogram (ECG) time series from patients in the terminal stage. Our results reveal a marked divergence in multifractal spectrum width: while neural activity exhibits a collapse of multifractality toward a more constrained state, cardiac signals undergo anomalous spectral broadening, indicating increased non-linear fluctuations and dynamical instability. A negative correlation between these spectral widths suggests effective functional decoupling and the emergence of anti-correlated dynamics between neural and cardiac systems. Rather than reflecting a uniform physiological decline, this divergence is consistent with a body-to-brain breakdown in which peripheral dysfunction progressively overwhelms central regulatory processes. In a broader context, the observed opposing trends resemble patterns reported in other body-driven adaptive processes, suggesting that inverse dynamics across coupled systems may emerge when constraints originate from peripheral rather than central mechanisms. Ultimately, the dying process appears to represent an extreme form of cross-system disintegration, marked by the collapse of the hierarchical coordination that normally sustains integrated physiological function.

The posture of the vocal folds produced by laryngeal muscle activation plays a central role in determining the dynamics of voice production. Abnormal vocal fold configurations are frequently associated with inefficient phonation and a variety of voice disorders. Although diverse glottal closure patterns have been observed clinically, the biomechanical mechanisms governing their dynamic behavior and resulting phonatory characteristics remain incompletely understood. Moreover, existing numerical models that incorporate the effects of the intrinsic musculature on posturing and glottal conformation are computationally expensive, which limits their suitability for large-scale parametric investigations. In this work, we introduce a computationally inexpensive vocal fold (VF) model wherein the body and cover VF layers are treated as a composite beam and a coupled membrane, respectively. Intrinsic laryngeal muscle activation, in addition to positioning the arytenoid cartilages and cricothyroid joint, introduces moments at the boundaries of the structure that influence glottal conformation. The model produces phonatory characteristics that are qualitatively consistent with those reported in high-fidelity finite-element models and clinical studies, thereby supporting its predictive capability while offering substantial computational advantage. The proposed framework provides biomechanical insights into the influence of incomplete glottal closure on phonation dynamics and may serve as a computationally tractable tool for investigating mechanisms underlying certain voice disorders.

Light propagation in disordered media is particularly complex due to the interference effects and polarization-mixing terms. Two-dimensional systems are particularly interesting because they feature two decoupled scattering channels, with distinct spectral statistics due to the presence of polarization-coupling terms in only one of them. We here show that in one channel, Anderson localization bounds the light transport both longitudinally and transversely. In the second channel, in contrast, diffusive transport takes place independently of scatterer density. Our results open new possibilities for engineering of light transport in two-dimensional waveguides.

Maritime chokepoints concentrate shipping traffic. Disruptions to this traffic can have a widespread impact on the global economy. However, the way in which these impacts are shaped by the shipping sector's adaptive behavior is not well understood. Here, we introduce an empirically calibrated full-scale agent-based model of the global commercial shipping fleet, representing 35,954 active ships moving among 1,651 ports. We use the model to quantify how rerouting changes arrival losses under chokepoint closures. Static route exposure alone does not predict realized losses. In the adaptive model, rerouting reduces losses at some directly exposed ports, while delayed vessel cycles create losses at later port calls and in dependent regions. Cumulative net shipping-day losses therefore continue to rise with closure duration because longer routes keep ships delayed after the initial adjustment. Each additional closure day reduces global shipping arrivals by 3.0% for Suez and 7.7% for simultaneous Suez, Panama, and Malacca closures. These losses are unevenly distributed in exposed regions and ports. Disruptions with known duration show different loss profiles from unexpected shocks with unknown duration, revealing that end-date information can reduce avoidable short-run losses. The results show that chokepoint risk is a dynamic problem of routing, timing, and regional exposure and not a static property of maritime-network topology.

We present a robust, ultrafast and highly efficient setup for distinguishing molecular enantiomers by combining ultrafast techniques with vector beams, a type of topological light with azimuthally varying polarization. An infrared vector beam generates high-order harmonics in a sample of randomly oriented chiral molecules, resulting in the emission of an ultraviolet vector beam whose intensity profile carries information about the handedness of the chiral molecules. Our approach allows for spatial discrimination of molecular enantiomers, opening a new route for studying ultrafast chirality.

Lonafarnib (LNF) is an investigational drug targeting hepatitis delta virus (HDV) but not hepatitis B virus (HBV), providing a unique opportunity to model HDV kinetics and how changes in HDV affect HBV. We performed a detailed kinetic analysis and developed a mathematical model to explain serum HBV DNA, HDV RNA and hepatitis B surface antigen (HBsAg) kinetics in 15 HBV/HDV coinfected patients receiving LNF-based treatment. After a delay of 0-2 days, patients experienced a rapid 1st-phase HDV-decline followed by either a viral plateau, 2nd slower-decline phase, or viral breakthrough (VB). LNF monotherapy led to a flat-partial-response (often followed by VB), while LNF combination therapy with ritonavir or pegylated interferon-$α$ (PEG-IFN$α$) was associated with a biphasic HDV decline (without VB). All treatments except LNF+PEG-IFN$α$ had at least one patient experiencing an increase in HBV on-treatment. Our model successfully reproduced the observed HDV and HBV kinetics. We estimated an HDV RNA half-life of 1.26 days [95% confidence interval, CI: 1.05--1.47] in serum and treatment efficacy of 94% in inhibiting HDV RNA production across all treatments [95% CI: 89%--97%], as reflected by the 1st phase HDV decline. The 2nd phase of HDV decline was explained by a time-dependent increase in efficacy, reaching a maximum of 98.9%. The model explained the increase in serum HBV DNA by a median 4-fold [interquartile range, IQR: 1--28] increase in HBV DNA production rate when HDV declined below an inhibitory threshold. The stability of serum HBsAg was explained by a constant number of HBsAg-producing cells.

In this work, we report on the design strategies and experimental validation of ultra-low threshold ($< 0.8\,\mathrm{mA}$) hybrid III--V/Si quantum dot (InAs/GaAs) micro-ring lasers with optical output powers $> 2\,\mathrm{mW}$ for $1.3\,μ\mathrm{m}$ emission. The multi-dimensional design exploration allows for the demonstration of record wall-plug efficiencies ($\sim 10\%$) and threshold current densities ($109\,\mathrm{A/cm^2}$) for these compact sources on silicon. We also demonstrate the thermal performance of several designs with record characteristic temperature values of $T_0 = 212\,\mathrm{K}$, indicating minimal temperature dependence of the threshold current. In addition, the high differential gain allows for the demonstration of 3-dB bandwidths up to $5\,\mathrm{GHz}$.

Present-day solution-processable single-photon sources are dominated by three-dimensionally confined colloidal quantum dot emitters, yet their particle-to-particle variation in single-exciton properties limits reproducibility and scalability. Here, to avoid such heterogeneity, we demonstrate reliable room-temperature single-photon emission from atomically flat two-dimensional (2D) colloidal quantum wells (CQWs) with inherently uniform one-dimensional quantum confinement, despite their long-standing limitations of efficient multiexciton emission and pronounced exciton-surface susceptibility. We resolve these challenges through volumetric quantum design (VQD) of CQWs, yielding a highly localized, single bandedge state. This design laterally confines the bandedge excitonic domain within the exciton coherent area and vertically decouples it from surface states via a thick, strain-relieved quantum-barrier shell that preserves strong confinement, overcoming the daunting thickness-confinement trade-off in 2D CQWs. Statistical single-particle spectroscopy reveals that VQD-CQWs deliver near-blinking-free (on-time >99.5%) and fluence-insensitive antibunching (g(2)(0): 0.041), protected by a bandedge-state-filling bottleneck, together with linear polarization of up to 73% under cavity-free conditions, originating from synergistic transition-dipole and electric-field anisotropies. These advances establish 2D CQWs as a viable, homogenous and scalable platform for quantum technologies.

Efficient and generalizable full-wave simulation is essential for nanophotonic analysis and inverse design, yet existing methods face a tradeoff between the high computational cost of numerical solvers and the limited generalizability of neural operator models for complex optical scattering. Here, we introduce FEMONet, a finite-element-constrained operator-learning framework that learns light scattering from an operator parameter space to a Galerkin-consistent solution space. The operator parameter space encodes the physical entities defining a wave-equation problem, while the variational weak form links this space to the coordinate and physical solution spaces. Integrated with operator-learning networks, FEMONet extends classical solvers from isolated problem instances to parameterized scattering operators. To our knowledge, FEMONet represents the first Galerkin-consistent operator-learning framework for complex-valued optical scattering, grounded in the variational weak form of the governing vector wave equations. Finite-element discretization absorbs spatial derivatives into assembled stiffness matrices and load vectors, removing coordinate-based derivatives of the neural-network output from the physics loss and improving training efficiency. By predicting finite-element expansion coefficients rather than unconstrained field values, the Galerkin-consistent formulation preserves compatible trial and test spaces, achieving high accuracy, stable training, and generalization across dielectric, metallic, arrayed, plasmonic, and three-dimensional nanophotonic structures.

We present Archê, an orbital-free molecular dynamics (OFMD) code designed to produce equations of state (EOS) in the plasma state. Unlike in other OFMD codes, Archê uses a self-consistent field (SCF) approach to compute the electronic density. This allows us to implement two algorithms that accelerate SCF convergence by a factor of up to six. First, the density is initialized using the results from the previous MD timestep to define a one-center profile, which is then applied to the new nuclei positions. Second, the initial and final densities are mixed at each SCF iteration in a proportion that minimizes an approximate free energy. We validate the code by comparing the calculated aluminum EOS to results obtained with the Kohn-Sham density functional theory software Abinit. Achieving agreement in the internal energy requires adding a correction related to the norm-conserving pseudopotential derived from an average-atom model. Performance is compared across CPU and GPU architectures, demonstrating an order-of-magnitude speedup for a single GPU compared to 256 CPUs. Archê exhibits an overall linear computational complexity with respect to the number of atoms, as well as the number of real and reciprocal grid points. Execution time is weakly dependent on density; however, interestingly, it decreases as temperature increases -- in contrast to simulations based on Kohn-Sham orbitals.

Multidisciplinary design analysis of coupled engineering systems requires the computation of equilibrium states in which all disciplinary coupling variables are mutually consistent. Conventional fixed-point iteration resolves this consistency problem separately at each design point, which can become expensive when disciplinary evaluations are costly and many analyses are required in outer-loop tasks such as multidisciplinary design optimization, uncertainty quantification, or digital twin updating. This paper introduces REMAL, a residual manifold surrogate modeling framework for coupled systems. Instead of approximating each discipline independently or directly learning converged coupling variables, the proposed method learns a surrogate model of the joint residual manifold via multitask Gaussian process models. An entropy-based active learning strategy selects additional residual evaluations near uncertain zero-contour regions, and equilibrium states for new design inputs are recovered by solving a nonlinear least squares optimization problem using only the trained surrogate. The method is evaluated on four engineering coupled system benchmarks: a satellite model, an aerostructural model, a finite-element gas-turbine heat-transfer and economics model, and a modified turbine model with added feedback coupling. Across these cases, REMAL consistently demonstrates the cost effectiveness when repeated evaluations of the fixed point across the design space are necessary. Theoretically, we show that, under mild assumptions, REMAL's predictive fixed point error is bounded.

Fibre lasers realise a large gain medium in a compactly coiled fibre. Disorder due to fabrication can negatively impact the stability of their lasing modes, especially in multi-core fibres. Recently, topological fibres (without gain) have been experimentally demonstrated to be robust against fabrication disorder, but topological fibre lasers have not yet been designed or modelled. Here, we use a combination of mode-coupling theory and finite-element simulations to design and model a topological laser based on a non-Hermitian Su-Schrieffer-Heeger (SSH) chain embedded in a photonic crystal fibre. Our design is based on a winding-number invariant in combination with a PT-symmetric SSH bulk. We show that the topological boundary mode is selectively amplified when extra gain is added at the topological interface. Even with nonlinearity added through saturable gain, the lasing supermode retains its robustness against disorder. We present a realistic design for a topologically robust fibre laser using readily available stack-and-draw methods with doped cores. This work establishes a new approach for imbuing non-Hermitian photonic systems with topological protection, with technological implications towards generating robust quantum and classical signals.

Skyrmions, topologically stable spin textures, have recently garnered significant attention in optics promising robust high-density information transition and nontrivial light-matter interaction. It was believed that the optical skyrmionic beams should be constructed by superposition of two orthogonal spatial modes with orthogonal polarizations to obtain topologically stable propagation. Here, we surprisingly find that propagation-stable skyrmionic beams can still be formed by superpositions of neither orthogonal spatial modes nor orthogonal polarizations. We theoretically present the mechanism to control the stable skyrmionics beams through the hybrid superposition of modes from the Hermite-Gaussian and Laguerre-Gaussian families and experimentally control the longitudinal on-demand dynamics of the skyrmions. This work redefines the topological stability of optical skyrmions, breaks limits and reduces the requirement for manipulating topologically structured light for practical multidimensional implementation of topologically robust information technologies.

Hard X-ray ptychography has become increasingly essential in both the life and physical sciences. However, pushing resolution down to a few nanometres often requires highly customized, chromatic diffractive or refractive X-ray nanofocusing optics, significantly limiting the practical broadband energy-scan applications. Here, we present the first known hard X-ray ptychographic imaging with a half-pitch resolution below 8 nm using total-reflection Advanced Kirkpatrick-Baez (AKB) mirror nanofocusing optics at the high energy photon source (HEPS), with clear potential for further extension. Despite leveraging the benefits of enhanced instrumentation, such as the high coherent flux of 4th-generation diffraction-limited storage rings (DLSR), state-of-the-art beamline X-ray optics and detectors, this is made possible by developing a reconstruction algorithm termed Generalized Wirtinger Projections (GWP). We derive the theory of GWP and experimentally demonstrate its capability for improved partial-coherence reconstruction and enhanced spatial resolution over conventional methods through imaging experiments on a Siemens star test chart at 12.4 keV. GWP provides a highly compact framework for jointly accounting for multiple coupled uncertainties that degrade resolution, enabling straightforward extension to other imaging modalities, such as burst ptychography, while delivering nearly an order-of-magnitude improvement in GPU memory efficiency. Furthermore, the ability to combine nanometre-scale spatial resolution with the inherently achromatic nanofocusing optics demonstrated in this work potentially opens new opportunities for in situ or operando broadband, energy-scan 3D spectroscopic imaging with element- or chemical-state specificity in complex environments at the nanoscale, holding significant promise for a wide range of applications from electronics and energy science to neuroscience.

Machine learning and computational inference, coupled with experimental data, promise to significantly accelerate our rate of learning in most scientific disciplines. In this study, we develop tools that connect microscopic observations to macroscopic device behaviour, a capability that is essential for accelerating the design of durable energy materials. To this end, we introduce a novel approach that integrates photoluminescence imaging with drift diffusion simulations to understand operation and degradation in fully fabricated perovskite solar cells. By employing Bayesian inference, we generate "inferred maps" of parameters that govern recombination processes present in devices. We track these parameter maps while the devices are aged (70 °C, full spectrum sunlight) to analyse their temporal evolution during degradation. Notably, our approach allows us to distinguish between degradation occurring at the hole or electron transporting layer interface, or within the bulk. Our analysis reveals pronounced spatially non-uniform degradation, with significant macroscopic heterogeneity observed in the optoelectronic parameter maps. We pinpoint the greatest degradation observed in specific regions to stem from the perovskite/transport layer interfaces. Finally, we demonstrate that an amino-silane molecular passivation treatment suppresses this degradation, highlighting its specific role in enhancing device stability. Our approach offers valuable insights for future device fabrication and is a clear exemplification of how advanced Bayesian inference can significantly increase the value of experimental data.

We theoretically investigate the transmission and diffraction of a weak probe field from an electromagnetically induced grating formed in a weakly driven two-level medium coupled to engineered quantum reservoirs. Using a perturbative solution of the optical Bloch equations in the weak-driving regime, we analyze how normal-vacuum, thermal, and broadband squeezed-vacuum environments modify the probe susceptibility and consequently reshape both the spatial transmission function and the far-field diffraction patterns. We show that reservoir statistics have a pronounced impact on the diffraction response by altering the amplitude and phase of the induced grating. Thermal reservoirs enhance the transmission modulation and increase the intensity of the dominant diffraction orders, whereas squeezed-vacuum reservoirs generate strongly phase-sensitive modifications that selectively redistribute optical power among diffraction channels. We further demonstrate that the detuning between the squeezed reservoir and the driving field provides an efficient mechanism for controlling diffraction directionality, leading to substantial amplification of selected angular orders. In two-dimensional geometries, squeezed-vacuum correlations produce highly structured phase landscapes and strongly anisotropic diffraction patterns, enabling directional enhancement of specific diffraction channels while suppressing others. These results establish reservoir engineering as a versatile approach for controlling transmission, diffraction efficiency, and angular selectivity in minimal two-level systems, with potential applications in programmable photonic devices, beam steering, and quantum optical platforms.

In this paper, we present an Intelligent Perception-Assisted Light Modulation (IPALM) technique, for femtosecond laser nanofabrication and particle manipulation. IPALM technique integrates real-time hand-motion recognition with dynamic spatial light modulation to achieve programming-free laser beam control. In contrast to the conventional programmed laser fabrication techniques which need setup a project prior to fabrication, IPALM offers a direct "mind-to-matter" pathway for laser nanostructuring and biological cells handling with high flexibility and multiple degree of freedom. In nanofabrication, IPALM provides high resolution with feature dimensions down to 280 nm. By mind-driven hand gesture control, the laser beam is regulated to enable direct writing of micro/nanostructures with a minimum feature size down to 280 nm via IPALM. In precise particle manipulation, multiple cells can be simultaneously moved to achieve cell coalescence. With these examples, IPALM showcases its potential for applications in photonics, biomedicine, and microfluidics, for high-dimension and flexible laser applications.

Bacteria live in environments that are constantly changing. To survive, they rely on different motility systems that let them move, explore, and interact with their surroundings. These motility systems not only control the movement of individual cells but also give rise to collective behaviors such as coordinated spreading, cooperative predation, and multicellular development. Understanding these processes requires not only a description of the underlying molecular machines, but also quantitative observations spanning single-cell behavior and collective dynamics. Imaging approaches now make it possible to follow motility across scales, from molecular machines to bacterial communities, while computational analyses extract principles that link mechanisms to emergent dynamics. By combining molecular, behavioral, imaging, and analytical perspectives, this review provides an integrated view of bacterial motility that links single-cell behavior to community-level dynamics across scales.

This short article - commissioned as part of the Quantum Flagship project OpenSuperQPlus - introduces the topic of fair access to quantum computing time and argues for its relevance in current European technology policy debates. Choices about access we make today affect which applications of quantum computing are developed in the future, and by whom. Among the issues at stake are: Europe's technological autonomy and sovereignty; Europe's future economic competitiveness; knowledge security and the future of open science; and delivering upon the EU's founding values and aims. The results of a survey of OpenSuperQPlus researchers reveal prevalent contractualist and transactional framings of access to European quantum computing capacity among those developing the technology. The article concludes with a call to go beyond framing access to quantum computing as a zero-sum-game, and to explore the possibility of an access regime premised on openness and solidarity.

Plasmonic metasurface-driven Optical Solar Reflectors (m-OSRs) offer a promising route towards lightweight and high-performance thermal management. By exploiting subwavelength structuring and intrinsic material losses, such systems enable tailored absorptance spectrum across the solar and thermal infrared domains, respectively. Here, a plasmonic m-OSR composed of an aluminum back-reflector, a silicon dioxide dielectric spacer, and a nanostructured aluminum-doped zinc oxide (AZO) layer is investigated. The optical response of the structure is governed by the interplay between reflection, localized surface plasmon resonances and Fabry-Perot cavity effects, leading to efficient spectral selectivity. An optimization performed with a multi-objective genetic algorithm yields a low solar absorptance of alpha = 0.16 combined with a high thermal emissivity of epsilon = 0.83, providing an alpha/epsilon ratio of 0.19. These results highlight the potential of plasmonic meta-OSRs as ultrathin, high-performance solutions for thermal management and in particular for the next-generation advanced spacecraft.

Omnidirectional flat-band resonances, characterized by an enhanced photonic density of states and inherent angular robustness, are highly sought-after in integrated nanophotonic devices, particularly when integrated with chiral functionality. Here we realize such resonances in a nonlocal silicon membrane metasurface patterned with periodic square-lattice air-hole arrays. Increasing the lattice period not only compresses the Brillouin zone but, crucially, weakens the evanescent coupling between neighbouring Bloch modes associated with the same-order guided resonances. Driven by the tight-binding model in the limit of weak inter-unit-cell coupling, the pronounced band flattening of the degenerate guided resonance along both $k_{x}$ and $k_{y}$ yields, giving rise to an omnidirectional flat-band resonance. Remarkably, both numerical simulations and experiments reveal a universal route for endowing flat-band guided resonances with optical chirality through the deliberate breaking of the mirror symmetry of air holes. As a result, the omnidirectional chiral flat-band resonance emerges along both principal in-plane directions, with $Q$-factors exceeding 10$^{3}$ and circular dichroism greater than 0.9 over a wide angular range of $\pm 5^{\circ}$. Nonlinear measurements further show that the resulting resonance not only drives highly efficient third-harmonic generation but also imparts a pronounced spin-selective character to the nonlinear process. Simultaneously, the highly efficient nonlinear process also enables chirality-controlled frequency-upconversion imaging. Our results establish a general paradigm for engineering omnidirectional chiral flat-band resonances in planar silicon platforms, opening new opportunities for nonlinear nanophotonics and chiral imaging.

The merits of thin-film surface acoustic wave (SAW) devices are pivotal to develop the high-performance intermodal acousto-optic modulators. In this work, we have proposed shear-horizontal (SH) SAW resonators for anticipated intermodal acousto-optic modulation on the thin-film lithium niobate platform. Through optimization of the cut angle of LN films, the SAW wavelength, and the thickness of interdigital transducer (IDT) electrodes, the calculated acousto-optic overlap factors utilizing SH0 modes are improved by more than an order of magnitude compared with those of Rayleigh modes. Furthermore, we have fabricated and characterized three kinds of proof-of-principle SH0 mode devices without/with grating reflectors. The electromechanical coupling coefficients (keff^2) and quality factors (Q) in the overtone resonators with grating reflectors are systematically evaluated, featuring the highest Q of 843 with the compromised keff^2 of 0.96%-4.72%. The results reveal that the temperature coefficients of frequency (TCF) of Rayleigh modes vary across various overtones, whereas the SH0 modes exhibit TCFs in the range of 32.3-68.9 ppm/C. Our fabricated SH0-mode overtone resonators demonstrate the capability of operating at power levels up to 29 dBm without electrode damage, offering a promising paradigm for robust and high-efficiency intermodal acousto-optic modulators with potential applications in integrated optical signal processing, microwave photonics,and quantum information technologies.

Here we address the problem of estimating the dimensionality and nature of parametric variation in an unknown generative process directly from time-series data, without specifying or fitting a model. In particular we suppose that inter-instance variation in collections of time series is caused by parametric variation in the generating model. We hypothesize that, given a sufficiently large library of time-series features, low-dimensional parametric variation will manifest as low-dimensional structure in feature space, enabling interpretable estimators of the underlying degrees of freedom to be constructed. We test our hypothesis using a library of over 7000 diverse and interpretable time-series statistics and thirteen simulated systems with known parametric variation, spanning linear stochastic processes, nonlinear oscillators, and chaotic dynamics. Our unsupervised, data-driven approach often reconstructs the underlying parametric variation across this extensive range of simulated dynamical systems while also yielding interpretable estimators for each underlying dimension. Applied to the movement dynamics of 1143 fruit flies, we use this method to extract biologically meaningful components corresponding to sex and circadian rhythmicity. Our results pave the way for much-needed data-driven methods to bridge the gap between interpretable theoretical understanding of dynamics and the large and complex datasets that characterize modern scientific problems.

A central challenge in nonlinear spectroscopy is the order-selective readout of weak higher-order responses that spectrally overlap with dominant lower-order signals. This bottleneck is particularly severe in two-dimensional (2D) spectroscopy, where extending conventional phase-cycling schemes to higher orders rapidly increases measurement and analysis complexity. Here we introduce a computation-assisted strategy that combines rotating-frame acquisition with a frame-shift tracking algorithm to separate signals by their frame-dependent spectral shifts. In a rubidium vapor experiment, we use this approach to isolate a 7th-order nonlinear contribution from coexisting 3rd-order components, enabling direct access to higher-order quantum-coherence dynamics without sacrificing operation at comparatively high pulse intensities. The method is broadly compatible with multidimensional spectroscopy platforms and provides a practical route to probing many-body and collective ultrafast dynamics beyond third order.

Accurate surface thermochemistry requires balanced error cancellation between extended slabs and molecular reference states. This balance can fail whenever the electronic-structure error is not transferable across the chemically distinct species entering a thermodynamic cycle. Here we examine this problem in single-determinant fixed-node diffusion Monte Carlo (SD-FNDMC) for oxygenated ORR intermediates on Pt(111). Gas-phase thermochemistry is used to diagnose the reference-state imbalance, and a hybrid cycle is introduced to separate slab-adsorbate binding from molecular formation. The hybrid cycle keeps the surface binding term at the SD-FNDMC level, where cancellation is expected to be most favorable, and replaces the molecular formation contribution with a benchmark coupled cluster reference. For Pt(111), the resulting correction is small for O and OH but larger for OOH, while the geometry-matched refinement gives only a secondary correction. Applying the same cycle to HCO and COH on Cu(111) gives corrections of opposite sign, showing that the bias is controlled primarily by the electronic structure of the molecular reference rather than by adsorbate geometry alone. This decomposition identifies molecular reference imbalance as a separable source of error in SD-FNDMC surface thermochemistry and reduces the corresponding bias without modifying the SD-FNDMC slab-binding contribution.

The resilience literature measures urban performance as recovery: the degree to which a city returns to its pre-shock baseline. This paper develops a stronger concept -- civic ascent -- as part of a broader research program on the ethology of coupled agent-environment systems, of which the city is the deepest available empirical instance. Civic ascent is defined as the condition in which a city emerges from shock with higher functional capacity than before. We develop a conceptual framework in the ethological tradition, treating the city as a coupled system of three slow state variables -- topos (physical structure), nomos (institutional structure), and hexis (civic judgment) -- together with a fast affective channel (delta) through which shocks to topos and nomos reach hexis. The framework distinguishes three structurally distinct pressures on civic systems: shocks (discontinuities in T or M), decay (continuous entropy), and leakage (active extraction of civic surplus into non-civic pools). The ascent condition is that reinforcement from cross-coupling of T, M, and H exceeds the combined loss from decay and leakage. Post-shock ascent is measured by a normalised improvement index A(T) applied to a composite civic performance signal P(t) constructed from scale-adjusted key performance indicators, distinguishing intrinsic civic ascent from demographically driven growth. New York City after September 11, 2001, is proposed as the primary empirical case; the operational measurement program is specified in the companion NYC Civic Data Map (Washburn 2026c, 133 KPIs) and executed in Paper 2. The reader for whom only the urban contribution is of interest will find it complete in itself; the reader interested in the larger program will find this paper its formal core.

Bichromatically pumped Kerr microresonators exhibit phase bistability and multistability arising from four-wave mixing between pump and generated modes. Here we analyze the symmetry structure of stationary solutions in the coupled-mode description of such systems. We show that the two-pump coupled-mode equations are a particular case of a more general model with pumps placed at equally spaced modes. For this model, any stationary solution generates a finite family of stationary solutions through a discrete phase transformation. This transformation leaves the equations invariant and preserves the stability type of the corresponding stationary states. As a consequence, the possible stationary values of each individual mode form a regular polygon in the complex plane. The number of vertices is determined by the order of the mode index $μ$ in the group $\mathbb{Z}_n$, where $n$ is the separation between the pumped modes. Thus, the phase multistability structure depends not only on nonlinear dynamics but also on the arithmetic relation between the mode index and the pump separation. These results provide a simple symmetry-based explanation for the discrete phase structure observed in multimode Kerr resonators under bichromatic pumping.

Efficient and stable microwave-optical transduction is a key enabling technology for distributed superconducting quantum computing and heterogeneous quantum networks. Electro-optic transducers based on thin-film lithium niobate (TFLN) have shown strong promise, but demonstrations to date have been limited by various factors such as low frequency bias drift, low efficiency, fabrication complexity, and scalability. Here we demonstrate the first integrated electro-optic microwave-optical transducers realized in thin-film lithium tantalate (TFLT), a material platform offering Pockels nonlinearity comparable to TFLN together with improved bias stability and high-power handling. We fabricate superconducting microwave resonators coupled to tunable photonic-molecule optical resonators using wafer-scale deep ultraviolet lithography, offering high-throughput production of hundreds of devices per wafer. Across six devices we observe coherent bidirectional conversion between C-band optical photons and 4.9-5.5 GHz microwave photons, with measured on-chip efficiencies and inferred single-photon coupling rates g_0/2π ~ 1 kHz consistent with theory. Continuous operation over multiple days is achieved using a static bias field with minimal feedback, demonstrating a major operational advantage. We further characterize optical loss statistics, microwave resonator performance, and optically induced added noise under pulsed pumping, finding less than one added photon for 100 microsecond pulses at the highest measured efficiencies. These results establish TFLT as a scalable and robust electro-optic platform for future quantum interconnects and modular quantum processors.

We present an asymptotic analysis and high-order boundary integral method for the three-dimensional Neumann Green's function in general geometries. The Neumann Green's function is a fundamental quantity which arises in numerous fields of science and engineering. In the application of singular perturbation methods to strongly localized reactions and diffusive transport, the Green's function plays the key role in mediating global dynamics. However, this essential quantity can only be determined in closed form for a limited set of geometries. The Green's function for the Laplacian is an elliptic problem with a Dirac forcing term. Accurate resolution of the solution requires a careful decomposition into a singular and a regular part. The bulk scenario is where the source is placed off surface and the singularity is given by the free-space function. In the surface case, where the source is placed at a curved point on the boundary, we use asymptotic analysis to determine a three-term singularity structure. With explicit knowledge of these singularities, we develop a high-order boundary integral method for the determination of the remaining regular part. To resolve the singular boundary data, our integral method uses a custom discretization with Duffy patches near the source. We validate our method using several test cases in which closed form solutions can be developed, including spheres, prolate spheroids and constructed domains. We demonstrate the applicability of our method to address some open problems in narrow capture theory.

We report on a phase-based lock-in thermography approach, combined with a multilayered thermal model (often employed in thermoreflectance analysis), to measure the thermal conductivity of bulk materials and layered structures. The spatial distribution of the material's thermal phase is monitored with an infrared camera, which is locked into the frequency of a modulated laser used to heat the material. This phase distribution is then fit with a thermal model, in which properties such as thermal conductivity are extracted as fit parameters. This approach enables non-contact, front-side measurements, which are insensitive to surface roughness. The technique does not strictly require the application of a transducer layer, but we highlight the practical benefits of applying a removable adhesive layer to serve as a near-surface absorber. We demonstrate the efficacy of the method by measuring materials with thermal conductivities that span over three orders of magnitude (approximately 1 W/m/K to > 2000 W/m/K).

Starting from our previously proposed line element and considering more ``surface color'' datasets, we derive a simplified version which matches experimental datasets equally well and resulted into a conformally-Euclidean line element, which is conceptually much simpler than any existing color difference metrics. The color difference is written as an Euclidean difference multiplied with a simple factor which depends on the luminance only. In a subspace with constant luminance, as considered by MacAdam, this factor becomes constant and the subspace is flat. The same holds for sufficiently large luminances. Based on this LE we derive perceptual coordinates $\left(A,l_{c},s_{c}\right)$ very similar to the CIELab $\left(L^{*},a^{*},b^{*}\right)$.

We develop a four-compartment susceptible--casual--addicted--resistant (SCAR) model for adolescent substance use in a high-school setting. The model divides students into susceptible non-users, casual or experimental users, students with sustained or substance-use-disorder (SUD)-level involvement, and resistant students in protective anti-use environments. It includes peer-driven initiation, escalation from casual to problematic use, protective peer influence, school disengagement, and partial re-entry after rehabilitation. Qualitative analysis and bifurcation diagrams show three main results. First, the return parameter \(φ\) separates two regimes: when \(φ=1\), the total population is conserved and interior equilibria may exist; when \(φ<1\), problematic use causes net school-population loss, so positive scaled equilibria may not represent true endemic equilibria. Second, initiation and escalation are governed by distinct thresholds, meaning first use and progression to problematic use are dynamically different. Third, the model can exhibit multistability, including bistability between a substance-free state and a stable high-use state, so long-term outcomes may depend on initial conditions. These findings suggest that effective school policy should combine universal prevention, early intervention for casual users, targeted support for students at risk of problematic use, recovery-supportive environments, and strong school re-engagement pathways.

This paper introduces peri urban dislocation as a structural condition that complements existing sprawl metrics by capturing hierarchical misalignments between inner city and peripheral areas. Whereas conventional measures emphasise density, land-use mix, or fragmentation, peri-urban dislocation reflects deeper divergences in the core periphery relational functional organisation of urban systems. We operationalise this concept using percolation analysis of street networks, revealing hierarchical patterns via clustering maps and dendrograms, providing a relational structure between urban elements. Two case studies, Valdivia, Chile, and Boston, USA, demonstrate contrasting manifestations: a structural reversal in Valdivia, where a homogeneous residential periphery dominates the hierarchical clustering process, and peri-urban voids in Boston, where isolated parcellations persist despite metropolitan consolidation. These findings position peri-urban dislocation as a structural dimension linked to sprawl yet distinct from metrics based on density or peripherality; one that may occur independently or represent a previously unidentified structural signature of sprawl. Methodologically, we apply established percolation techniques to expose this previously unarticulated structural phenomenon, enabling the detection of hierarchical misalignments within urban systems. Conceptually, we introduce peri urban dislocation as a new dimension of urban structure, helping articulate debates on sprawl and peri urbanisation through a complexity informed lens and enabling core periphery diagnostics across diverse urban contexts.

The efficiency of multi-photon emission is an important characteristic of quantum light sources. Bright multi-photon emission is desirable for high-power lighting and lasers, while its complete suppression is required for high-purity single-photon generation. In colloidal quantum emitters, multi-photon emission can vary significantly between individual particles. Resolving this heterogeneity remains challenging with conventional particle-by-particle approaches. Here, we introduce a crosstalk-suppressed SPAD-array photon-correlation approach for high-throughput quantification of multi-photon emission from more than 1000 colloidal quantum shells. By projecting two images of the same sample onto distant regions of the detector array, we avoid short-range crosstalk between detector pixels. Time gating suppresses dark-count coincidences and distinguishes individual emitters from clusters. Applying this method to quantum shells reveals a near-Gaussian distribution of biexciton emission efficiencies, with a mean of 0.55 and an estimated intrinsic standard deviation of 0.12. Intra-batch correlations between the biexciton efficiency and the particle brightness are consistent with the volume scaling of Auger quenching. These results establish SPAD-array photon correlation as a scalable route to resolve multi-photon heterogeneities in nanoparticle ensembles.

This work presents a finite element formulation for coupled chemo-mechano-thermodynamical large deformation contact. The formulation is based on the contact theory of Sauer et al. (2022) that contains six coupled (but separate) fields: the deformation and temperature of the two contacting bodies, as well as an interfacial bonding field and interfacial temperature. The latter is governed by the chemical and mechanical energy dissipation at the interface. Here the focus is placed on the evolution of bonding and debonding, and how it is coupled to the mechanical and thermal contact state. Several elementary models are proposed for this based on a quadratic contact potential. The resulting contact formulation becomes very general and versatile, which is illustrated by several challenging examples. They include pressure- and gap- depended bonding, exothermic bonding reactions, thermal hardening and thermal expansion, as well as simultaneous bonding and debonding. They are based on a monolithic finite element implementation using classical and isogeometric shape functions together with implicit time integration. Its full linearization, required for the Newton-Raphson solution method, is also provided. If bonding sites are material points, the bonding variable can be condensed-out locally.

Kinetic collisionless plasma equations are commonly solved via spectral methods in velocity space. The most commonly used spectral method is based on Hermite polynomials with a Maxwellian weight, as this basis efficiently represents near-Maxwellian distributions with relatively few degrees of freedom. An alternative approach uses Legendre polynomials, which are better suited for resolving strongly non-Maxwellian features. In this paper, we propose a mixed method that combines the Hermite and Legendre expansions. The mixed method is particularly advantageous for problems in which non-Maxwellian features are localized in velocity space, such as beams and plateaus. We demonstrate analytically and numerically that the mixed method conserves total mass, momentum, and energy by imposing certain constraints. The numerical results show that, for the same number of degrees of freedom, the proposed mixed method can achieve improved accuracy in comparison to the individual Hermite or Legendre methods, while maintaining comparable computational cost.

Laser-induced forward transfer (LIFT) is a nozzle-free laser-assisted printing method that provides an advanced manufacturing route for spatially selective deposition of functional inks, nanoparticle suspensions, polymers, hydrogels, biological materials, and other difficult-to-nozzle formulations. The apparent simplicity of LIFT, however, conceals a strongly coupled laser-liquid interaction. Laser energy is absorbed within a confined donor architecture, converted into thermal and plasma responses, and then transformed into bubble-mediated motion of the donor material. The cavitation bubble provides the transient mechanical bridge between optical energy deposition and the hydrodynamic ejection process. This chapter presents LIFT from a multiscale perspective centered on bubble dynamics and material ejection. It first reviews major LIFT donor architectures. Then, it examines how donor ribbon design, absorbing-layer properties, laser parameters, material rheology, control bubble inception/growth, jet formation, droplet breakup, and final deposition. Modeling approaches are discussed as tools for connecting experimental observations across time and length scales, ranging from reduced-order estimates to interface-resolving simulations and data-driven process maps. As one illustrative mechanistic example, thermal-only, plasma-mediated, and coupled plasma-thermal-thermoelastic frameworks for early-stage bubble inception are briefly compared to show how different inception assumptions can provide initial conditions for downstream bubble growth and jetting models. This chapter concludes by identifying opportunities for bubble-aware donor design, time-resolved diagnostics, benchmark datasets, and predictive LIFT process maps based on intermediate bubble and jet observables.

The Stokes parameters are three real parameters that completely characterize partially coherent optical fields spanned by two modes -- whether a pair of polarization or spatial modes -- and their use is thus ubiquitous in optics. Because the Stokes parameters are defined through an expansion of the $2\times2$ coherence matrix in terms of the Pauli matrices, they cannot be applied to optical fields comprising three modes, which are described by a $3\times3$ coherence matrix. Examples of such fields include the polarization of non-paraxial fields (spanned by three orthogonal polarization modes), and fields comprising three spatial or temporal modes. It has long been theorized that the $3\times3$ Gell-Mann matrices -- developed in high-energy particle physics -- can serve as a basis for $3\times3$ optical coherence matrices, with 8~expansion coefficients known as the Stokes-Gell-Mann (SGM) parameters, but the measurement procedure is daunting, and the SGM parameters have not been measured directly to date in optics. Here we present the first measurements of the SGM parameters for partially coherent three-mode light in a photonic integrated platform comprising a hexagonal mesh of Mach-Zehnder interferometers. Measuring the SGM parameters on chip, from which we reconstruct the $3\times3$ coherence matrix facilitates exploring the full space of iso-entropy fields that can be inter-converted into each other unitarily, and those that share the same value of entropy and yet cannot be inter-converted unitarily. These results pave the way to utilizing multimode partially coherent light in applications involving optical communications, sensing, and information processing.

Positron annihilation lifetime spectroscopy (PALS) spectra are commonly analyzed by fitting multi-exponential lifetime models convoluted with the detector resolution function. In practice, this inverse problem is sensitive to initial parameter choices, parameter bounds, source corrections, and correlations between lifetime and intensity parameters. This paper presents fitPALSpectra, an open-source Python workflow for configurable PALS spectrum simulation, fitting, visualization, and reporting. The implementation uses an analytically integrated exponential--Gaussian response model, configurable source and sample components, constrained optimization, optional least-squares refinement, and machine-readable output of fit results, correlation matrices, and fitted curves. Validation on fully synthetic spectra with known ground-truth parameters shows accurate recovery of the simulated lifetimes, intensities, detector full width at half maximum, prompt shift, and background.

The chaotic rotation of Saturn's moon Hyperion is a paradigmatic example of Hamiltonian chaos in a natural system. Although its tumbling motion is well established theoretically, identifying a robust observational signature of chaos from sparse and noisy astronomical time series remains a major challenge, making phase-space reconstruction techniques impractical under realistic conditions. In this work, we show that multifractal detrended fluctuation analysis (MFDFA) provides an effective alternative for detecting chaotic dynamics directly from photometric observations. Using historical ground-based light curves and synthetic datasets, we demonstrate that the intermittency associated with chaotic tumbling produces a broad multifractal singularity spectrum. While multifractality is a known feature of Hamiltonian chaos, we show that it can serve as a practical observational diagnostic when traditional chaos indicators fail because of sparse sampling. In particular, the multifractal spectrum remains detectable after realistic observational filtering and distinguishes chaotic tumbling from aliased regular rotation. By contrast, regular resonant rotation exhibits a significantly narrower spectrum, approaching the monofractal behavior expected for uncorrelated noise. For the observational data, we measure a broad spectral width consistent with the synthetic chaotic model, statistically distinct from surrogate datasets, and robust against finite time-series length. These results establish multifractal scaling as a viable observational signature of Hamiltonian chaos in sparse astronomical datasets, bridging nonlinear dynamics and planetary photometry.

A theoretical framework is developed to analyze the generation of the optical GKP state using a single-photon-added squeezed vacuum. This state, defined by the squeezing parameter $r$, is injected into a 50:50 beam splitter, and the optical GKP state is obtained through conditional measurement at one output port. The single-photon-added squeezed vacuum is especially prominent in this context because it provides a simpler and more experimentally accessible ingredient than Schrodinger cat states, while conditional measurement ensures projection onto a state that closely approximates the finite-energy GKP form. Fidelity is employed to quantify this closeness, and the analysis demonstrates that the scheme achieves a maximum fidelity of 85% at a squeezing level of $3.76 \ \text{dB}$. This performance surpasses approaches based on squeezed optical odd Schrodinger cat states, underscoring the single-photon-added squeezed vacuum as a practical and effective pathway toward fault-tolerant photonic quantum computing.

Human behaviour and epidemic dynamics are intertwined, yet accounting for this feedback remains one of the key challenges of epidemiological modelling. The COVID-19 pandemic was an opportunity to overcome the traditional limitations of the field, raising expectations that data-informed endogenous approaches to behaviour modelling would advance substantially. To quantify the progresses made, we conducted a systematic review of SARS-CoV-2 transmission models endogenously including human behaviour in response to epidemic dynamics. The COVID-19 pandemic saw great strides in terms of the expanded use of empirical data in epi-behavioural modelling. However, it also showed shortcomings with respect to limited use of behavioural empirical data, lack of innovation in model structure, and limited engagement with other disciplines and decision-makers. Overall, our results suggest that identifying priorities in model design and behavioural data, building an adequate data collection infrastructure, leveraging on AI advancements, and fostering interdisciplinarity are strategies of utmost importance for pandemic preparedness.

This paper studies sovereign stress avalanches and network amplification in Latin American credit markets using monthly J.P. Morgan EMBI Global Diversified spreads for eleven sovereigns over 2007-2026. Country stress events are defined as positive log-spread innovations exceeding country-specific volatility thresholds, and regional avalanches count the number of stressed countries in each month. The empirical design combines finite-sample power-law diagnostics, threshold robustness checks, a country-level reshuffling placebo, and rolling correlation, partial-correlation, and minimum-spanning-tree networks. Avalanche sizes are heavy-tailed, with an estimated exponent of 1.77, while spread changes and inter-event times lie in a heavy-tail boundary regime. The placebo shows synchronization far above independent stress timing, with p-values below 0.001. Large avalanches coincide with denser and more spectrally amplifying raw-correlation networks, but not after partial-correlation filtering, indicating common-factor co-movement rather than conditional regional propagation. Network metrics describe contemporaneous stress regimes rather than early-warning signals. The results provide a finite-size criticality framework for monitoring sovereign fragility in emerging markets.

In recent years, neural ordinary differential equation frameworks such as Biologically-Informed Neural Networks (BINNs) have shown promise for learning mechanistic laws from sparse data. However, most existing approaches implicitly assume homoscedastic Gaussian noise, and therefore do not account for potentially meaningful structure in biological variability. Here, we present an extension to the existing BINNs framework that includes a learnable noise model, allowing discovery of the noise model directly from data. Using population growth as an example, we demonstrate that the framework accurately recovers the underlying noise structure and improves predictions of the underlying growth laws compared to existing approaches. As such, this work establishes a general likelihood-based framework for jointly learning dynamics and heteroscedastic noise within mechanistic neural network approaches.

AI decision-support systems can benefit from anticipating biases in human decision-making. Many such biases may arise from human cognitive limitations. The policy compression framework models decision-making as a trade-off between reward maximization and the cognitive cost of encoding state-dependent action policies, formalized as the mutual information between states and actions (policy complexity). We argue that this account is incomplete because it treats conditional entropy--the irreducible uncertainty about which action should be selected given a state--as costless, even though empirical evidence suggests that it modulates reaction times. We therefore extend the framework by defining cognitive cost as the sum of policy complexity and a weighted conditional-entropy term, governed by a new parameter, $η$. The resulting optimal policy retains the standard exponential form but becomes sharper as $η$ increases, allowing policy precision to vary more independently of reward sensitivity. This modification implies that the standard policy compression framework may underestimate the cognitive cost of action selection, and it has the potential to better account for biases in human decision-making. At the same time, it introduces additional complexity for fitting the model to human data, which future work will need to address.

Computational design of nanobodies that bind user-specified protein epitopes could transform therapeutic development, but current methods either rely on stochastic sampling requiring days of GPU computation or inverse folding approaches unable to target epitopes directly. Here we present EasyNano, a practical pipeline for rapid, epitope-targeted nanobody complementarity-determining region (CDR) design that operates in approximately 10-20 minutes on a high-end personal workstation. EasyNano optimizes CDR residue logits via gradient descent through the ESMFold2 pairwise distance distogram, using the lightweight ESMFold2-Fast model (721M) as a differentiable oracle guided by a composite loss including a dedicated epitope proximity term. A full ESMFold2 (1.3B) CA-coordinate structure prior prevents framework pose drift. The wild-type logit initialization bias emerges as a critical practical parameter controlling CDR mutability. Across six target-framework pairs spanning self-recovery and de novo design scenarios, EasyNano improves ipTM by up to +0.559 -- from 0.143 to 0.702 (Ty1/RBD) -- and achieves a 4.6-fold improvement (ipTM 0.117 to 0.538) on a manually docked AQP4-targeting framework, while preserving ipTM on already-strong binders. Random CDR baselines (n=30 per target) confirm statistical significance (5.7 sigma above random mean for Ty1). Multi-seed analysis reveals diverse local minima, underscoring the importance of replicate runs. Kabsch cross-validation against crystal structures confirms that designed CDRs preserve the framework pose basin. EasyNano demonstrates that ESMFold2-based differentiable optimization provides a fast, practical, and epitope-specific approach to nanobody CDR design.

A key question in theoretical biology is how effectively biological systems preserve information about their inputs while operating under physical and functional constraints. We examine that question at the neuromuscular junction (NMJ) by studying how neurotransmitter concentration is transformed into current at both cholinergic and glutamatergic NMJs. An information maximization analysis was used to derive a theoretical distribution over neurotransmitter concentrations based on biological understandings of dose-response relationships. These theoretical distributions were compared to an experimentally derived distribution obtained from a Drosophila NMJ. The theoretical and experimental distributions showed very little agreement, indicating that the Drosophila NMJ does not shape its distribution of synaptic vesicle release probabilities in order to maximize information transmission from nervous system to muscle. Predictions for cholinergic systems are provided.

Neural assemblies, transiently coordinated groups of neurons, observed in the hippocampus are thought to underlie the formation of episodic memories. Acetylcholine (ACh), a neuromodulator, that is received by the hippocampus, plays a critical role in memory and learning. A well supported hypothesis suggests that high levels of ACh during active exploration and rapid eye movement (REM) sleep promote memory encoding, while low levels during quiet waking and slow-wave sleep (SWS) support memory consolidation. We study this bidirectional role of ACh in neural assembly formation through its effect on the synchrony among neurons. We consider a network model of pyramidal neurons, each equipped with a slow, voltage-dependent, non-inactivating potassium current (M-current), which is downregulated in the presence of ACh. Neural assemblies are represented as cluster solutions to this system. Using a one-dimensional phase model reduction of a pair of weakly coupled pyramidal neurons under different levels of the M-current, we predict the symmetric cluster solutions that may emerge in larger networks equipped with all-to-all globally homogeneous, symmetric distance-dependent and nearest-neighbours coupling architectures. We find that under low ACh conditions, the network can fully synchronize, whereas high levels can desynchronize the network into multiple stable symmetric cluster solutions representing distinct neural assemblies.

Evolutionary accumulation models (EvAMs), also known as cancer progression models (CPMs), infer dependencies in the order of accumulation of mutations during tumor progression from cross-sectional data. It has been suggested that EvAMs could be used to identify therapeutic targets, but there is no procedure in the literature for how to extract predictions under intervention from these models. A simple approach of conditioning on the absence of a mutation gives incorrect predictions. We address this gap by formalizing what ``intervene'' means for all currently available EvAM methods (OT, OncoBN, CBN, H-ESBCN, MHN, HyperHMM, HyperTraPS), using Pearl's do operator and conditional interventions. For each model, we show how to implement the intervention (in most cases as specific parameter modifications), identify equivalent implementation procedures, and analyze whether the modularity assumption -- required for the intervention to be well-defined -- is justified. Drawing on individual-level causal DAGs that make fitness an explicit variable, we distinguish two types of intervention (killing and inactivating) that are conflated in standard EvAM representations. Since the goal is to prioritize intervention candidates, we recast the problem as one of ranking: we define three intervention objectives and provide a protocol for evaluating how well EvAMs rank targets. Our framework is not specific to cancer or EvAMs; it applies wherever fitted computational models can be interpreted as structural causal models. Code available from https://github.com/rdiaz02/scm-interv-evams.

Performing statistical inference is an essential component of data science. Our focus in this work is on two inference techniques, viz. regression and interpolation. We propose a reaction network based approach that can implement linear regression (both univariate and multivariate) and linear interpolation. We do this by encoding the steady state concentration of species as the output of these inference techniques. Towards this, we use a novel generalized division module that can handle division of negative numbers. We verify our results by comparing them with in-silico implementation on standard synthetic datasets.

In this paper, we investigate a secure integrated sensing and communication (ISAC) system in which multiple communication users (CUs) coexist with multiple untrusted sensing users (SUs) that may eavesdrop on the confidential information intended for the CUs. To promote security fairness among users, we formulate a max-min secrecy rate optimization problem subject to a transmit power budget and sensing quality requirements characterized by beampattern matching error constraints. The resulting design problem is highly non-convex due to the secrecy rate expressions and non-convex sensing constraints. To address these challenges, we first reformulate the problem using semidefinite relaxation (SDR). Based on the reformulated problem, we develop a branch-and-bound (BB) framework combined with convex relaxations to obtain the globally optimal solution within a prescribed accuracy. To further reduce computational complexity, we propose a low-complexity algorithm based on successive convex approximation (SCA), which iteratively solves a sequence of convex subproblems and converges to a local solution. Numerical results demonstrate that the proposed BB algorithm achieves the global optimum and provides a benchmark for performance evaluation. Moreover, the proposed SCA-based algorithm attains near-optimal secrecy performance with significantly lower computational complexity, making it attractive for practical ISAC deployments.

The frequency bands between 7 and 24 GHz, also known as upper midband or Frequency Range (FR) 3, are being considered as an enabler of 6th Generation (6G) mobile networks. This portion of the spectrum exhibits different propagation characteristics compared to frequencies above 24 GHz, while also offering the potential to provide larger bandwidth allocations for mobile systems than those available in the sub-6 GHz range. 6G technology and spectrum policy, however, will need to guarantee coexistence with the incumbents that already use these frequency bands, which include a variety of services, from radiolocation to satellite-based communications, remote sensing, and radioastronomy. In this paper, we consider the challenge of coexistence between 6G terrestrial systems and satellite incumbents in different portions of the FR3 bands. Using a large-scale 3D model of a terrestrial deployment in the city of Boston and an open-source ray tracing solution, we evaluate the level of Radio Frequency Interference (RFI) that tens of terrestrial Next Generation Node Bs (gNBs) generate toward satellites at different elevation angles. Our model, based on realistic obstruction, clutter, diffraction, and reflections, shows that sidelobes and Non-Line-of-Sight (NLoS) paths can significantly contribute to RFI. Besides directionality, the spatial distribution of gNBs also plays a key role in defining the RFI levels, suggesting that a careful design and operation of terrestrial deployments can create coexistence opportunities.

Low-resolution data converters can significantly reduce the power consumption and silicon area of all-digital massive multi-user (MU) multiple-input multiple-output (MIMO) basestations. However, the existing literature almost exclusively focuses on idealistic quantization models, neglecting the inherent non-idealities present in real-world analog-to-digital converter (ADC) implementations. To overcome this limitation, we propose two affine models, one based on Bussgang's decomposition and one that maximizes the signal-to-distortion ratio (SDR), both accounting for the most prominent non-idealities in successive approximation register (SAR) ADCs. Subsequently, we utilize these models to devise low-complexity methods that mitigate SAR-ADC non-idealities in massive MU-MIMO wireless systems.

Affine frequency division multiplexing~(AFDM) has emerged as a compelling waveform candidate for future wireless networks, owing to its strong resilience to doubly selective channels and its ability to enable the seamless integration of communication and sensing functionalities. Against this context, this article provides a systematic study of AFDM from a standardization perspective. We first introduce the principles of AFDM and discuss the major considerations involved in waveform standardization. We then examine the backwards compatibility of AFDM with 4G/5G multi-numerology frameworks and their anticipated evolution, frequency-modulated continuous-wave (FMCW) radar waveforms, and long-range (LoRa) modulation, demonstrating that AFDM can be incorporated into legacy processing chains with limited modification. Key standardization-critical capabilities are further discussed, including multiple-antenna and multi-user support, and peak-to-average power ratio (PAPR). Finally, we investigate the potential of AFDM in several emerging scenarios, including non-terrestrial networks~(NTN), integrated sensing and communications (ISAC), vehicle-to-everything (V2X), and underwater acoustic (UWA) communications, whereby severe delay-Doppler dispersion places stringent demands on waveform robustness. Through these explorations, it is shown that that AFDM represents a timely and compelling technology for future wireless networks.

Practical implementations of phased arrays suffer from per-antenna gain, phase, and delay mismatches, which can significantly worsen the maximum sidelobe level (SLL) of beampatterns. The existing literature either analyzes specific structured mismatch patterns or derives per-angle marginal statistics under random mismatches, which fail to characterize global beampattern metrics such as the maximum SLL. To address this limitation, we propose a frequency-domain framework in which the beampattern is described by a tapering-window-dependent base function evaluated along a deformation determined by the array architecture and signal bandwidth. This formulation enables a spectral analysis of mismatches, revealing that element-wise errors generate weighted replicas of the ideal beampattern whose amplitudes are given by the discrete Fourier transform of the mismatch sequence. Building on this insight, we derive an approximation of the maximum SLL distribution under random gain and phase mismatches. The resulting expressions enable yield-oriented design and rapid design-space exploration without relying on computationally intensive Monte-Carlo simulations.

Chopper stabilized amplifiers are popularly used for realizing amplifiers with low offset and for rejecting flicker noise. One of the main limitations of these amplifiers is the low Input Impedance (Zin) produced by the switch capacitor input network. Zin here is resistive due to the switch capacitor action and is inversely proportional to the product of Chopping frequency (Fch) and Input Capacitance (Ci). Since Fch should be greater than the flicker noise corner frequency, this results in a low Zin. When interfacing sensors with high Sensor Output Impedance (Zo), chopper stabilized amplifiers load the sensors resulting in reduced sensitivity. This paper presents a novel input impedance boosting technique - Differential capacitor flipping technique for chopper based Capacitively Coupled Instrumentation Amplifier (CCIA), which prevents discharge and recharge of Ci's in every cycle by reconfiguring the capacitor positions while preserving the chopping operation. This ideally results in a purely capacitive Zin which is independent of Fch. The proposed architecture is used to demonstrate Electrocardiogram (ECG) signal acquisition with dry electrodes that have Zo in the order of a few Mega Ohms. This circuit implemented in TSMC 65 nm CMOS technology node features Zin of 21 GOhms at DC. The circuit has a power consumption of 2.6E(-6)W (2.8E(-6)W including clock generation circuits), with 7.2E(-6)Vrms (1 Hz-150 Hz) of total integrated input referred noise. ~

6G is expected to bring unprecedented advancements in the capabilities of vehicular networks. However, the advent of 6G will also introduce changes in the operation of vehicular communication infrastructures such as roadside units (RSUs), including the incorporation of autonomous intent-based network paradigm and integrated sensing and communication (ISAC) capabilities. While ISAC enables sensing and communication within a single 6G network node, intent-based network design paradigm ensures that network nodes such as RSUs, act as autonomous cognitive agents to fulfill the objectives of their respective communication service providers. This paradigm shift necessitates the development of V2I communication strategies that learns and adapts to the sensing-assisted communication and the autonomous decision-making strategies of RSUs. We model the RSU as a constrained utility maximizer, where the utility function characterizes the RSU intent, and formulate an inverse learning (IL) problem to infer the underlying utility function from observed ISAC RSU actions, for example the adaptive beamwidth allocation in response to the kinematic states of vehicles within a vehicular micro-cloud (VMC). The main contributions of this paper are: (i) ATIL, a nonparametric method based on Afriat theorem for fixed utility learning; (ii) FICNNIL, a parametric approach using fully input-concave neural networks, for structured fixed utility learning; and (iii) PICNNIL, a parametric approach based on partially input-concave neural networks, for inverse learning of state-dependent utilities. (iv) Federated inverse learning algorithms FedFICNNIL and FedPICNNIL for fixed and state dependent utility, respectively. We demonstrate the proposed IL-based framework for two V2I communication applications in VMCs, namely predictive scheduling for cooperative data downloading and dynamic cluster-head selection.

Scanning Photothermal Radiometry (SPR) is an active thermal technique that is simultaneously non-destructive, contactless, and allows for temporal resolutions on the order of nanoseconds, spatial resolutions down to the sub-micrometre scale and at different depths. This scanning method can be time consuming thus this work shows that it is possible to reduce the amount of measurements taken by 6 when using SPR on a sample consisting of carbon fibres in an aluminium matrix. It uses irregular sampling on sparse signals, and a weighted random technique to further decrease the amount of samples needed.

Gears are an integral component of electromechanical applications, but accurate condition monitoring methods, including data-driven predictive maintenance, are strongly dependent on high-quality data, especially from faulty components. To address the scarcity of data, we proposed a multibody simulation using an advanced polygonal contact method to replicate torsional vibrations from an experimental azimuth thruster test rig. The key novelty is the ability to simulate both healthy and faulty gears with arbitrary fault geometries. The simulated signals closely matched the measurements in both time and frequency domains. In the time domain, average torque levels and periodic fluctuations aligned well, although measured signals exhibited higher peak-to-peak amplitudes and greater noise, particularly in healthy conditions at lower rotational speeds. In the frequency domain, the simulations accurately reproduced expected fault frequencies and corresponding sidebands, with larger faults producing higher amplitudes. While the simulations tended to overestimate peak amplitudes and underestimate external noise, the results were highly comparable to measurements and consistent with the physical expectations. These findings provide a robust foundation for enhancing data-driven condition monitoring methods, particularly those employing machine learning or deep learning.

Driven by the massive video transmission requirements in the Internet of Everything, semantic communication holds great promise for striking a balance between transmission efficiency and quality. This paper introduces a large-model-driven generative video semantic communication (LGVSC) framework, enabling efficient video semantic transmission under extremely low bandwidth conditions. First, by decoupling the encoder and decoder as well as exposing explicit intermediate semantic representations, LGVSC maintains interpretability, avoiding the black-box behavior commonly observed in end-to-end systems. Next, we introduce a new metric, i.e., the probability-based semantic similarity score (PSSS), which quantifies semantic similarity for complex modalities within a continuous range, allowing for more precise evaluation of semantic content. Building on PSSS, we propose a semantic-guided keyframe extraction module driven by a multimodal large model. This module can enhance fine-grained semantic consistency during keyframe selection at the transmitter, optimizing transmission bandwidth without compromising semantic fidelity. Additionally, we design a generative large-model-driven dynamic semantic-adaptive decoder at the receiver, which can adapt to videos of arbitrary lengths. Simulation results demonstrate that LGVSC significantly outperforms traditional schemes, achieving a channel bandwidth ratio on the order of 10^-4 to 10^-3, while maintaining strong zero-shot generalization across downstream tasks.

The monomial parameterization of finite-memory Volterra identification is ill-conditioned under non-Gaussian input, and the Wiener--Hermite expansion removes this ill-conditioning only for Gaussian white-noise input. We construct the distribution-matched Volterra--Wiener--Kunchenko (VWK) basis by oriented Gram--Schmidt orthogonalization of monomials in $L^2(P)$ and use it as an arbitrary-polynomial-chaos coordinate system for finite-memory Volterra identification from data, following the generalized polynomial chaos of Xiu and Karniadakis (2002) and the data-driven arbitrary polynomial chaos of Oladyshkin and Nowak (2012). The basis itself is classical; the contribution is the Volterra-estimation reading. First, an order-2 misspecification-penalty theorem shows that a self-normalized diagonal estimator in the variance-matched Gaussian basis incurs an excess $L^2(P)$ risk governed by the skew coefficient $δ=μ_3/σ^2$, vanishing exactly for symmetric inputs. Second, conditioning experiments separate the constructional fact that the population matched Gram is the identity from the finite-sample design Gram: at $n=2000$, the centered-exponential empirical VWK Gram remains far better conditioned than the power Gram, although it degrades with degree. Third, a machine-checked Lean 4 proof establishes the Binomial$(N,p)$ Krawtchouk row for arbitrary $N$. Full least squares over a fixed span is basis-invariant, so VWK stabilizes diagonal cross-correlation and regularized coordinate fits rather than claiming universal prediction superiority. The analysis is moment-based, finite-memory, and restricted to product input laws.

In this paper, we propose leveraging rotatable antennas (RAs) to enhance near-field communication and sensing by exploiting a new orientation-domain spatial degree-of-freedom (DoF) provided by element-wise antenna rotation. Specifically, we investigate an RA-enabled near-field integrated sensing and communication (ISAC) system with sub-connected hybrid beamforming, where each transmit RA can independently adjust its boresight direction under a practical rotation constraint. A spherical-wave channel model incorporating orientation-dependent antenna gains is established to characterize multi-user communication and target sensing in the presence of clutters. Based on this model, a weighted communication-sensing utility maximization problem is formulated by jointly optimizing the receive beamformer, digital beamformer, analog beamformer, and RA boresight directions. To solve the resulting non-convex problem, an alternating optimization algorithm is developed by combining fractional programming, Riemannian optimization, and a spherical-cap Frank--Wolfe-based boresight update. To further understand the impact of RA rotation on near-field sensing, we derive a closed-form root Cramer--Rao bound (RCRB) expression. Simulation results demonstrate the convergence and effectiveness of the proposed algorithm. It is shown that the RA-enabled hybrid design can match or even outperform the fully-digital FPA benchmark in some regimes, indicating that the orientation-domain DoF introduced by element-wise rotation can compensate for limited RF chains. The RCRB and beampattern results further show that RA rotation improves off-broadside sensing accuracy, enhances range-domain focusing, and suppresses same-angle clutters in the near field.

Radio maps provide the essential foundation for low altitude networking systems. Unlike terrestrial radio maps that are typically generated via drive test measurements, mapping the air-ground environment requires the deployment of unmanned aerial vehicles (UAVs). This shift introduces two formidable challenges in uncharted 3D scenarios. First, sparse radio measurements and incomplete geometric observations hinder accurate reconstruction. Second, the large 3D action space and strict power constraints from high spectrum scanner energy consumption make informative exploration difficult. To address these issues, this paper proposes 3D uncertainty aware radio active mapping (3D-URAM), a closed loop active perception framework that decouples the mapping process into two offline trained stages. In Stage I, a Bayesian UNet is developed to recover radio maps from sparse measurements and partial geometry while providing calibrated predictive uncertainty. In Stage II, a dynamic probabilistic roadmap and a transformer based waypoint selection policy trained via proximal policy optimization maximize long horizon uncertainty reduction under travel budgets. Experimental results demonstrate that 3D-URAM reduces reconstruction error by over 50% compared to representative baselines. Real-world field tests within a 300mx200mx100m space also validate the potential of active radio map reconstruction.

This paper proposes a beamforming optimization scheme with joint antenna sub-array selection (SAS) and angular perturbation-based nulling (APN) for full-duplex (FD) massive multiple-input multiple-output (mMIMO) systems, to simultaneously suppress self-interference (SI) and multi-user interference (MUI). A comprehensive over-the-air SI channel measurement campaign, conducted with an 8x8Tx-8x8Rx FD array prototype, reveals significant variations across sub-arrays at different spatial locations, as well as reconfigurable characteristics of the SI channel under diverse Tx and Rx sub-array configurations. To exploit the selective SI channels, a particle swarm optimization (PSO)-based algorithm is developed to jointly determine optimal sub-array indices and perturbed steering angles, thereby effectively nullifying potential interference. Selecting sub-arrays with inherently lower SI channels notably enhances the beam-level isolation, while the added selection flexibility among comparable SI channels ensures more uniform SI suppression across diverse DL/UL locations and significantly improves worst-case isolation. Experimental evaluation based on the measured SI channel demonstrates that the proposed SAS technique achieves residual Tx-Rx beam-level SI suppression improvements of 29.2 dB and 26.6 dB for the sample 1x2 and 1x4 sub-arrays, respectively. A worst-case improvement greater than 30.7 dB is observed. Overall, the joint SAS and APN optimization scheme achieves average beam-level isolation of 85.2 dB and 83.3 dB with the 1x2 and 1x4 sub-arrays, respectively. With the application of a baseband precoder, all tested sub-array configurations achieve average MUI suppression better than -181.3 dB. These results confirm the potential of the proposed optimization algorithm to successfully reduce interference to the noise floor, thereby guaranteeing reliable FD mMIMO operation.

Affine frequency division multiplexing (AFDM) exhibits excellent Doppler robustness and the ability to characterize doubly selective channels. However, its signal dispersion characteristics make it challenging to directly adopt traditional time-frequency multiple access schemes. To address this issue, we introduce cooperative rate splitting multiple access (RSMA) for AFDM systems. The flexible configuration of AFDM chirp parameters can reduce the correlation between users' equivalent channels, which decreases the interference from RSMA private streams. We conduct a theoretical analysis of the cooperative RSMA-AFDM system and demonstrate that minimizing the overlap in the channel column spaces among users can effectively enhance the system performance. Guided by this analysis, we design a chirp parameter optimization scheme that reduces multi-user interference and maximizes diversity gain. To fully exploit the diversity gain brought by the proposed chirp parameter optimization, two expectation propagation (EP)-based distributed cooperative detection schemes are proposed. First, a decision-fusion-based method is developed, where local information and cooperative information are fused by maximum ratio combining, achieving a globally consistent estimate of the common stream. Second, we develop a belief-consensus EP-based detection scheme. In each iteration, user nodes exchange and fuse the first- and second-order statistics of the common stream, and the resulting beliefs gradually converge to a consistent global decision, which significantly improves the overall reliability.

Coordinated multi-satellite (CoMS) transmission and non-orthogonal multiple access (NOMA) are envisioned to jointly enhance coverage, capacity, and spectrum efficiency for satellite networks. Their integration into a unified CoMS-NOMA framework will allow more efficient, reliable, and energy-efficient multi-user access. This paper investigates the downlink performance of CoMS-NOMA networks from a system-level perspective, in which multiple satellites cooperatively serve multiple users via NOMA. Leveraging tools from stochastic geometry, related angles and distances in CoMS-NOMA are first derived as intermediate results. Then, we obtain the combined signal power distributions and analyze coverage and spectrum performance under both inter- and intra-satellite interference, accounting for potential imperfect successive interference cancellation (SIC). The analytical model is validated across a range of system parameters, including the number of satellites, service region angle, error-propagation factor, and power allocation coefficients. Numerical results indicate that increasing the number of cooperative satellites does not always improve coverage and spectrum efficiency. Additionally, while a higher main-lobe gain improves coverage, a near-perfect SIC provides only slightly greater benefits than a reasonably good SIC. With properly selected power allocation coefficients, CoMS-NOMA achieves up to a 270% improvement in coverage and a 56% gain in sum spectral efficiency, compared with conventional orthogonal and single-satellite schemes, indicating potential for green, energy-efficient satellite networking.

In this work, we investigate maximum-confidence discrimination, which encompasses minimum-error and unambiguous discrimination, for ensembles of separable states by considering global and separable measurements. We demonstrate that global measurements outperform separable ones, thereby establishing nonlocality without entanglement (NLWE) in terms of confidence in a detection event, a fine-grained state-identification strategy that maximizes the probability of a correct guess given a measurement outcome. Conversely, verifying achievable confidence in measurement outcomes can certify global measurements, namely, semi-device-independent certification of NLWE. Our results make it feasible to experimentally demonstrate NLWE using present-day quantum measurement devices, even with non-unit detection efficiencies, since maximum-confidence measurements rely only on detected measurement outcomes.

Quantum-enhanced displacement sensing with bosonic systems is typically formulated assuming that the oscillator is cooled close to its ground state before nonclassical probe preparation. We investigate whether such near-ground-state initialization is necessary, or whether sensitive probes can instead be generated directly from thermal states. We analyze hot quantum probes produced by squeezing, number-raising, and Schrödinger-cat-state generation applied to thermal inputs. We identify two distinct mechanisms by which thermal mixedness can remain compatible with enhanced displacement sensitivity. First, projecting a mixed probe onto a definite parity sector removes the usual thermal suppression of the displacement quantum Fisher information, which can then increase with initial thermal occupation. Second, coherent superpositions of opposite displacements can retain sensitivity through coherence between their displaced components, even when the underlying state is mixed. We use these two mechanisms to classify hot-state protocols according to whether their sensitivity comes from parity selection, coherence between displaced components, or both. Finally, we formulate an experimentally relevant optimization problem comparing initial cooling with direct hot-state preparation under realistic decoherence and show that complete cooling is not universally optimal. Our results establish hot-state engineering as a route to quantum-enhanced bosonic displacement sensing without mandatory ground-state initialization.

The detection of high-frequency gravitational waves (HFGWs) above 10 kHz provides a crucial probe of exotic astrophysical phenomena and new physics. We report the first search for HFGWs via their conversion to electromagnetic radiation through the inverse Gertsenshtein effect in Earth's magnetic field, utilizing radio telescopes including the Very Large Array (VLA) and the Atacama Large Millimeter/submillimeter Array (ALMA). Since no statistically significant signal is observed, we obtain new upper limits on the characteristic strain across the 1 GHz -- 1 THz band, with the most stringent constraint reaching $h_c \lesssim 10^{-18}$, improving upon existing bounds by up to three orders of magnitude. These results significantly advance the exploration of uncharted parameter space for exotic gravitational-wave sources, paving the way for future discoveries with next-generation facilities such as the Square Kilometre Array (SKA).

Projected quantum feature maps provide a strategy for using quantum processors as feature generators for classical machine-learning models. Building on counterdiabatic Ising-glass and one-dimensional Heisenberg PQFMs, we introduce a generalized two-qubit Hamiltonian-based PQFM that provides a unified way to encode classical features through local Pauli fields and pairwise two-qubit Pauli interactions. This construction allows distinct classical variables to be embedded along different Pauli axes of the same qubit, increasing the information density of shallow circuits while remaining compatible with hardware constraints. We develop and implement these methods in pqfmlib, a publicly available Python library for constructing, executing, and benchmarking Hamiltonian-based PQFMs.We then benchmark the generalized Hamiltonian PQFMs against reference PQFMs on four biomedical classification datasets under a nested cross-validation protocol with paired statistical tests. Quantum features are generated using both IBM quantum processors with up to 156 qubits and statevector simulations. Our results show that the generalized two-qubit Hamiltonian family provides the most consistent pattern of statistically supported gains over matched classical baselines, although the performance of all methods depends on the dataset, encoding strategy, measured observables, and hardware conditions. These findings support generalized Hamiltonian PQFMs as a promising route toward near-term quantum utility.

Full tomography of an unknown quantum evolution is resource-intensive and often unnecessary when the goal is only to predict selected properties. This motivates the study of classical shadow estimation of unitary channels (CSEU), a task in which one queries an unknown $d$-dimensional unitary $U$ and stores classical data that can later be used to predict expectation values $\mathrm{tr}[O \cdot UρU^\dagger]$ up to additive error $\varepsilon$ for arbitrary input states $ρ$ and observables $O$. We propose a parallel, non-adaptive CSEU protocol using $\mathcal{O}(d\varepsilon^{-1})$ queries when the input states or observables have constant rank. This achieves Heisenberg scaling with respect to $\varepsilon$ and is query-optimal, as we prove a matching $Ω(d\varepsilon^{-1})$ lower bound that remains valid even with stronger access to the unknown unitary. Our query-optimal CSEU protocol provides a versatile and powerful tool for quantum learning theory, pushing the performance limits of several fundamental learning tasks, including unitary channel tomography, Hamiltonian learning, boundary-regime quantum channel tomography, Pauli transfer matrix learning, inverse-free amplitude estimation, pure-state property estimation, and shallow-circuit learning. Remarkably, we show that optimal unitary channel tomography can be achieved using only parallel queries, closing the gap between the best achievable efficiency of parallel and sequential tomography protocols. Together, these applications establish our framework as a fundamental tool for learning properties of quantum processes, particularly for certain key tasks that require high precision.

We develop an affine-null characteristic formulation of the Einstein-Maxwell system coupled to a charged complex scalar field in $n$-dimensional spherical symmetry. By introducing suitable auxiliary variables, the main field equations are cast into a hierarchical system of radial hypersurface equations, supplemented by a transport equation for the scalar field. We discuss the associated characteristic initial-boundary value problem for asymptotic, vertex and null-boundary configurations, and derive the corresponding asymptotic quantities and balance laws. As a consistency check, we recover the scalar-free Reissner-Nordström-Tangherlini family, including both the non-extremal and extremal branches, directly from the hierarchy. The resulting framework provides a systematic setting for the study of charged scalar dynamics and exact black-hole solutions in higher-dimensional affine-null coordinates.

We investigate the construction of regular compact objects in the recently proposed lower-dimensional Einstein--Gauss--Bonnet (EGB) gravity obtained through regularized dimensional reduction. Unlike the standard BTZ black hole, the corresponding vacuum EGB solution develops a genuine curvature singularity at the origin, providing an interesting setting in which higher-curvature corrections deteriorate the ultraviolet behavior of spacetime. To address this issue, we reconstruct matter sectors capable of restoring regularity while preserving the BTZ-like asymptotic structure. First, we derive regular black-hole solutions supported by nonlinear electrodynamics and determine the corresponding electromagnetic Lagrangians directly from the field equations. We then extend the analysis to Simpson--Visser black-bounce geometries, obtaining smooth throat configurations with finite curvature invariants throughout the spacetime. As an alternative regularization mechanism, we formulate a unimodular extension of lower-dimensional EGB gravity and show that standard Maxwell fields can support regular geometries through a dynamical exchange between the vacuum and matter sectors mediated by a spacetime-dependent cosmological function. We further investigate the thermodynamic properties of the regular black-hole and black-bounce solutions, showing that the matter sector modifies the evaporation process, allows for remnant formation, and produces nontrivial phase transitions. In the black-bounce case, the thermodynamic quantities smoothly recover the EGB-BTZ behavior in the appropriate limit. These results demonstrate that lower-dimensional EGB gravity provides a useful laboratory for exploring the interplay between higher-curvature corrections, regular compact objects, nonlinear electrodynamics, and unimodular gravity.

We construct a graphical coaction for Friedmann-Robertson-Walker (FRW) integrals at all loop orders in conformally-coupled scalar theories with non-conformal polynomial interactions. Our construction makes use of intersection theory in the context of (partial/relative) twisted (co)homology, which we use to decompose FRW integrals (and their discontinuities and derivatives) into building blocks that can be represented as decorations of the original Feynman diagram. This facilitates a purely graphical description of the coaction, up to rational prefactors that can be read off from the graph. Our construction provides a comprehensive combinatorial framework for dissecting the analytic properties of cosmological observables; in particular, we demonstrate that the combinatorics of the differential equations that govern FRW integrals -- their so-called kinematic flow -- is a natural consequence of our coaction. We have also developed a user-friendly web application that computes the graphical coaction of any graph: https://frwcoaction.ca. Whenever possible, the web application also computes the differentials and discontinuities. A Mathematica notebook with the same functionality is also hosted at on a public GitHub repository.

Symmetries shape the quantum-information dynamics of many-body systems, but their effect on nonstabilizerness, the resource complementary to entanglement, is less understood. We compute the stabilizer Rényi entropy, a measure of nonstabilizerness, in $\mathrm{U}(1)$-symmetric one-dimensional random circuits. The disorder-averaged dynamics is captured by a four-replica tensor network, which we evaluate by $S_4$-adapted infinite time-evolving block decimation (iTEBD) directly in the thermodynamic limit. Together with a hydrodynamic argument, our results identify a diffusive universality class for the late-time approach of nonstabilizerness to its random-state value, with the stabilizer Rényi entropy gap closing as $1/t$. The same scaling is verified in an energy-conserving nonintegrable Ising chain. More broadly, our framework provides a hydrodynamic perspective on nonstabilizerness generation and offers insight into the design of approximate Haar-random states in Hamiltonian dynamics.

We develop an on-shell framework for thermal dissipation and radiation by macroscopic objects, whose large degeneracy of internal states is encoded in their entropy. In this framework, equilibrium asymptotic states are represented as on-shell particles, while non-equilibrium processes are described by on-shell transition amplitudes between them. A central observation is that spinning states remain essential even for macroscopically non-rotating objects. Consistency with macroscopic symmetries then implies spin universality, whereby all spinning states are governed by a single universal coupling. A key consequence is that absorption and emission probabilities are controlled by the same coupling, yielding local detailed balance directly from on-shell data. Applied to black holes, our framework reproduces the thermal emission spectrum and relates the Hawking temperature to the condition of maximal absorption consistent with unitary time evolution.

We formulate a local geometric criterion for weak cosmic censorship in black hole overcharging and overspinning thought experiments. Under the null convergence and generic conditions, matter injection turns a horizon cross section into a closed trapped surface. Any final spacetime unable to accommodate this surface is ruled out. This trapped surface criterion excludes superextremal Reissner-Nordström, Reissner-Nordström-de Sitter, and Kerr-Newman final states, as well as Weyl-class naked singularities. Our criterion does not rely on asymptotic charges or on an extremal condition characterizing naked singularities.

For general max-k-XORSAT with $k \geq 3$, no polynomial-time algorithm can do substantially better than random guessing on worst-case instances unless $\mathsf{P} = \mathsf{NP}$: approximating beyond the random-assignment value of $1/2$ is $\mathsf{NP}$-hard. The picture changes when each variable appears in at most $D$ constraints. In that bounded-degree setting, polynomial-time algorithms can provably beat the random baseline by an additive amount of order $1/\sqrt{D}$. For Boolean instances, this scaling is known to be optimal: the matching hardness result is due to Trevisan, while the corresponding algorithmic guarantee was established by Barak et al. Whether the same holds over general finite fields, and what it implies for quantum algorithms, has not been established. We make this connection explicit and extend the hardness to max-E$k$-LINSAT$(q,r)$ with bounded degree $D$ and over arbitrary finite fields $\mathbb{F}_q$, proving that it is $\mathsf{NP}$-hard to exceed $r/q + \mathcal{O}_{q,r}(1/\sqrt{D})$. These results provide the complexity-theoretic benchmark for the bounded-degree instances targeted by decoded quantum interferometry (DQI), QAOA, and classical heuristics. Any quantum advantage on bounded-degree instances is therefore confined to the constant prefactor. We further show that in the context of DQI and on $(k,D)$-regular instances, this prefactor is sensitive to the nature of the decoder: DQI with classical decoders faces an information-theoretic $1/\sqrt{D \log D}$ barrier that prevents it from matching the hardness scaling, while DQI with quantum decoders is compatible with the $1/\sqrt{D}$ scaling -- identifying quantum decoding as the key ingredient for matching the complexity-theoretic scaling with DQI.

Non-isometric encoding arises in various important contexts in quantum error correction, most notably in the finite-energy, non-ideal codewords inevitable in experimental realizations of continuous-variable codes, and holographic quantum gravity. In this work, we present a general and systematic theory of non-isometric quantum error-correcting codes. In particular, we employ the approximate quantum error correction framework to quantitatively study the fundamental limitations imposed by non-isometric encodings on the accuracy of quantum error correction and implementation of logical operations. We apply our theory to analyze GKP and tiger codes under energy constraints, and discuss the implications to holography.

Measurements can be used to monitor the evolution of quantum systems and may lead to a universally quantized time statistics. It is known that the mean return time is quantized for strong and indirect monitoring through the winding number of the return amplitude in a one-dimensional space. Here we discuss that under multi-channel strong or indirect monitoring, where the latter is achieved through ancilla coupling, the mean return time of a quantum walk in the projected subspace is also quantized. This reflects a universal time quantization for a higher dimensional evolution.

ZX calculus provides a graphical formalism for reasoning about quantum processes, built from two interacting Frobenius algebras associated with the Z and X bases of a qubit. While it has found widespread application in quantum information and computing, its relationship to quantum field theory has only recently begun to be explored. In this work, we further develop this connection by providing a generalization of ZX calculus to two-dimensional Yang Mills theory with a compact gauge group. The key observation is that both frameworks can be organized around the Hopf Frobenius algebraic structure associated with a group algebra, which can in turn be described by the diagrammatics of two dimensional topological quantum field theory. Given the well known relationship between gauge theory and gravity in two and three dimensions, our work paves the way for applications of ZX to low dimensional gravity.

We study the one-loop correction to the entropy of near-extremal black holes in three-dimensional massive gravity at the special point where the theory exhibits a unique maximally symmetric vacuum and non-constant curvature hairy black holes can achieve extremality even in the static case. Focusing on the near-horizon AdS$_2\times S^1$ geometry, we evaluate the contribution of boundary graviton modes that become exact zero modes in the extremal limit. We show that the resulting one-loop partition function generates logarithmic corrections to the semiclassical entropy, providing a new extension to higher-curvature gravity of what has been recently obtained for near-extremal black holes in General Relativity.

When matter fields are integrated out in a large N approximation, the conformal anomaly induces an effective action up to conformally invariant correction terms. In the present work, we consider the implications on black hole evaporation of such a term involving a conformal invariant found by Fefferman and Graham. Working in an approximation where the spacetime is static, we compute the induced stress tensor around a Schwarzschild black hole. The boundary conditions select an Unruh-like asymptotic sector, but not a unique stress tensor. A new one-parameter family of quantum hair emerges which changes the stress tensor in the near-horizon region. This suggests a new semiclassical mechanism by which information about black hole formation could be encoded outside the horizon.

In this paper, we analyze quantum information measures in fractional quantum mechanics using the Riemann-Liouville derivative formalism adopted here. In this case, we initially reconsider the conventional definitions of Shannon entropy and Fisher information, subsequently extending them to fractional quantum systems described by nonlocal differential operator frameworks adopted. Within this generalized formulation, fractional expressions of Shannon entropy and Fisher information are constructed and their mathematical structures examined thoroughly. Also, the formalism is then applied to the quantum harmonic oscillator, yielding explicit analytical expressions derived as functions of the fractional parameter therein. The obtained results demonstrate that fractional derivatives alter the localization properties of probability densities and generate nontrivial variations in information content and sensitivity across system behavior. In this context, the fractional parameter plays a central role in controlling deviations from the standard quantum information measures framework. Also, the study establishes a consistent framework for describing information-theoretic properties of quantum systems governed by nonlocal dynamics.

This paper develops an epistemological constraint on quantum gravity grounded in the empirical meaning of general relativity. The central claim is that a complete recovery of general relativity requires an effective metric, a continuum limit, or Einstein-like dynamics together with the physical conditions under which relational geometrical quantities can be objectively determined. These conditions concern the dynamical stability of measuring devices and reference systems, causal accessibility among physical systems, record formation, and invariance under admissible descriptions. In classical general relativity, they are usually implicit in the use of clocks, rods, light signals, freely falling bodies, detectors, and gauge-invariant observables. In quantum gravity, however, they become non-trivial because spacetime geometry may be emergent, effective, thermodynamic, relational, or frame-dependent. This claim is developed through four cases: Rindler horizons and the Unruh effect, black-hole thermodynamics and Jacobson's equation-of-state derivation, gravitational-wave detection, and Weyl and conformal gravity. The latter is discussed as a critical limiting case in which conformal invariance raises a sharp question about whether scale-dependent measurements of space and time can be physically fixed. Implications for quantum gravity are also discussed using emergent gravity and quantum reference frames as examples. The perspective developed in the study suggests a general epistemological constraint on quantum gravity: any viable approach must recover the physical possibility of objective geometrical measurement together with geometry itself.

We study the structure and transversal logical capabilities of stabilizer quantum error correcting codes. Among our results, we identify universal lower bounds on circuit depth to generate a full logical Clifford algebra, and develop novel constructions of logical transversal gates including a new depth-one transversal phase $\mathrm{\overline{S}}$ gate in the rotated surface code and a depth-one intra-block $\mathrm{\overline{CZ}}$ gate in the 2D-toric code that generalizes to all odd distances and all lengths $L\ge3$, respectively. Finally, we construct a high-rate non-LDPC CSS code family with parameters $[[n,\sqrt{n},Θ({n^β})]]$ where $β\approx 0.2823$ in one demonstrated case, that provably possesses a constant-depth complete 2-local transversal logical Clifford basis instruction set architecture (ISA) composed of all individually targeted $\mathrm{\overline{S}}$, $\mathrm{\overline{SHS}} = \sqrt{X}$, and $\mathrm{\overline{CZ}}$ gates. This ISA is depth-one for certain subfamilies that we design and generally constant-depth under certain conditions. The code family is built from a small code with parameters $[[n_0, 2, d_0]]$, and is tunable in the standard way: it tiles out to form utility-scale logical qubit counts, and it scales up through concatenation to achieve higher distances and error suppression. We show that this construction preserves the depth-one complete transversal logical Clifford basis ISA when composed with these commuting construction actions, inheriting structure from the core codes so that at scale the complete logical Clifford basis ISA remains depth-one up to depth-two addressable operations between tiled cores. We call these Quantum Logic Codes.

We propose a novel mechanism to trigger a first-order cosmological electroweak phase transition using non-Gaussian primordial fluctuations of a $\mathbb{Z}_2$-symmetric spectator scalar field. We show that the large fluctuations of the spectator field can modify the Higgs thermal mass and enhance the thermal barrier, thereby enabling a strong first-order phase transition. Non-Gaussianities in the primordial fluctuation spectrum significantly increase the probability of large-amplitude fluctuations, allowing a substantial fraction of the Universe to undergo the transition. The spectator field also naturally serves as a cold dark matter candidate through its coherent oscillations, reproducing the observed relic abundance. The resulting stochastic gravitational wave background peaks in the $10^{-3}$-$10^{-1}$ Hz band, making it detectable by future space-based interferometers.

We present a systematic thermodynamic analysis of nonsecular master equations. We consider master equations resulting either from the partial secular and the geometric-arithmetic approximations, two approximations ensuring the positivity of the system's dynamics when some of its transition frequencies are too small to enable the full secular approximation. Both cause the system to relax towards a steady state which is not the Gibbs state of its bare Hamiltonian. Nonetheless, we build a unified, consistent thermodynamic framework for those dynamics. Starting from a microscopic expression of the second law based on system-environment correlations, we employ a systematic perturbation theory to preserve the positivity of the second law despite the approximations done on the dynamics. We show that, in spite of the weak system-bath coupling, the system-bath interaction energy participates to the energy balance, as well as the Lamb-shift. Those extra contributions give rise to work performed by the system on the bath when the former is out of equilibrium. We compare this microscopic entropy production with the definition based on the contractivity of the reduced system dynamics (Spohn inequality). We show that, unlike for secular master equations, the two entropy production rates differ because of the presence of non-vanishing stationary coherences in the energy eigenbasis. However, in the case of a single thermal bath, the difference is purely transient, and no work can be cyclically extracted from the steady-state despite its non-Gibbs form. Finally, we illustrate our results with a simple example, clarifying and completing the thermodynamic picture of Markovian dynamics in the quantum regime.

Non-Gaussian evolution of high-order spin correlations characterizes important properties of quantum many-body systems. In practice, decoherence, statistical fluctuation and miscalibration of experimental parameters all hinder the witness of non-Gaussian dynamics. Here we demonstrate the crossover between Gaussian and non-Gaussian dynamics on a two-dimensional XY model with long-range and spatially structured interaction using a trapped ion quantum simulator. We prepare different initial densities of magnon excitations and verify the dynamics of single-spin observables for the engineered Hamiltonian. Then we compare the high-order spin correlations with the mean-field solution and the Holstein-Primakoff approximation, and demonstrate the non-Gaussian behavior in a way independent of the calibration errors. Our work provides a verifiable path from classically simulatable dynamics to regimes where quantum advantage may emerge.

Monitored quantum circuits provide a natural setting in which scrambling, measurements, and measurement-conditioned updates compete within a stochastic many-body dynamics. From the viewpoint of nonstabilizer resource theory, this competition is especially relevant because Clifford-compatible operations preserve the stabilizer structure, while weak non-Clifford perturbations inject magic resource. Most of the existing understanding of monitored quantum circuits has been shaped by numerical simulations and phenomenological descriptions, while a rigorous dynamics theory remains less developed. In this paper, we address this gap by developing an analytical framework which lays a rigorous mathematical foundation for the study of random monitored quantum dynamics. Specifically, we study a class of monitored quantum circuits driven by random Clifford. We prove the existence and uniqueness of the stationary law, which gives an ergodic description of the long-time dynamics. We then resolve the leading asymptotics of steady magic in the weak-magic-injection limit. This tangent description makes the contrast between resource measures transparent: in odd-prime local dimension, the steady Gross--Wigner mana has a linear leading asymptotic, whereas in qubit systems the steady 2-stabilizer Rényi entropy has a quadratic leading asymptotic. These different powers reflect the distinct local geometries of the two resource measures near the stabilizer layer. In this way, this work develops an analytical framework that first establishes the stationary ergodic dynamics of random monitored quantum circuits.

As quantum computing moves toward scientific computing applications, nonlinear differential equations remain a central challenge since quantum evolution is intrinsically linear. In this work, we introduce a reduced basis algorithm (RBA) for polynomial nonlinear ordinary differential equations (ODEs) and spatially discretized partial differential equations (PDEs). After time discretization, the method composes the resulting polynomial update map over $m$ timesteps, identifies the reduced monomial basis appearing in this composed map, and constructs a linear RBA operator whose action recovers the exact $m$-timestep nonlinear dynamics. Thus, at the level of the chosen discrete update rule, the method introduces no additional approximation error beyond the time discretization error. The qubit number requirement is governed by the size of the reduced monomial basis. For an $n$-dimensional polynomial ODE system of degree $p>1$, the lifted register requires at most $q_m^{\mathrm{ODE}} = O(nm\log p)$ qubits in the full basis scenario. For PDEs discretized on $N^D$ grid points, a locality-based construction requires at most $q_m^{\mathrm{PDE}} = O(D\log N + n m^{D+1}\log p)$ qubits. Hence, the dependence on the grid size remains logarithmic, while the nonlinear overhead is controlled by local reduced basis size. The main computational burden is moved from the quantum computer to a classical preprocessing step, where the reduced monomial basis and RBA operator are constructed for the chosen timestep window. Through numerical tests on the Lorenz system and the one-dimensional Burgers equation, we verify that the RBA reproduces the corresponding discrete time nonlinear dynamics exactly, while exposing the trade-off between timestep composition, reduced basis growth, and locality.

We present a general theoretical framework for multiphoton processes driven by quantum light fields, establishing a direct link between photon statistics and photoelectron observables. Our results show that the autocorrelation and cross-correlation functions, which quantify the underlying photon statistics, are directly mapped onto the resulting photoelectron spectra. Although our framework is broadly applicable, we demonstrate specifically in the example of reconstruction of attosecond beating by interference of two-photon transitions (RABBIT) the influence of the light statistical properties. In this approach, the amplitude, contrast and phase of the oscillations of the sideband signal as a function of pump-probe delay reveal the quantum nature of light. We analyze these observables across several quantum configurations, including correlated infrared and harmonic modes, as well as the uncorrelated case with non-classical harmonic statistics, thereby establishing a general framework for quantum-light RABBIT spectroscopy. We compare the analytical theory with numerical simulations for the case of classical harmonics and an infrared field in a squeezed coherent state, obtaining excellent agreement. Our results reveal how the interplay between classical and quantum correlations dictates the coherence of the photoemission process, providing a new window into the quantum-optical foundations of attosecond science.

We propose a scheme for realizing nonreciprocal transparency, Fano resonances, and slow/fast light in a hybrid cavity magnomechanical system containing two YIG spheres and a mechanical resonator. The nonreciprocal behavior originates from the magnon Kerr nonlinearity, which induces direction-dependent frequency shifts and modifies the interference pathways among cavity photons, magnons, and phonons. We show that the hybrid system supports multiple transparency windows arising from magnon- and magnomechanical-induced interference processes. The Kerr interaction strongly reshapes these transparency features, producing asymmetric Fano line shapes and enabling controllable nonreciprocal transmission. Furthermore, the associated dispersion exhibits pronounced directional asymmetry, leading to giant differences in the group delay for opposite propagation directions and allowing reversible switching between slow- and fast-light regimes. We investigate the roles of hybrid coupling strengths and dissipation channels and identify parameter regimes where the nonreciprocal response is maximized. These findings establish Kerr-engineered magnomechanical systems as promising platforms for integrated nonreciprocal microwave photonics and quantum information technologies.

Quantum coherence is understood to play a role in excitation energy transfer in open quantum systems, yet a quantitative approach to assessing its influence on the transfer process is still missing. Using Nakajima-Zwanzig projection operators, we derive a general memory kernel identity that enables us to characterize and quantify the impact of coherence in the eigenenergy basis on a generalized rate of energy transfer. Applying our approach to the electronic dynamics of a dimer coupled to a structured phonon bath, we demonstrate how quantum coherence acts to modulate energy transfer.

We study Yang-Mills theory coupled to a particle on a cylinder, where gauge invariance and compactness reduce the dynamics to a finite dimensional quantum system. In the Abelian case, this yields a model equivalent to the Landau problem on a torus, with a degenerate ground state structure. We generalize this construction to non-Abelian gauge groups and show that, for SU(N), the system reduces to a one dimensional quantum many body problem with a singular Calogero-Sutherland-type interaction.

Evaluating the trainability of adaptive quantum chemistry algorithms under multi-reference static correlation requires understanding how representation topologies intertwine with molecular geometry. We systematically expose a deep physical dependence on point-group symmetry by evaluating a spin-conserved SUSD operator pool across highly stretched configurations (2 x Re) of asymmetric LiH, symmetric BeH2, and asymmetric H2O. Under asymmetric distortions, the non-local mapping constraints of the Bravyi-Kitaev transformation create an optimization trapping effect--an encodement-locked manifestation of the broader barren plateau crisis. Crucially, by comparing these to the symmetrical stretching baseline of BeH2, we demonstrate that the preservation of point-group symmetry structurally protects the optimization landscape, proving that ansatz symmetry restrictions are necessary but insufficient without accounting for the underlying fermion-to-qubit representation. While current methods rely on numerical pruning to throttle pool sizes, our structural approach establishes that the mapping representation remains a critical factor in maintaining landscape trainability. Furthermore, exploiting structural overlap within our pool, we introduce a covariance-driven, adaptive shot-allocation filter. Diverging from static energy-variance minimization frameworks, our allocation engine operates as a dynamic runtime diagnostic tool. By continuously monitoring the gradient precision threshold epsilon, it aggressively prunes dead symmetry channels and triggers an automated circuit-termination sequence upon detecting representation-induced flat-lined states (dE/dtheta approx 0). This integration of algebraic measurement reuse with topology-aware statistical filtering provides a promising, resource-efficient strategy for executing deep variational algorithms on early fault-tolerant architectures.

The black hole information paradox originates from the thermal character of Hawking radiation, which appears to erase information about the collapsing matter. However, thermality constrains only observables defined at a single time and leaves the structure of temporal quantum correlations largely unexplored. Here we show that multi-time quantum-field correlations provide a concrete mechanism for the survival of pre-collapse information in black hole evaporation. Using a two-dimensional model of gravitational collapse and evaporation, we demonstrate that late-time multi-time correlations are not fully reproduced by the Unruh vacuum. In particular, they contain a contribution that depends explicitly on parameters characterizing the pre-collapse state, despite the thermal character of the asymptotic radiation. Our results identify measurable multi-time correlations as carriers of information in Hawking radiation and suggest that formulations of the black hole information paradox based solely on single-time observables are incomplete.

Quantum interconnects distribute entanglement via controlled light-matter interactions for quantum computing and sensing applications. Many entanglement generation schemes use coherent, reversible interactions that require precisely calibrated pulses to execute. In contrast, driven-dissipative protocols use a continuous-wave drive in the presence of correlated dissipation to stabilize entanglement in protected (dark) states. However, the same dissipation that generates the entanglement also limits its utility once the stabilization protocol ends. Here, we engineer a superconducting system of two giant artificial atoms coupled sequentially to a waveguide, with tunable individual and correlated dissipation enabled by interference between coupling points. Continuously driving the atoms through the waveguide exploits correlated dissipation to generate remote entanglement. We then tune the qubit frequencies in situ to suppress individual dissipation and thereby preserve the entanglement, achieving a Bell-state fidelity F = 0.89 +/- 0.02. This demonstration indicates that the driven dissipation of giant atoms is a viable approach for distributing entanglement across quantum networks.

Recent advances in multimessenger astronomy, particularly through gravitational-wave observations of compact-object mergers, have significantly improved our understanding of dense matter. Nevertheless, the internal composition of compact stars remains uncertain. Depending on the underlying equation of state (EoS), these objects may be neutron stars composed primarily of nucleons, quark stars made of deconfined quark matter, or hybrid stars containing both hadronic and quark phases. More exotic constituents, such as hyperons, meson condensates, or dark matter, have also been proposed. In this work, we investigate whether the internal composition of compact stars can be inferred from observable quantities, including mass, radius, and tidal deformability. To address this problem, we employ machine-learning and deep-learning techniques trained on a larg dataset of EoSs describing both neutron stars and quark stars. From these EoSs, we generate the corresponding mass radius relations spanning a wide range of stellar configurations. The resulting dataset is used to train and evaluate classification models aimed at identifying the nature of compact objects from their macroscopic properties. Our results indicate that suitable combinations of observables can distinguish neutron stars from quark stars with very high accuracy. These findings demonstrate the potential of machine-learning approaches as tools for probing the composition of dense matter. However, further studies incorporating additional scenarios, including hybrid stars and other exotic forms of matter, are required to establish the robustness and general applicability of this methodology.

We constrain the cosmological constant $Λ$ using astrometric and spectroscopic observations of the S2, S1, and S14 stars orbiting Sgr A$^*$. The stellar motion is modelled by numerically integrating timelike geodesics in Schwarzschild-de Sitter spacetime, including relativistic redshift and time-delay corrections. Orbital and spacetime parameters are inferred using a Bayesian MCMC analysis. The resulting posterior distributions place upper bounds on the magnitude of $Λ$ at the Galactic Center (GC). Combining the independent constraints from the S2, S1, and S14 orbits yields upper bounds of $Λ\lesssim 6.9\times10^{-48} \mathrm{m}^{-2}$ at 68\% credibility and $Λ\lesssim 1.0\times10^{-38} \mathrm{m}^{-2}$ at 95\% credibility.

Recently, a new class of deformed black-hole exact solutions was constructed in four-dimensional general relativity. The deformation is controlled by a parameter $B$, which survives after demagnetizing a black hole immersed in an external Bertotti-Robinson magnetic field and changes the global structure of the spacetime into a non-asymptotically flat spindle geometry. Although no astrophysical mechanism for generating such a deformation is currently known, it is natural to ask phenomenologically how large such a geometric deformation could be if it extended over the weak-field solar exterior. Using two classical Solar-System tests, we derive the leading corrections to planetary perihelion precession and to the light travel time in a Shapiro-type configuration. Requiring the \(B\)-induced perihelion advance to be smaller than the observational uncertainties in the supplementary perihelion precessions of planets gives the strongest bounds, \( |B|\lesssim 10^{-24}\text{--}10^{-23}\ {\rm cm}^{-1}\), while a Cassini time delay estimate gives a complementary null-geodesic sensitivity at the level \( |B|\lesssim 10^{-21}\ {\rm cm}^{-1}\). These results show that any such spindle deformation, if extended to the solar exterior geometry, must be extremely suppressed on Solar-System scales.

New interactions mediated by novel particles propose solutions to several important questions in modern physics. Axions serve as examples of such particles; they are lightweight and interact weakly with ordinary matter. This category of particles, including those similar to axions-termed Axion-Like Particles (ALPs)-arises from diverse theoretical frameworks, such as the Peccei-Quinn mechanism addressing the strong CP problem, string theory, and spontaneous supersymmetry breaking. Given their light mass and weak coupling, ALPs are also possible candidates for cold dark matter. Introducing these new interactions mediated by novel particles not only tackles several challenges in modern physics but also raises a crucial question: Are there undiscovered interactions beyond the Standard Model? Many of the interactions predicted by these theories are spin-dependent, which is the primary focus of this review. In this review, we first outline the theoretical foundations for investigating exotic spin-dependent interactions, highlighting their importance in various models beyond the Standard Model. We examine the potential roles of new lightweight particles in mediating these interactions, which may enhance our understanding of dark matter. Relevant formulas derived from theoretical models are included to support experimental investigations. Following this theoretical framework, we conduct a detailed review of recent experimental efforts to detect these exotic interactions. A systematic review of current constraints on these interactions is presented, along with an assessment of various detection approaches.

In quantum reservoir computing, a fixed quantum system transforms an input signal, while learning reduces to training a simple linear readout on its measured outputs. Since the quantum dynamics themselves are never optimized, the method is well suited to today's hardware. Yet these dynamics must still be chosen carefully, because their settings remain fixed throughout training and inference. It therefore remains an open question where, in its control space, a fixed quantum system learns well. We address this question for a dissipative reservoir by mapping performance over three central physical controls: the strength of the input drive, the coupling between neighboring qubits, and the rate of dissipation. Good performance concentrates in a single, well-defined operating region of this control space. This region transfers across tasks and reservoir initializations, and the same memory-defined regime persists under architectural changes. It is also mechanistically grounded, since it disappears whenever any of the mechanisms that create it is removed. Finally, the region can be located cheaply before any task is run, using a simple memory diagnostic.

In this study, we probe the cosmological evolution of the universe within the framework of modified teleparallel gravity by considering a power-law form of the function $f(T)=α(-T)^{n}$. To characterize the expansion dynamics, we employ logarithmic and log-periodic parametrizations of the deceleration parameter. These specific parametrizations provide a flexible and well-structured description of the cosmic expansion history across different cosmological epochs. The corresponding Hubble parameter is obtained as a function of redshift, facilitating a systematic investigation of the background dynamics. The model parameters are constrained using cosmic chronometer (CC) and joint (CC+Pantheon) datasets through a Bayesian analysis based on the $χ^{2}$-minimization approach. The evolution of key cosmological quantities, including the deceleration parameter, energy density, pressure, equation of state parameter and energy conditions, is examined in detail. The geometrical diagnostics indicate a clear departure from the standard cosmological constant behavior, pointing toward a dynamically evolving dark energy scenario. Further, the proposed models remain thermodynamically consistent and the estimated age of the universe is found to be compatible with observational constraints, thereby reinforcing the robustness and viability of the framework.

We investigate the extension of the Kubo--Martin--Schwinger (KMS) thermal equilibrium condition to non-Hermitian Hamiltonians with real spectra and biorthogonal eigensystems, providing a systematic analysis through three complementary routes. Our central result is a thermodynamic characterisation of quasi-Hermiticity: for $H \in M_d(\mathbb{C})$ diagonalisable with real spectrum, the biorthogonal Gibbs functional $ω_{\rm{bi}}(A) = Z_{\rm{bi}}^{-1} \sum_n e^{-βE_n}\langleφ_n|A|ψ_n\rangle$ satisfies $ω_{\rm{bi}}(A^†A) \geq 0$ for all $A$ if and only if $H$ is quasi-Hermitian. The proof constructs the metric $η$ directly from the eigenprojectors of $ω_{\rm{bi}}$ via the Riesz representation theorem, with no prior choice of $η$, providing a metric-free certificate of quasi-Hermiticity outside the Mostafazadeh--Scholtz framework. Under the full quasi-Hermitian hypothesis, we prove that the $η$-Gibbs state $ω_η(A) = Z_η^{-1}\, \rm{Tr}[ηe^{-βH}A]$ satisfies all three analytic KMS conditions, using the Hadamard three-line theorem and Bari's theorem on Riesz bases. The result is non-trivial: the transported state $\hatω(X) = \rm{Tr}[e^{-βh}Xη]/Z_η$ differs from the Gibbs state of the isospectral Hermitian partner $h = η^{1/2}Hη^{-1/2}$ whenever $[η,h]\neq 0$, so the KMS property cannot be deduced from the Hermitian theory by similarity. The gap between this result and the full Haag--Hugenholtz--Winnink $C^*$-algebraic framework is identified. Failure modes at exceptional points and for complex spectra are analysed, and the relation to the Fagnola--Umanità quantum detailed balance condition for open systems is discussed.

Selecting anomalous samples and explanatory features under fixed budgets defines a coupled constrained-optimization problem. Sequential feature-first selection ranks features before choosing samples, which can overlook features whose utility depends on which samples are selected, especially when scores are calibrated from reference data that may be limited, noisy, or drifting. We instead formulate the task as joint sample-feature selection under the same fixed counts. In the analyzed formal model, calibration-error sensitivity grows linearly with the number of samples for feature-first ordering but stays constant for joint selection. We introduce Coupling-Grouped XY-QAOA, a constraint-preserving grouped-angle variant for the resulting optimization problem. On matched sparse IBM Heron R3 benchmarks, a hardware-aware implementation reduces circuit depth by 45.9%-61.3% and two-qubit gates by 2.6%-5.2% relative to Qiskit optimization level 3 on the CZ-basis target. It enables, to our knowledge, the largest reported width-depth configurations for constraint-preserving bipartite-selection QAOA hardware executions with feasible-sector retention: 64 qubits at p=2 and 36 qubits at p=3. The 20-qubit p=5 runs retain 63% valid samples. Across 36-64 qubits, fixed-angle runs yield lower-energy feasible samples than matched random-feasible sampling. Warm starts reduce the gap to strict-feasible classical references by 57.5%-80.5%, and near-budget repair matches the sparse classical reference at 36 qubits. Benchmarks show gains in balanced fixed-budget regimes, and noiseless simulations show that problem-structured angle grouping improves over same-depth XY-QAOA and matched-parameter, type-preserving randomization controls. Overall, the results support calibrated joint selection and hardware-realizable constrained-mixer execution in the tested regimes.

In the present work, we study the absorption cross section of a Schwarzschild black hole for a massive vector field over arbitrary frequencies. Working in the Frolov-Krtouš-Kubizňák-Santos (FKKS) basis, we show how the conserved flux of the normalized Proca field naturally leads to the usual definitions of the absorption cross section in terms of the reflection and transmission coefficients. We then numerically compute these quantities over arbitrary frequencies. In contrast to the massless (photon) case, massive vector bosons exhibit new features arising from the field mass. In particular, the mass term introduces a longitudinal degree of freedom in addition to the transverse modes. Furthermore, it leads to a scalar-type branch in the even parity transverse modes, thereby breaking the usual degeneracy with the odd parity sector found in the massless case. We illustrate how these characteristics manifest themselves in the transmission and absorption spectrum. Given the recent developments in the study of ultralight bosonic fields in black hole spacetimes, the present analysis of the transmission properties of massive vector bosons bears particular significance.

Quantum Zeno dynamics (QZD) and dynamical decoupling (DD) are useful tools that enable the effective suppression of noise in quantum systems. We consider the problem of when (i) noise can be suppressed and (ii) Heisenberg limit (HL) can be achieved in quantum metrology, and prove necessary and sufficient conditions for when QZD and DD are useful for achieving these two goals. We also show that in the Markovian regime, there are scenarios where preventing errors using QZD/DD may enable HL to be achieved where current QEC methods may not. Finally, we demonstrate that the combination of both techniques can allow individually imperfect QZD and DD strategies to saturate HL.

We derive a class of vacuum regular black holes inspired by effective loop quantum gravity dynamics and extend the construction to asymptotically anti-de Sitter spacetimes. The derivation is based on a deparameterized Lemaître--Tolman--Bondi formulation, where an auxiliary dust field is introduced only to define an internal time and does not act as a matter source. In spherical symmetry, the dynamics reduces to a set of independent radial shells, giving rise to a factorized shell Hamiltonian and to a Birkhoff-type property: for a fixed reconstruction function and cosmological constant, the static geometry is uniquely determined by the mass. Within this framework, we construct several regular black hole models with de Sitter cores and corresponding models with anti-de Sitter cores. We then study their thermodynamics in the extended phase space, with particular emphasis on the Hawking--Page transition. For the class of models considered, the dominant transition is of Hawking--Page type, determined by the crossing of the black hole free energy with the corresponding thermal-AdS background. The regularization affects the quantitative transition temperature by deforming the physical outer-horizon branch, including its endpoint structure. In the large anti-de Sitter radius regime, the de Sitter core solutions exhibit a higher Hawking--Page temperature than their anti-de Sitter-core counterparts, while the ordering can be modified close to the lower admissible range of the AdS scale. Thus, the thermodynamic differences between the two classes are not a consequence of regularity alone, but arise from how the core deformation modifies the horizon branch relative to the thermal-AdS reference background.

Pretty Good Measurement (PGM) is a near-optimal strategy for quantum state discrimination, but its practical realization becomes unstable when the ensemble operator is singular or ill-conditioned. We introduce a numerically robust PGM formulation based on the Moore-Penrose pseudoinverse, replacing the standard inverse square root with a threshold-regularized variant that remains well-defined across different spectral regimes. We develop a hybrid classical-quantum framework that combines pseudoinverse-based spectral preprocessing with quantum circuit realizations using block-encoding and spectral-transformation techniques. The framework incorporates support awareness, yielding physically meaningful measurement operators even in rank-deficient cases, and employs oblivious amplitude amplification to improve circuit-level success probabilities. Extensive numerical and circuit-level simulations show close agreement between theoretical predictions and quantum circuit outputs. Experiments on synthetic and real datasets, including ill-conditioned and degenerate scenarios, demonstrate stable discrimination performance where standard PGM becomes numerically unstable. The results establish a practical hybrid classical-quantum framework for robust quantum state discrimination and extend previous circuit-based implementations of the PGM testing stage toward pseudoinverse-aware measurement design.

Designing compact ansätze in Variational Quantum Eigensolver (VQE) is crucial for solving energetic problems of practical molecules on near-term quantum devices. However, existing Adaptive Derivative-Assembled Pseudo-Trotter (ADAPT) ansätze face two challenges: improper operator selection and accumulation of degraded operators. In this paper, we propose the Hamiltonian-Aware (HA) ADAPT-VQE algorithm to address these issues. First, we establish a novel excitation operator selection criterion. It breaks the local constraint of existing criteria by incorporating Hamiltonian information, prioritizes physically meaningful excitation operators, and incurs no extra classical or quantum computational overhead. Furthermore, we develop a problem-adaptive method for discriminating and pruning redundant excitation operators stemming from improper selection and inevitable degradation. This method balances redundant operator pruning and convergence guarantee, and is applicable to ansätze with arbitrary scales. Systematic numerical experiments on typical strongly correlated molecular systems demonstrate that our HA-ADAPT-VQE avoids energy plateaus and outperforms baseline algorithms in terms of energy error, ansatz size, and measurement cost. This work offers an efficient, robust ansatz construction paradigm, facilitating the development and practical deployment of large-scale VQE in quantum chemistry.

Fast, high-fidelity two-qubit gates are a key requirement for fault-tolerant quantum computation. Tunable coupler architectures provide a flexible approach for implementing entangling gates through flux control with large on-off ratios, but fast flux modulation can induce diabatic transitions and population leakage to non-computational states, limiting gate performance. Here we present an analytical flux control method enabling derivative removal by adiabatic gate ($Φ$-DRAG) for suppressing leakage in flux tunable two-qubit gates. We show that $Φ$-DRAG differs fundamentally from conventional microwave implementations and derive modified flux modulation protocols that suppress leakage below $10^{-4}$ for fast entangling gates. The method remains effective across a range of asymmetry between qubit anharmonicities and different circuit parameters, enabling high-fidelity two-qubit gates within the fifteen nanosecond range.

We present a quantum algorithm for random number generation that achieves a provable quadratic speedup over classical Markov chain mixing, building on the Diaconis-Shahshahani Fourier analysis of the top-to-random card shuffle. The algorithm integrates three quantum primitives into a unified mixing circuit: the Quantum Fourier Transform (QFT), which diagonalizes the Markov transition operator; controlled phase rotations, which encode the shuffle eigenvalue spectrum; and the Grover diffusion operator, which acts as a quantum analogue of the Aldous-Diaconis strong uniform stopping time by reflecting amplitudes about their mean at each iteration. For an n-qubit register, the mixing time is O(\sqrt{n \log n}) iterations. Extending to m qudits of local dimension d reduces this to O(\sqrt{\log_d N}) iterations, where N = d^m, compared to the classical O(n \log n) bound. The qudit formulation further reduces QFT circuit depth from O(\log^2 N) to O(\log_d^2 N) gates per layer by encoding the same N-state space using m = \log_d N subsystems instead of \log_2 N qubits. We validate both variants on IBM superconducting hardware.

In this work, we study non-radial perturbations of AdS bubbles and their tidal Love numbers (TLNs). The odd- and even-parity TLNs are computed up to $l=6$ in the limit $k \to \infty$. The odd-parity TLNs are found to be negative, while the even-parity TLNs are positive for $\upsilon^2_s=-1$. As $l$ increases, the tidal Love numbers approach zero. The TLNs of the even-parity sector up to order $l=41$ are also calculated over the entire parameter space of $k$, from $0$ to $\infty$. We find that in the region where $p/σ>0$, an increasing number of TLNs become negative as $l$ increases. For $l = 41$, the highest order we have examined, the TLNs are negative everywhere except in a narrow region very close to the zero of $p/σ$, which agrees well with the instability criterion in the eikonal limit for self-gravitating membranes proposed by Yang {\it et al.}\ [P. R. L. {\bf 130}, 011402 (2023)].

Achieving high-fidelity operation in large-scale superconducting qubit systems requires not only control hardware with broad frequency coverage, low crosstalk, and tight synchronization but also software that coordinates system configuration, experiment execution, and data analysis. Here we present an integrated qubit-control system that combines broadband microwave hardware with a pulse-level software stack for scalable superconducting qubit experiments. The hardware provides broadband microwave coverage, including an instantaneous span of up to 1.6 GHz from a control output, while the software reduces setup and calibration overhead through automated configuration and built-in experiment workflows. We validate the system on a 64-qubit fixed-frequency transmon chip through full-chip frequency identification and representative demonstrations, including multi-unit far-detuned cross-resonance calibration and benchmarking that yields a measured two-qubit gate fidelity of 98.34%, and multilevel readout beyond the computational subspace. By disclosing the hardware architecture and releasing the software stack as open source, this work provides an inspectable hardware-software foundation for scalable superconducting qubit control experiments.

Ground-state preparation is a critical task in quantum simulation and quantum computing, as it enables the study of correlated phases and the generation of entangled resource states. While measurement--feedback control has emerged as a promising route to state preparation, existing schemes either rely on handcrafted, task-specific policies or are designed using full quantum-state information that is unavailable in real experiments and becomes impractical for large many-body systems. Here we develop an adaptive measurement--feedback protocol based on reinforcement learning under partial observability. The controller uses only the history of experimentally accessible measurement outcomes to choose both the measurement operator and the feedback action in real time. To make training compatible with experiments, we introduce a stochastic terminal reward built from one-shot measurements of randomly sampled Hamiltonian components, avoiding unphysical full-state reconstruction while remaining an unbiased estimator of the target energy. We demonstrate the method by preparing ground states of the Bose--Hubbard model and by generating GHZ states, establishing a scalable and hardware-compatible route to quantum state preparation.

In all known schemes for the non-Hermitian skin effect, the non-Hermitian ingredient that drives the skin localization, whether asymmetric hopping or gain and loss, is invariably introduced by hand as an independent model parameter along the skin direction. Here we show that when two spatial coordinates do not commute, the skin effect can break free of this paradigm: a gain-loss potential applied along one coordinate automatically generates non-reciprocity along the other through the coordinate noncommutativity, driving all eigenstates to pile up exponentially at a boundary. We term this phenomenon the noncommutative skin effect. The inverse skin length is proportional to the noncommutativity parameter and is given by an analytic formula, exact in the thermodynamic limit and verified by exact diagonalization of lattice models; the reflection symmetry of the imaginary potential furnishes an exact criterion for the presence or absence of the effect, valid rigorously for finite-size systems. For a sinusoidal imaginary potential, the skin direction of all eigenstates flips collectively at parameter points fixed purely by geometry. Because the flip point is independent of the potential strength, the reversal constitutes a zero-crossing measurement scheme intrinsically robust against systematic errors, from which the noncommutativity parameter can be extracted directly. The qualitative transition of the eigenstates from uniform to exponentially localized renders the effect a nonperturbative probe of spatial noncommutativity, and the Peierls-phase structure of its lattice model is in principle accessible to cold-atom synthetic dimensions, photonic resonators, and topolectrical circuits.

Gravitational wave (GW) observations provide an unprecedented laboratory for testing general relativity (GR) in the strong-field, highly dynamic, and relativistic regimes. Within the parameterized post-Newtonian (PN) formalisms, waveform generation tests have conventionally been limited to constraining inspiral coefficients up to the 3.5PN order. Leveraging the recent theoretical breakthrough that extended the analytical compact binary phasing to the 4.5PN order, we present the first observational constraints on these higher-order effects. Our analysis utilizes two exceptional events detected by the LIGO-Virgo-KAGRA (LVK) network: GW250114\_082203, which boasts the highest signal-to-noise ratio (SNR) recorded to date, and GW230627\_015337, which features a uniquely prolonged inspiral phase and the highest inspiral phase SNR to date. By performing Bayesian inference on the dimensionless deviation parameters ($δφ_i$) associated with the 4PN and 4.5PN coefficients, we find that our results are fully consistent with the predictions of GR. While the current 90\% credible intervals for the four deviation parameters are of order $\mathcal{O}(1) \text{-} \mathcal{O}(10)$, the general relativistic null values ($δ\hatφ_a= 0$) are entirely encapsulated within the bounds. This investigation establishes the first empirical baseline for 4PN and 4.5PN inspiral tests of GR, paving the way for high-precision null tests of GR with current and next-generation GW detectors.

In the Cartan teleparallel formulation of gravity, a scalar field coupled to the torsion tensor can acquire an environment dependent effective mass. This gives rise to a screening effect: Inside dense bodies such as the Sun, the scalar field is suppressed, causing the mass varying neutrino mass to vary with local density. We investigate the impact of this torsion-induced screening on solar neutrino oscillations. The mechanism modifies the standard MSW resonance condition in a distinct way, leading to detectable shift in the flavor conversion probability. Using data from solar neutrino experiments (Super-Kamiokande, Borexino, SNO), we place constraints on the torsion-scalar coupling parameters. Our work provides concrete test of teleparallel gravity through mass varying neutrino oscillations.

In this work, we show how the interaction between propagating modes of the quantized gravitational field and two massive particles trapped in a harmonic oscillator potential can cause the two particles to become entangled. To demonstrate this, we employ an operator-based approach within the framework of the Feynman-Vernon influence functional. Through this method, we find that the effect of the gravitational field on the generated entanglement is encoded in the commutation relations of the gravitational field. This result indicates that, within the framework of the model considered, entanglement arises through the quantum contributions of the gravitational field. Furthermore, this work also shows that entanglement is not formed instantaneously after the two particles interact with the gravitational field. Instead, there exists a time delay, proportional to the distance between the particles, before entanglement is established. This result reflects the causal propagation nature of gravitational interactions. In general, the entanglement generated through this mechanism is extremely small. Nevertheless, if the initial quantum states of the two massive particles are chosen to be squeezed states, the amount of generated entanglement can be enhanced, although the resulting effect remains very small.

We study a quantum Otto heat engine whose working substance is an anisotropic two-qubit Heisenberg XYZ model. Independent local magnetic fields are used to control each spin individually. The influence of the longitudinal coupling, anisotropy, transverse coupling, and local fields on the net work output and efficiency is systematically examined. Reducing the longitudinal coupling is found to markedly improve both the maximum work and the peak efficiency. The engine performance reaches an optimum at a particular value of the anisotropy parameter. A local work analysis clarifies how work is produced during the cycle. Because of the asymmetric local fields and the intrinsic spin-spin interaction, the two qubits play markedly different thermodynamic roles; the interaction term itself contributes crucially to the total work. We further analyze the variation of quantum entanglement, quantified by concurrence, along the cycle. The results indicate that a pronounced change in entanglement between the hot and cold isomagnetic strokes is closely correlated with the efficiency enhancement. This work offers new insight into the operating principles and control of quantum Otto heat engines.

The capacitated vehicle routing problem (CVRP) is an appealing candidate for quantum optimisation due to its combinatorial complexity and practical importance. However, the problem's constrained search space poses a challenge for such quantum algorithms. We introduce a quantum walk-based optimisation algorithm (QWOA) for the CVRP with homogeneous or heterogeneous vehicle fleets, addressing this challenge through a continuous-time quantum walk over a product space that coincides with combinatorial structures intrinsic to the CVRP solution space. Relative to the prior QWOA-based formulation, this approach reduces the per-layer gate complexity from $\mathcal{O}(n^{3}\log n)$ to $\mathcal{O}(n^{2}\log n)$ and supports a circuit parameterisation schedule generated by a fixed number of classical parameters. Exact state-vector simulation on instances with up to $n=8$ customers and $K=3$ vehicles demonstrates improved convergence to low-cost solutions using markedly fewer objective function evaluations, with the advantage broadening as problem size increases. These results identify structured product-space walks as a promising tool for optimisation over constrained combinatorial spaces.

We consider the deflection of light by a spherically symmetric and electrically charged black hole solution in the Einstein-Maxwell-Dilaton theory with specific values of the dilaton coupling constant where the deflection angle can be represented as elliptic integrals. We show that the deflection angle as a function of two dimensionless variables $(s,z)$, which are related to the background charge and the impact parameter, respectively, satisfies a system of 2nd-order linear partial differential equations called the Picard-Fuchs (PF) equations. For each case of the dilaton coupling, the PF equations lead to 1st-order ordinary differential equations with respect to the variable $s$ for the constants $\bar{a}$ and $\bar{b}$ in the log-formula for the strong deflection limit. Using the Hamiltonian system associated with Painlevé VI equations, which holds as a result of the integrability of the PF equations, we solve the equations for $\bar{a}$ and $\bar{b}$. By requiring consistency with the Schwarzschild case in the zero-charge limit, $\bar{a}$ and $\bar{b}$ for nonzero charge are uniquely determined.

We study the capability of future space-based gravitational-wave (GW) detectors to constrain supermassive black hole (SMBH) properties through observations of inspiraling stellar-mass binary black holes (BBHs) orbiting them. Focusing on stable hierarchical triple systems, we model the BBH motion in Kerr spacetime and compute the modulated GW signals using the post Newtonian waveform combined with moving-source transformation. Based on the LISA configuration and second-generation time delay interferometry technology, we estimate parameter uncertainties with the Fisher information matrix. Our results show that the outer semimajor axis has the strongest influence on parameter precision, while the SMBH spin and eccentricity mainly affect their own uncertainties. For high-SNR signals, the SMBH mass and orbital parameters can be measured with relative uncertainties on the order of $10^{-5}$, while the spin magnitude and its orientation can be constrained to within a few percentages. Applying the method to an M87*-like system, GW observations provide more precise measurements of the SMBH mass and spin compared with current electromagnetic observations, highlighting the potential of space-based GW astronomy to probe SMBH properties with high accuracy.

Quantum networks encounter unavoidable channel noises and erasure errors, presenting a huge obstacle in designing protocols that attain both high reliability and efficiency. Typically, quantum networks fall into two categories: those utilize quantum entanglements for quantum teleportation, and those directly transfer the actual quantum messages. In this paper, we present SurfNet, a quantum network that inherits the main advantages from both categories. It employs surface codes as logical qubits for encoding messages, and utilizes two parallel communication channels to fault-tolerantly transfer each surface code in a modular manner. Our approach of using surface codes can timely correct both operational and photon loss errors within the network, and the integration of the two channels within the network can greatly improve network throughput. For the implementation of SurfNet, we propose a novel network architecture, designed to better integrate surface codes into quantum networks. We also propose a novel error correction decoder, designed to fully utilize the modular characteristic of surface codes within our network. Simulation results demonstrate that SurfNet with its decoder significantly enhances the communication fidelity within quantum networks.

Primordial black holes (PBHs) in the asteroid-mass window ($10^{17}$-$10^{22}$ g) can account for all of the dark matter without violating any observational constraint, yet are routinely dismissed as fine-tuned. I put that dismissal to the test by applying three complementary fine-tuning measures uniformly across a broad landscape: three non-inflationary PBH production mechanisms, six classes of inflationary PBH models, and seven particle dark matter benchmarks, all evaluated against the same observable target. Three distinct naturalness universality classes emerge, determined entirely by the analytic structure of the abundance map rather than by the nature of the dark matter candidate. Biased-domain-wall PBHs are as natural as off-resonance weakly interacting massive particles and freeze-in particles; early-matter-domination and first-order phase transition PBH mechanisms occupy an intermediate tier alongside coannihilating WIMPs, unified by a structural identity in which the fine-tuning measure equals the logarithm of the ratio of the formation scale to the matter-radiation equality scale; and single-field ultra-slow-roll inflationary collapse is severely tuned for a distinct reason: a double exponential in which the power spectrum amplitude is itself exponentially sensitive to the inflaton potential coefficients, on top of the exponential collapse sensitivity of the abundance map. My main conclusion is that {\em the claim that PBH dark matter is generically fine-tuned conflates the worst case with a landscape spanning every naturalness tier}. The three-measure protocol also resolves a tension in the recent literature: the Barbieri-Giudice and Iovino-Riotto fine-tuning measures answer complementary questions and are reconciled within the two-layer decomposition developed here.

Quantum computation of fluid dynamics has attracted growing attention as a key application of fault-tolerant quantum computers anticipated in the coming decade, with lattice Boltzmann methods emerging as a particularly promising approach. Explicit and efficient elementary-gate-level circuit simulations, however, have so far been demonstrated only in the linear case. Here we include the leading nonlinearity through second-order Carleman linearization of the one-dimensional Boltzmann equation, and demonstrate, via explicit quantum-circuit simulation, the preparation of the final-time state using a Taylor-expansion-based ODE solver based on the quantum singular value transformation. With this construction, we analyze the gate and qubit complexities, which scale logarithmically with the grid size, the nonlinearity captured by the higher-order Carleman linearization, and the practical utility of higher-order expansions in the Taylor ODE solver. The construction provides a concrete baseline for computational cost reduction and further developments such as extensions to higher dimensions, complex geometries, and the extraction of physical quantities, towards industrially useful quantum CFD.

We introduce a general phase-space representation that includes global quantum symmetries in the basis expansion. This method, called matrix phase-space, projects the basis onto a reduced Hilbert space, which can greatly reduce sampling errors of many-body quantum simulations and unifies several previous phase-space methods. The purpose of this paper is to provide detailed proofs of basic theorems and operator identities. We also treat several different types of symmetries. To illustrate the benefits of matrix phase-space methods, we give a detailed derivation of a recent application to the topical problem of verifying the outputs of Gaussian boson sampling (GBS) quantum computers with photon number resolving detectors. This has exponential complexity, and using parity symmetry reduces sampling errors by very large factors relative to earlier methods.

We investigate the harvesting of quantum steering and its directional asymmetry between two Unruh-DeWitt detectors in a Lorentz-violating BTZ black hole spacetime. Since the detectors are located at different radial positions outside the black hole, they experience inequivalent local environments induced by gravitational redshift, causing Alice to undergo stronger effective thermal noise than Bob. Remarkably, we uncover a counterintuitive phenomenon in which the detector subjected to a higher effective temperature exhibits stronger steerability than the other one, revealing a nontrivial inversion of thermal intuition in curved spacetime. Furthermore, quantum steering survives only within a finite window of detector energy gaps and reaches its maximum within an optimal regime. We find that Lorentz violation suppresses steering most strongly near this optimal energy gap, indicating an enhanced sensitivity of maximal correlation extraction to symmetry breaking effects. Our results demonstrate that Lorentz violation acts as a geometric constraint on the quantum information capacity of spacetime, simultaneously restricting both the strength and the directionality of quantum correlations.

Quantum clocks can acquire relativistic phases from motional or gravitational proper-time differences, but reduced clock dephasing alone does not certify nonclassical proper-time histories. We formulate this distinction as a channel-certification problem. First, we show that any two-level single-time dephasing signal, including one generated by an effective quantum proper-time label, admits a classical random proper-time representation. We then define the convex set of classical mixtures of experimentally specified proper-time histories and prove a Choi-rank separation criterion for conditioned coherent history recombination. A two-branch Ramsey protocol gives explicit bright- and dark-port population witnesses outside this classical set. The certification is operational and relative to the specified history set: it rules out classical mixtures of the same implemented proper-time histories, not arbitrary classical protocols with different histories or controls.

Morphing circuits are a new paradigm for quantum error correction that relaxes hardware requirements. We present four constructions for CNOT-based CSS morphing circuits with explicit qubit connectivity degrees. All four constructions are specified in block algebra notation, with entries in algebras generated by permutation matrices. The first three are obtained by rewriting existing surface- and color-code morphing circuits; the fourth is a new three-round construction modeled on the 6.6.6 color code. The surface-code construction recovers the morphing circuit of Ref. [ST25] for two-block group algebra codes. Numerical search then instantiates these permutation matrices using regular representations of finite groups. [ST25] M. H. Shaw and B. M. Terhal, Phys. Rev. Lett. 134(9), 090602 (2025).

Quantum random access codes (QRACs) ask how well N classical bits can be encoded into M qubits while allowing any single bit to be recovered. Although the Nayak bound remains the standard general upper bound on the decoding probability, numerical evidence suggests a stronger upper bound in the small-qubit regime. In this work, we formulate the optimal decoding probability in terms of decoding measurements, reformulating QRAC design as a spectral problem for noncommuting measurements. Using this formulation, we give an elementary proof of the Nayak bound by simplifying the Chernoff-bound argument. Moreover, we refine the argument to obtain upper bounds that improve over Nayak's bound in the entire finite-size regime. The equality conditions of our bounds justify defining mutually unbiased projector-valued measurements (MUPVMs), a generalization of mutually unbiased bases. We show that decoding measurement of any two-qubit QRAC attaining the conjectured bound must form MUPVMs. We also show that any MUPVM, assisted by one ancillary qubit, yields a QRAC with optimal N-scaling decoding probability. Finally, we propose a new MUPVM-based construction for the (M+2,M)-QRAC family attaining the conjectured bound.

We investigate the dynamical and optical properties of a higher-dimensional loop-quantum-corrected black hole in a de Sitter background. The quasinormal modes of massless scalar perturbations are computed using time-domain evolution with Prony extraction, the matrix method, and the WKB approximation, showing good agreement among the three approaches. We find that loop quantum corrections induce moderate shifts in the quasinormal spectrum, whereas the spacetime dimensionality has a much stronger impact, leading to higher oscillation frequencies and damping rates. The negative imaginary parts of all modes indicate dynamical stability against massless scalar perturbations within the explored parameter range. We also analyze null geodesics and construct the corresponding black hole shadow. The shadow radius depends sensitively on the loop quantum parameter, the cosmological constant, and the number of spacetime dimensions, with extra dimensions generally reducing the shadow size. By comparing the theoretical shadow radius with the Event Horizon Telescope constraints for M87$^{\ast}$, we obtain bounds on the parameter space of the model. These results suggest that quasinormal modes and black hole shadow observables can provide complementary probes of loop quantum gravity effects and higher-dimensional spacetime structure in the strong-gravity regime.

Symmetry-protected topological phases have attracted significant interest at the fundamental level and as a potential platform for quantum information processing, owing to their protected edge states and resilience to perturbations. Applying these features for practical and efficient quantum computation is highly desirable, but remains an open challenge. Here, we demonstrate the partitioning into multiple independent Haldane phase subsystems of a single spin-1/2 ladder system and propose this as a scalable architecture for gate-based quantum computation, which takes advantage of the symmetry-protected topological order. We encode qubits in the two topological states of the $S^{z}=0$ sector of each subsystem. Finite-size effects, typically viewed as detrimental, instead provide a controllable energy splitting that enables single-qubit rotations using only local magnetic fields. An Ising-type interaction between neighboring subsystem edges generates entangling gates, enabling universal quantum computation driven by two control parameters that are easily accessible experimentally. Our results demonstrate how symmetry-protected topological phases can be directly harnessed for circuit-model quantum computation in realistic systems.

We demonstrate how supersymmetries of Hamiltonians for coupled Bose-Fermi systems can be used to place the Hamiltonians of the Jaynes-Cummings model and Dicke model under the rotating wave approximation in matrix form and provide explicit analytic solutions for their eigenvalues. We then use this supersymmetry to place the Liouvillians of the associated Markovian open systems in matrix form and provide explicit solutions for their eigenvalues. These results are a consequence of the fact that the Hamiltonian of the Jaynes-Cummings model commutes with the linear Casimir invariant of the superalgebra $u(1|1)$ and that the Hamiltonian of the Dicke model commutes both with the linear invariant of $\sum_{i} u_{i}(1|1)$ and with the invariant of an additional $su(2)$ algebra. Our methods apply to various coupled Bose-Fermi systems with $u(1|1)$ and more generally with $u(n|m)$ dynamical superalgebras, and may provide efficient tools for studying more complicated examples.

Pure two-qubit entanglement is usually characterized by scalar quantities such as concurrence. Here we show that it also has a natural geometric form. In the Pauli correlation tensor, maximally entangled states appear as improper orthogonal maps between two local Bloch spheres. These maps are roto-reflections. For partially entangled pure states, the same roto-reflection geometry is recovered after separating the contraction associated with concurrence. We call the corresponding geometric object the Entanglement Roto-Reflection Plane (ERRP). It organizes the maximally correlated directions of the two-qubit state and provides a covariant geometric complement to the scalar magnitude of entanglement.

We formulate stochastic inflation in an open quantum system framework. The field coarse-grained in a patch of fixed physical size, and the total momentum of that patch, form a canonical pair and act on a one-mode Fock space which we identify as the "bulk". At each time step, new comoving modes join the coarse-grained patch and the bulk has to be redefined. This redefinition produces an entangled mode that is traced over, yielding a non-unitary evolution equation for the bulk's density matrix. For a free test field in de Sitter, one obtains GKLS dynamics, generated by an effective Hamiltonian and a single non-Hermitian Lindblad operator, hence diffusion and Hubble friction originate from the same quantum channel. The Wigner-Weyl transform of the GKLS equation leads to a Fokker-Planck equation for the Wigner function, which matches the one that applies to the classical phase-space distribution of stochastic inflation. We also provide several schemes under which one can unravel the GKLS dynamics into stochastic Schrodinger equations when continuous measurements of the decoupled mode are performed, making contact with Langevin formulations of stochastic inflation. In the light-field regime, an additional overdamped reduction can be performed by integrating out the momentum variable in the Wigner distribution, leading to Starobinsky's slow-roll Fokker-Planck equation. In that regime, the purity of the patch is strongly suppressed. In contrast, for heavy fields, field diffusion is suppressed and the coarse-grained patch remains close to a pure underdamped oscillator, which prevents a classical stochastic treatment.

Classical electrodynamics in curved spacetime is formulated within a tetrad-based framework that preserves the direct physical interpretation of the electromagnetic fields measured by an observer. The formalism is applied to Schwarzschild spacetime for two distinct families of observers described in the same coordinate system: static observers and radially free-falling observers. For static observers, Maxwell equations retain their usual spherical structure, while the gravitational field introduces geometrical corrections in the radial and temporal sectors through the metric function. These corrections are interpreted in terms of proper radial distance, proper time, and the radial variation of the lapse function. For radially free-falling observers, additional kinematical contributions arise from the local radial boost relating the free-falling frame to the static one. As a consequence, charge density mixes with radial current, and the electric and magnetic sectors become intertwined in the temporal-radial projections of Maxwell equations, whereas the angular Ampère-Maxwell and Faraday equations retain the same structure found for the static observer. The weak-field and near-horizon regimes are also examined, and the results are discussed through an effective-medium analogy, in which the Schwarzschild geometry behaves as an inhomogeneous geometrical medium at rest for static observers and as a radially moving effective medium in the temporal-radial sector for free-falling observers.

Scalar fields provide perhaps the simplest playground in which to develop our understanding of quantum field theory. In this lecture, we consider what it means to write down a scalar quantum field theory and how we can give geometrical interpretations to the space of such theories: the theory space.

In the private simultaneous message (PSM) setting, $k$ players obtain inputs $x_i\in\{0,1\}^n$ and then each send messages to a referee, who should learn $f(x_1,...,x_k)$ but no other information about $(x_1,...,x_k)$. The PSM setting was introduced as a minimal model for secure multiparty computation and has connections to Boolean function complexity. In the quantum setting, PSM has been related to non-local quantum computation (NLQC). The communication and correlation cost of implementing PSM remains poorly understood. Here, we give new upper and lower bounds on the (quantum) PSM model. For lower bounds, we show: 1) Nečiporuk's measure lower bounds the entanglement required for $k$-player quantum PSM with perfect correctness. This leads to quadratic lower bounds for explicit functions. 2) The rank of the communication matrix of $f(x_1,x_2)$ lower bounds 2-player quantum PSM with perfect privacy but imperfect correctness. This implies a previously unknown lower bound on classical PSM with imperfect correctness. When allowing quantum communication and shared entanglement, these are the first lower bounds on quantum PSM that make use of the privacy condition. For upper bounds, we show: 1) Letting $s$ be the size of a quantum circuit computing $f$, $d_f$ be the circuit depth, $k$ the number of players, $n$ the number of bits received by each player, and $ε$ a correctness parameter, we obtain $\mathsf{PSM}_k^*(f) \leq (kn +s) \cdot \log^{O(d_f)}(s/ε)$. 2) The square of the Fourier 1 norm of $f$, $\Vert \hat{f}\Vert_1^2$, upper bounds the classical PSM complexity, $\mathsf{PSM}(f)\leq O(\Vert \hat{f} \Vert^2_1)$. In proving the first upper bound, we generalize existing $T$-depth based techniques for NLQC from $2$ to $k\geq 2$ parties, and consider cases where the Clifford layers are restricted to having small light cones.

We propose a generic, largely inflaton-potential-independent mechanism in which a light axion spectator, initialized near the hilltop of its potential, reshapes inflationary observables through purely gravitational multi-field dynamics. During inflation the axion is frozen and the background follows an effectively single-field trajectory. After inflation ends, the axion rolls, inducing a turn in field space and transient tachyonic phases of the isocurvature mode. The resulting ``tachyonic encore'' occurs entirely on super-horizon scales. These phases generate a nearly scale-invariant enhancement of the curvature power spectrum, suppressing the tensor-to-scalar ratio and shifting the scalar tilt to a weighted combination of adiabatic and entropic tilts at horizon crossing. We show that these effects can reconcile otherwise disfavored inflaton potentials with current CMB constraints. The same dynamics predict local non-Gaussianity, $f_{\rm NL}^{\rm loc.}\sim \mathcal{O}(1)$, within reach of upcoming surveys.

We develop a perturbative description of vacuum radiative processes in quantum electrodynamics with a prescribed external electromagnetic background capable of producing electron-positron pairs. Since the initial vacuum is then unstable and the in- and out-vacua are inequivalent, radiative observables require a real-time formulation beyond the ordinary in-out approach of vacuum-stable QED. Using the Keldysh-Schwinger-Fradkin nonequilibrium technique, we derive the mean number density of emitted photons through the second nonvanishing order in the fine-structure constant. The leading term, of order $α$, reproduces the known vertex and tadpole mechanisms, while the complete order-$α^2$ correction contains interference, loop, and induced-current contributions. We also give an independent derivation based on the spectral decomposition of the identity operator in the in-Fock space, where the photon number density is represented as a sum of squared transition amplitudes and vacuum-disconnected terms are canceled by the optical theorem generalized to an unstable vacuum. In addition, we compute the mean electromagnetic field through order $e^3$, including the electromagnetic dressing of the induced vacuum current, and verify it using the corresponding Schwinger-Dyson equations. The final formulas are expressed in terms of exact solutions and propagators of the Dirac equation in the external background and apply to general spacetime-dependent field configurations.

We develop a new grid-based multi-grid Poisson solver in numerical relativity. We report the performance of the multi-grid Poisson solver in the initial value problems for two-puncture black holes, a static spherical neutron star, a uniformly rotating neutron star in equilibrium, and a gravitationally collapsing massive star. As a demonstration, we conduct a numerical-relativity neutrino-radiation-transfer hydrodynamics simulation of the gravitational collapse of the $9M_\odot$ massive star in Ref.~\cite{Aguilera-Dena:2020mfh} up to the core bounce. During the simulation, we employ the constraint-preserving regird prescription with the newly developed multi-grid Poisson solver to improve the resolution. It shows that the baryonic mass, the Arnowit-Deser-Misner (ADM) mass, and the ADM-like angular momentum are, respectively, preserved with $O(10^{-3})\%$ and $O(10^{-2})$--$O(10^{-1})$\% accuracy.

Recent efforts to formulate quantum mechanics in a way that treats space and time on a more equal footing have led to a large variety of spacetime-oriented approaches. In this work we present a detailed study of spacetime states, the objects that play the role of quantum states in the recently introduced framework of spacetime quantum mechanics, and show that the main proposals in the literature are different manifestations of the same underlying object. Path integrals, quantum states over time, pseudo-density matrices, the Page and Wootters mechanism, superdensity operators, and timelike-entanglement proposals all arise from spacetime states through particular evaluations, reduced information, linear maps, or quantum channels. This unification provides explicit mathematical representations of these formalisms, reveals relations among them, and clarifies the spacetime information each one captures. We also study the broader relevance of the spacetime-state point of view for Leggett-Garg inequalities, OTOCs, temporal tensor networks, fermionic systems, relativistic QFTs, quantum reference frames, and classical physics, together with additional insights and perspectives revealed by the common unifying framework.

Given a quantum critical wavefunction in any dimension, we propose a reconstructed Hamiltonian, analogous to the ones previously found for 1+1d CFT and for 2+1d bosonic liquid topologically-ordered states. We test numerically that, for known regularized approximate CFT groundstates (on the icosahedron and the fuzzy sphere), (1) they are close to the groundstate of their reconstructed Hamiltonian, and (2) the spectrum of their reconstructed Hamiltonian on the unit sphere has CFT properties (integer spacing of descendants) and matches known low-lying energies. We show that this provides an automated method to improve the finite-size effects in a fixed Hilbert space.

We study the ability of the upcoming Laser Interferometer Space Antenna (LISA) to constrain gas torques acting on extreme-mass-ratio inspirals (EMRIs) when these are embedded in accretion disks, using recently developed relativistic models for the binary-disk interaction. Using a fully Bayesian setup, we find that, contrary to previous forecasts based on Newtonian results, these observations can provide simultaneous estimates of the disk surface density and the accretion rate (or, equivalently, its total luminosity) without the need for an electromagnetic counterpart. Our analysis also indicates that simpler measurement constraints based on the linear-signal (Fisher matrix) approximation are not valid for these systems. For typical EMRI observations, the torque amplitude can be constrained to within ~10%, strengthening the prospect of probing accretion physics at (sub)microparsec scales, deep in the strong-field gravity regime and complementing electromagnetic observations. This also strengthens LISA's ability to help answering questions such as how massive black holes grow and coevolve with their host galaxies and, by helping to identify the EMRI's host galaxy through cross-correlation with AGN catalogues, to improve the use of these sources as (dark) sirens for cosmology.

We study perturbations of Yang-Mills plasma, represented by scalar quasinormal modes of AdS black branes, as functions of the wave number $q$ in the entire range from zero to infinity. At finite $q$, these modes can be computed by classical spectral methods based on truncating the boundary value problem. We show that this truncation admits a natural analytic interpretation in terms of quantum Seiberg--Witten periods in the Nekrasov--Shatashvili limit, with the spectral condition organised as an instanton expansion around small values of the counting parameter. The physical black-brane problem corresponds to evaluating this series at a finite value of the counting parameter, and the Seiberg--Witten formulation provides a systematic way to analyse when the truncation is under control. In particular, it reveals that, as $q$ or the mode number $N$ increases, the physical point approaches the boundary of the domain of convergence of the instanton expansion, limiting the validity of the truncation approach. We overcome this limitation through an exact WKB analysis in which $q^{-1}$ acts as the expansion parameter. The resulting exact quantisation conditions, expressed in terms of period integrals and the associated Stokes geometry, incorporate both perturbative and non-perturbative corrections. The resummed quasinormal modes remain accurate far beyond the strict large-$q$ regime and can be analytically continued all the way to $q=0$, notably by using the Seiberg--Witten approach, providing a consistent description of the QNM spectrum.

To answer the above question, we study low-frequency scattering in the four-dimensional traversable wormhole of Maldacena, Milekhin, and Popov. The resulting transmission probabilities reveal that wormhole traversability depends strongly on the nature of the probe. For scalar probes, both neutral and charged, traversability depends on the time scale. On time scales of order the light-crossing time after sending in a signal, the transmission is parametrically suppressed, with most of the incoming signal reflected or temporarily trapped inside the wormhole throat. As time progresses, the trapped signal gradually leaks out, so that at late times the accumulated transmission cross-section approaches one half of the corresponding black hole absorption cross-section. Despite this generic suppression at low frequencies, the transmission spectrum also exhibits resonant frequencies at which transmission becomes perfect. Charged massless fermions tell a very different story. Unlike scalars, they traverse the wormhole with essentially unit probability at low energies. The same mechanism underlies their efficient absorption by magnetic black holes and realizes a channel closely analogous to the Callan-Rubakov effect, revealing unexpected connections with monopole-fermion scattering. Putting everything together, we conclude that scalar probes are best suited for uncovering distinct features of these magnetic wormholes, while charged massless fermions are the ideal carriers of information through them.

Polarized images of a black hole encode the direction of electromagnetic energy flow near its event horizon. Measuring polarization from near-horizon emission can determine whether this energy flow is powered by the accreting plasma or the black hole spin. Here we consider the linear polarization of synchrotron radiation emitted from the base of horizon-threading field lines in a time-stationary, axisymmetric, and degenerate Kerr magnetosphere. We show that the observed polarization pattern displays universal behavior: it is completely determined by the black hole spin and observer inclination and is independent of the magnetic field geometry. We derive a simple analytic formula for this spin-dependent horizon polarization pattern. We find that this predicted pattern is also approached in time-averaged images from General Relativistic Magnetohydrodynamic simulations. Future observations with Very-Long-Baseline Interferometry at microarcsecond resolution could detect the trend of polarization toward the unique horizon value in M87*. Such observations may enable new measurements of black hole spin and provide evidence that magnetic field lines thread the horizon and extract spin energy via the Blandford--Znajek process.

We construct the phase space for an arbitrary cut of a null hypersurface in Minkowski spacetime and demonstrate that it is symplectomorphic to the infrared phase space of asymptotically flat gravity. Fluctuations of the cut are mapped to the leading soft graviton mode. Furthermore, the supertranslation Goldstone mode is mapped to the product of the size of the cut with its symplectic partner, the null time offset.

The asymmetric three-qubit W-class state $|\overline{W_3^L}\rangle$ defines an isosceles entanglement-network geometry, (a) two vertex-base (VB) links form stronger bipartite connections, (b) while the base-base (BB) link is weaker. This suggests that concentrating entanglement into a super-link may be advantageous for quantum-network tasks. Here, we show that this intuition is incomplete. We analytically compare the bipartite concurrence dynamics of the symmetric |W> state and the asymmetric $|\overline{W_3^L}\rangle$ state, which differ both in entanglement-network geometry and excitation sector under standard noise models. In the absence of noise, the concurrence hierarchy is $C_{VB} > C_W > C_{BB}$. Under phase damping, this hierarchy is preserved for all noise strengths and no entanglement sudden death occurs. Under amplitude damping, however, the hierarchy is reordered. The symmetric |W> state becomes the most robust, while the base-base concurrence of $|\overline{W_3^L}\rangle$ vanishes at the finite threshold of parameter $γ$. We term this reordering as the \textit{Super-Link Fragility Effect}. The same structural asymmetry that produces a stronger vertex-base link also makes it more vulnerable to energy dissipation when coupled with multi-excitation amplitudes. Under depolarization, the asymmetry advantage is erased, with $C_W$ and $C_{VB}$ sharing the same sudden-death threshold for some value of the parameter p, while $C_{BB}$ disappears earlier at some other value of the parameter p. The generalized amplitude damping channel continuously connects the damping-dominated regime to the pure-excitation limit, where the initial hierarchy is restored. These results show that entanglement robustness in $W$-class resources is controlled not by initial concurrence alone, but by the joint structure of entanglement-network geometry, excitation sector, and noise symmetry.

Quantum gates are usually described through matrix and tensor-product formalisms that often obscure their geometric structure. In this work, we formulate the Pauli and Clifford groups within the complex Geometric Algebra (GA) framework. We show that the Pauli group is naturally identified with the group of blades up to a global phase, thereby providing a geometric interpretation of Pauli operators and their commutation relations in terms of oriented subspaces. We further prove that Clifford operators are generated by products of π/4-Pauli rotors and introduce a greedy Pauli rotor decomposition algorithm whose empirical behavior suggests unexpectedly compact decompositions for Clifford operators. Finally, we show that Clifford+T universality admits a natural geometric interpretation through π/8-rotors within this framework.

In this paper, we investigate the sparsity of black hole radiation, the Bekenstein entropy loss, and the total particle emission of a Schwarzschild black hole (SBH) within the framework of non-commutative (NC) gauge theory of gravity. First, we provide a brief review of black hole (BH) thermodynamics, computing both the deformed Hawking temperature and entropy. In this geometry, the divergent behavior of the temperature is removed, and a logarithmic correction to the entropy emerges. We then present the NC corrections to the Bekenstein entropy loss of the SBH alongside the total number of emitted particles. Our results show that the total number of emitted particles is proportional to the entropy behavior of the NC SBH, which is consistent with non-thermal radiation. Finally, we analyze the sparsity of Hawking radiation in this geometry, finding that the NC SBH exhibits extremely sparse radiation ($\hatη \gg 1$), which diverges at the final stage of evaporation as the black hole ceases to radiate.

We developed the third-order hydrodynamic equations using relativistic extended thermodynamics of gases with 14 independent fields. The resulting fluid equations are based on the relativity principle, the entropy principle, and the requirement of hyperbolic, and hence finite, propagation of disturbances, which is automatically incorporated. The expressions of entropy, four-current, shear-stress tensor, dynamic pressure, and heat flux are expanded up to third order (cubic). We explicitly present the newly calculated coefficients in the equilibrium properties of an ultra-relativistic gas regime and the non-degenerate relativistic gas. Contrary to the general cases, the non-degenerate regime eliminates fugacity from the coefficients, allowing for the easy normalization of these coefficients, and the ultra-relativistic regime provides us with the upper bounds of these coefficients. We found good agreement on some of the coefficients as compared to calculations from earlier models, specifically in kinetic theory, and other coefficients had slightly different values to those obtained in kinetic theory.

We use the exactly-solvable t Hooft model of 1+1D large-N_c QCD as a rigorous laboratory for the breakdown of analyticity of a causal response function, the meson two-point function. A PT-symmetric deformation i gamma(x-1/2) of the light-cone meson operator, the analogue of an imaginary chemical potential, drives the lowest two mesons to an exceptional point (EP) at gamma_c. Recasting the resolvent as a Jacobi continued fraction yields gamma_c in closed form: 2 pi g^2 N_c at the two-pole level, converging to 7.966 g^2 N_c by depth five -- an analytic, not numerical, threshold. The square-root exponent nu=1/2 is fixed by the 2x2 Jordan form and confirmed by finite-size scaling to N=1999. The breakdown has an unambiguous time-domain signature: the propagator norm is bounded for gamma < gamma_c, grows linearly at gamma_c (the Jordan secular law), and exponentially beyond -- observable, since the deformed operator is a non-Hermitian Wannier-Stark ladder, in photonic and topolectrical analogues. The threshold is locked to confinement, gamma_c propto g^2 N_c, and recurs as a uniform EP cascade; a second, non-reciprocal deformation yields an exactly-exponential non-Hermitian skin effect. This is the first analytically-controlled instance of exceptional-point analyticity breakdown in a confining gauge theory.

Arbitrary initial conditions allow solutions of Einstein's field equations for General Relativity to have arbitrarily large relative rotation of matter and inertial frames. The ``Rotation Problem'' is to explain why the measured relative rotation rate is so small. Nearly any reasonable theory of quantum gravity can solve the rotation problem by phase interference. Even as early as about a quarter of a second after the initial singularity, quantum cosmology would limit the cosmologies that contribute significantly to a path integral calculation to have relative rms rotation rates less than about $10^{-51}$ rad/year. Those calculations are based on using 50 e-foldings during inflation. For 55 or 60 e-foldings, the cosmologies contributing significantly to the path integral would have even smaller relative rotation rates. In addition, although inflation dominates the calculation, even if there had been no inflation, the cosmologies contributing significantly to the path integral would have relative rotation rates less than about $10^{-32}$ rad/year at about a quarter of a second after the initial singularity. These calculations are insensitive to the details of the theory of quantum gravity because the main factor depends only on the size of the visible universe, the Planck time, the free-space speed of light, the Hubble parameter, and the number of e-foldings during inflation. These calculations use the Einstein-Hilbert action in quantum gravity, including large-scale relative rotation of inertial frames and the matter distribution, in which each ``path'' is a cosmology with a different rms relative rotation rate. The calculation shows that the action is an extremum at zero rms relative rotation rate.

The canonical Weyl form for four-dimensional static axisymmetric vacuum metrics is obtained by identifying the area function of the two Killing orbits with a harmonic coordinate on the twodimensional orbit space. This construction is valid in Ricci-flat vacuum, but it is no longer available in Einstein spaces with nonzero cosmological constant. In this paper, we consider the generalized orthogonally transitive static axisymmetric line element and derive the reduced Einstein- $Λ$ field equations. We show that the canonical Weyl choice $W=ρ$ is locally admissible if and only if $Λ=0$. The Kottler metric gives the simplest explicit example of the resulting equation for the area function. Thus, the statement that "Weyl metrics do not allow $Λ\neq 0$ " is precise only when the metric is assumed to be in canonical Weyl coordinates. The issue is not staticity or axisymmetry, but rather the fact that the area function is no longer harmonic.

We investigate how a negative cosmological constant affects the Gauss-Bonnet (GB) scalarization in the Einstein-Maxwell-scalar-Gauss-Bonnet theory with a scalar coupling constant $η$ to GB term. We focus on the instability of Reissner-Nordström-AdS (RN-AdS) black holes under a scalar perturbation governed by an effective mass $μ^2_{\text{eff}}$ sourced by the GB term. Unlike the asymptotically flat spacetime case, the onset of scalarization is not merely determined by $μ^2_{\text{eff}} < 0$, but it is constrained by the Breitenlohner-Freedman (BF) bound. In case that the BF bound is violated ($η>2.25$ with $Λ=-0.5$), one may find AdS-tachyoinic instability. We find that for $0<η<2.25$, the GB$^+$ scalarization may be performed through spontaneous scalarization, while for $η<0$ the GB$^-$ scalarization is found to give the single branch of scalarized AdS black holes. For the GB$^+$ scalarization in $η_{th}\leη<2.25$ with $η_{th}$ threshold instability, we obtain the single branch ($n=0$ fundamental branch) of scalarized AdS black holes, contradicting to infinite branches. Finally, it is observed from Gibbs free energy that a phase transition from scalarized AdS black holes to RNAdS black hole occur natually and it is confirmed to be second-order.

The one-loop five-point function of stress tensor multiplets in N=4 super Yang-Mills theory in four dimensions has previously been studied by integrability methods. The finite-size viz gluing corrections defining it were originally matched on the known field theory result - establishing many features of the formalism on the way - and later to a good extent also analytically evaluated. For the first time, we present a full analytic evaluation of all processes. The starting point is the re-summation of series of residues into Euler integrals. These are directly integrated where possible or otherwise recovered from intersection theory for generalised hypergeometric functions.

We calculate, for the first time, heavy quark transport and thermal dilepton production from QGP using viscous correction up to second order in gradients. We use the form of viscous correction obtained from Chapman-Enskog like expansion of the Boltzmann transport equation in relaxation time approximation, and compare our results with that of Grad's 14-moment approximation. By employing the temperature and shear stress evolution profiles of QGP obtained from second-order causal relativistic viscous hydrodynamics, we study the heavy quark transport coefficients and thermal dilepton production from an evolving QGP. In the case of HQ transport, the CE corrections suppress the drag force substantially, induce a non-trivial momentum dependence in transverse momentum diffusion, and result in a comparatively less modification in longitudinal momentum diffusion. Whereas, for thermal dileptons, the CE corrections result in an enhanced early-time contribution which decreases and become converging to the first-order CE correction with the evolution of QGP, and remain well behaved compared to that of Grad's correction. Our results indicate that the modification of the observable due to viscous corrections is governed by the magnitude of the corrections as well as the interplay between their momentum dependence and momentum weighting of the transport and emission kernels. We demonstrate that the momentum structure of the various viscous corrections at the level of distribution function is not directly translated to the observables since the different observables are sensitive to distinct regions of momentum space.

We report a measurement of the charged-current coherent pion production cross section on argon using the MicroBooNE liquid argon time projection chamber exposed to the Booster Neutrino Beam at Fermilab. The measurement uses the MicroBooNE data set corresponding to $1.26 \times 10^{21}$ protons on target with a mean neutrino energy of $0.8$~GeV. The flux-averaged cross section is measured to be $(9.1 \pm 1.2_{\text{stat}} \pm 1.2_\text{syst}) \times 10^{-40}\,\text{cm}^2/\text{Ar}$. This result represents the first measurement of charged-current coherent pion production on argon at sub-GeV neutrino energies. Due to its clean two-body kinematics, where the neutrino interacts coherently with the entire nucleus producing a forward muon and pion with no nuclear breakup, this process provides a useful tool for constraining neutrino flux uncertainties in current and future oscillation experiments such as DUNE.

We study the cosmological dynamics of the thermal scalar the winding string mode that becomes massless at the Hagedorn transition by coupling it to the string-frame gravi dilaton effective action. This provides a field-theoretic framework for investigating winding mode dynamics near the Hagedorn temperature and their role in string cosmology. Below the Hagedorn temperature, the phase space contains static configurations in which the thermal scalar balances the shifted dilaton evolution. These configurations are boundary states rather than attractors, and when the thermal scalar mass depends on the scale factor, winding mode back reaction opposes expansion and can reverse it. Above the Hagedorn temperature, the tachyonic thermal scalar generates negative effective energy density while preserving the null energy condition, enabling branch changes of the Brustein Veneziano type. These transitions do not provide the graceful exit required to connect the Hagedorn phase to standard cosmological evolution. At the Hagedorn temperature itself, the quadratic effective theory breaks down and higher order interactions become essential. Because the thermal scalar originates as a Euclidean order parameter, our Lorentzian treatment should be viewed as an effective dynamical model of the Hagedorn transition. Within this framework, the thermal scalar clarifies the dynamical structure surrounding the Hagedorn transition and shows how the Hagedorn exit problem cannot be resolved within the quadratic effective theory alone.

We study supercooled first-order phase transitions in a supersymmetric hidden sector with a spontaneously broken $U(1)_X$, focusing on the frequency range of the Einstein Telescope and Cosmic Explorer. Along the D-flat direction the tree-level quartic vanishes, so the barrier is generated radiatively by soft SUSY-breaking splittings. In the $\overline{\rm DR}$ scheme the gaugino mass $M_{\tildeλ}$ sets the barrier depth, while the soft scalar mass $m_0$ stabilizes the broken vacuum. For $M_{\tildeλ}/v_X\simeq0.05$--$0.23$, the predicted signal reaches $Ω_{\rm GW}h^2\sim3\times10^{-10}$ near the percolation boundary. The observable amplitude depends sensitively on the portal coupling $δ$ through the hidden-to-visible temperature ratio at percolation: for a cold initial hidden sector the signal rises from the ET floor at $δ=10^{-6}$ to $Ω_{\rm GW}h^2\simeq7\times10^{-11}$ as the sectors approach thermal contact at $δ=10^{-4}$, while a hotter initial hidden sector gives a large signal already for weak portal coupling. We follow this evolution with an 11-variable Boltzmann system that separates the cold nucleating exterior from the reheated true-vacuum interior; reheating mainly enters through the energy budget and redshift factors. The same hidden sector can reproduce $Ω_{\rm CDM}h^2=0.12$ through relativistic dark-quark freeze-out followed by entropy dilution from hidden-Higgs decay, with $m_q\simeq30$--$800\;$keV and $N_{\rm eff}\lesssim{\rm few}\times10^{-5}$.

Sub-eV axion dark matter behaves as a coherent classical field that can induce macroscopic current in quantum materials. We explore the possibility of axion detection via the chiral magnetic effect in zero-bias Weyl semimetals under a static external magnetic field. We demonstrate that for a $1 \, {\rm cm^2}$ sample in a realistic $10 \, {\rm T}$ magnetic field, the signal remains in the observable femto-ampere range. Utilizing state-of-the-art SQUID-based current readout, the setup can probe axion-electron couplings below existing stellar cooling bounds across a broad range of axion masses.

We study the interplay between the chiral density wave (CDW) and the single-plane-wave Larkin-Ovchinnikov-Fulde-Ferrell (LOFF) phase of color-superconducting matter in two-flavor quark matter at vanishing and non-vanishing temperature $T$, quark number chemical potential $μ$ and isospin chemical potential $δμ$. The analysis is performed within the two-flavor Nambu--Jona-Lasinio (NJL) model in the chiral limit, using a three-momentum cutoff scheme. Treating the CDW wave vector $\vec{q}$ and the LOFF pair momentum $\vec{q}\,'$ as independent variational parameters, we minimize the mean-field effective potential with respect to both amplitudes and both wave vectors, without constraining their relative orientation, and map out the $T$-$μ$ and $μ$-$δμ$ phase diagrams for a range of diquark couplings $G_D$. Our central result is that $\vec{q}$ and $\vec{q}\,'$ are never simultaneously nonzero: inhomogeneous chiral and diquark condensates do not coexist across the entire parameter range.

Particle physics collider experiments provide Rivet routines as part of the analysis preservation strategy for model-independent measurements. Rivet is a C++ toolkit that allow new theoretical models to be compared to the measurements, thus aiding the development and tuning of Monte Carlo event generators as well as searches for physics beyond the Standard Model. However, analysis coverage is known to be incomplete, with only 39% of measurements having documented and publicly available Rivet routines. In this article, we design and implement an automated workflow based on Large Language Models with the goal of providing the missing routines. This multi-step workflow, referred to as AgentRivet, extracts the physics analysis information from published papers and writes the missing Rivet routines, with intermediate code- and physics- reviews as part of an autonomous quality control. We report the results obtained using commercial Large Language Models, provided by OpenAI, Anthropic, and Google, for two recent measurements from the ATLAS and CMS experiments. We find that AgentRivet produces competent Rivet routines with few syntax errors. The physics fidelity of the routines is reasonable and follows the explanations given in the relevant publications. Nevertheless, physics-implementation issues do arise and are investigated using the artefacts produced by AgentRivet. The majority of physics implementation issues arise from subtle-but-ambiguous definitions in the given publication, although some models struggle to implement complex observables even when clear definitions are given.

Following our earlier work on studying 3-dimensional nuclear recoil response to Galactic Weakly Interacting Massive Particles (WIMPs) in directional direct Dark Matter detectors, in this paper, we simulate 3-D coherent elastic neutrino-nucleus scattering (CEvNS) events induced by Solar B-8 neutrinos. Our numerical results show that, in contrast to the approximately fixed patterns of the WIMP-induced signals, the characteristic ring-like angular distributions of the nuclear recoil flux/energy of CEvNS events show clearly annual variations along the trajectories of the moving direction of incident Solar neutrinos in different celestial coordinate systems without experimentally distinguishable target dependence.

Yang-Mills theories at $θ$ and $θ+2π$ are unitarily equivalent, but their $2π$ periodicity has a nontrivial realization. Recent developments in generalized global symmetries show that confinement vacua at $θ=0$ and $2π$ should belong to different symmetry-protected topological (SPT) states with the $1$-form center symmetry. For its examination, we measure the Wilson-'t Hooft loop operators at $θ=2π$ for the $SU(2)$ Wilson lattice gauge action and discuss their long-distance behaviors. This requires us to identify the gauge topological charge in the presence of defects, and we employ the $1$-form covariant DBW2 gradient flow to smear lattice gauge fields. We then obtain numerical evidence consistent with dyon condensation at $θ=2π$, rather than monopole condensation, as theoretically predicted.

In QCD sum rules for light-quark systems, infrared (IR) divergences can appear in the Wilson coefficients of certain condensates. These divergences manifest explicitly in the coordinate-space expressions of the light-quark propagators. We propose an improved method to eliminate these IR divergences at the propagator level and present a subtraction formula that implements this procedure. Compared to the existing methods that rely on the mixing between quark and gluon condensates of the same dimension to eliminate IR divergences, this method is more intuitive and easier to apply in practical QCD sum rule calculations.

We consider an extension of the Standard Model by an additional Higgs doublet and a Majorana neutrino, which we call the Grimus-Neufeld Model (GNM). For certain parameter choices the GNM can be compared to the Inert Doublet Model (IDM), which has a scalar dark matter candidate. This motivates that the scalars of the GNM could possibly contribute to dark matter. To check this, we present the tree-level two-body decays of the heavy scalars of the GNM and compute the lifetime of the pseudoscalar in the IDM limit.

We study the impact of inelastic gluon-radiation processes on the transport coefficients of the quark-gluon plasma within the dynamical quasiparticle model (DQPM) in the temperature-baryon-chemical-potential plane $(T,μ_B)$. Extending the elastic baseline established in previous DQPM calculations, we include radiative $2\to3$ scattering channels with massive partons and effective DQPM propagators and vertices. The corresponding momentum-dependent interaction rates and relaxation times are used within the relaxation-time approximation to calculate the shear viscosity, bulk viscosity, electric conductivity, and baryon diffusion coefficient as functions of temperature $T$ and baryon chemical potential $μ_B$. We find that radiative channels systematically reduce all considered transport coefficients relative to the elastic-only results, in accordance with the decrease of the relaxation times. In the thermal regime explored here, however, this reduction remains moderate, since the inelastic rates stay below the elastic ones over the considered $(T,μ_B)$ range. The radiative channels become more relevant mainly for partonic scatterings at large momenta, which are thermally suppressed in the strongly interacting QGP. At $μ_B=0$, the resulting $η/s$, $ζ/s$, and $σ_Q/T$ are compatible with available lattice-QCD estimates within uncertainties. At finite $μ_B$, our results provide predictions for the transport properties of QCD matter relevant for beam-energy-scan programs.

We consider different scenarios which lead to a damping of the neutrino oscillation probability and investigate their impact on the first JUNO results. First, we study decoherence effects due to wave packet separation. In addition, we consider an open quantum system framework and adopt a phenomenological approach which allows us to parameterize the energy dependence of the decoherence effects more freely. Finally, we study the effect of invisible neutrino decay. In all cases with the first data JUNO can already place competitive bounds on the parameter space of the scenarios under consideration, while also maintaining a robust measurement of the standard neutrino oscillation parameters.

A reliable description of direct CP asymmetries in hadronic $B$ decays, among the most sensitive probes of the CKM mechanism and of physics beyond the Standard Model, remains unresolved. Conventional factorization approaches adopt distinct treatments for the strong phases originating from long-distance QCD interactions, leading to predictions that differ by factors of several for identical decay channels. Existing final-state interaction (FSI) models are limited to one-loop approximations that break unitarity and rely on channel-specific phenomenological cutoffs, severely restricting the predictive capability. We introduce a unitarized FSI framework based on the Lippmann-Schwinger equation solved to all orders, restoring unitarity that is manifestly broken in one-loop treatments. Using the coupled $D\bar{D}$ system, we show that the interaction kernel can be derived from chiral effective field theory constrained by heavy-quark spin symmetry, and low-energy constants are fixed by the $γγ\to D\bar{D}$ cross-section data from BaBar, leaving no adjustable parameter in the FSI sector when applied to $B$ decays. The resulting predictions for CP asymmetries and branching fractions show excellent agreement with the available experimental data. A defining consequence is that CP asymmetries in the pure-annihilation channels $\bar{B}^0\to D^0\bar{D}^0$ and $\bar{B}^0\to D_s^+D_s^-$, identically zero in any short-distance treatment, are dramatically enhanced by FSI, and a measured nonzero asymmetry in these modes is therefore a direct experimental signature of long-distance dynamics. We present definite predictions for the partial widths and CP asymmetries of all yet-unmeasured channels, establishing final-state interactions as a predictive, data-driven ingredient of the Standard Model with direct implications for CKM parameter extraction and BSM searches.

Geoneutrinos, electron antineutrinos produced by radioactive decays of heat-producing elements (HPEs) within the Earth, provide unique insights into Earth's interior and heat budget since their first detection in 2005 by KamLAND. Conventional geoneutrino detectors currently provide integrated global information and lack the capability to spatially resolve structures deep within the Earth. Here, we evaluate the ability of angular-sensitive geoneutrino detectors to distinguish between homogeneous and heterogeneous mantle models, focusing on Large Low Shear Velocity Provinces (LLSVPs). Our results show that LLSVPs enriched in Th and U yield a distinct flux of geoneutrinos with distinctive angular patterns. An oceanic site above the Pacific LLSVP is considered a particularly favorable detector location. The Ocean Bottom Detector (OBD) project aims to leverage this spatial resolving advantage by deploying a kiloton-scale liquid scintillator detector directly on the ocean floor, enabling unprecedented sensitivity for mantle geoneutrino detection. These findings demonstrate the critical role of combining geophysical and geochemical data to guide detector site selection, ultimately improving constraints on Earth's internal heat and the HPE distribution.

Hadronic tau decays provide a unique low-energy laboratory for the charged-current interaction between quarks and leptons. Their use as an electroweak precision sector is, however, limited by nonperturbative QCD dynamics in the resonance region, encoded in hadronic form factors. In this work we show that angular information in two-hadron tau decays can be used to construct observables in which these form factors cancel, leading to first-principles Standard Model predictions up to light-quark-mass and radiative corrections. We derive the fully differential distributions for arbitrary two-pseudoscalar final states, including tau polarization, and extend them to the Weak Effective Field Theory at linear order in new-physics couplings. We then illustrate our strategy with several benchmark observables, for both unpolarized and polarized taus, whose Standard Model predictions can be obtained without modelling the dominant hadronic form factors and which receive characteristic beyond-the-Standard-Model corrections, in particular from tensor currents. These results provide concrete targets for Belle II, polarized-tau proposals such as Chiral Belle, and future high-statistics facilities, including super-tau-charm factories and FCC-ee, to test the electroweak structure of hadronic tau decays.

We study a renormalizable SU(5) grand unified theory (GUT) supplemented by a 45-dimensional Higgs field and a DFSZ axion sector, imposing a Georgi--Jarlskog flavor structure at the unification scale. We perform a one-loop gauge coupling unification analysis, explicitly including the threshold masses of the light multiplets arising from the 45-dimensional Higgs field. This analysis identifies the viable region in the $(M_{\mathrm{GUT}},M_{S_1})$ parameter space. Through the relation between the unification and PQ scales, this region yields correlated predictions for the QCD axion mass. The Georgi--Jarlskog assumption substantially reduces the flavor ambiguity of the dimension-six baryon-violating operators, enabling robust constraints and predictions not only for antineutrino modes but also for charged-lepton proton decay modes such as $p \to e^+ π^0$. We present the combined implications for proton decay and axion searches, showing how the GUT-selected parameter region maps onto the axion mass, the axion-photon coupling, and the axion-induced EDM coupling.

We analyze the flow of the background graviton propagator driven by the running of asymptotically safe form factors in four-dimensional quantum gravity at quadratic order in the curvature expansion. We construct the propagator in the $k=0$ limit and investigate its momentum dependence. The non-local form factors are then analytically continued to Minkowski spacetime, from which the corresponding Minkowskian propagator is obtained. We find a single pole at $q^2=0$ with positive residue and no additional ghost poles at this level of approximation. The corresponding Newtonian potential is found to be regular at $r=0$.

Motivated by the observation of numerous charmed-strange hadrons lying close to the $K^{(*)}\bar D^{(*)}$ and $K^{(*)}D^{(*)}$ thresholds, we systematically investigate their mass spectra and electromagnetic characteristics, explicitly incorporating the $S$-$D$ wave mixing and coupled-channel effects. For the mass spectra, we employ the one-boson-exchange model and obtain several promising $K^{(*)}\bar D^{(*)}$ and $K^{(*)}D^{(*)}$ molecular candidates awaiting future experimental confirmation. The coupled-channel dynamics is found to play an essential role: it not only enhances binding in existing channels but also generates new loosely bound states that are absent in the single-channel approximation. Regarding the electromagnetic characteristics, we discuss the M1 radiative decay widths and magnetic moments of these charmed-strange molecular tetraquarks based on our obtained mass spectra and spatial wave functions within the constituent quark model. Our results demonstrate that the electromagnetic observables serve as valuable discriminants for clarifying the nature of these hadronic molecules. We encourage experimental collaborations to focus on such predicted charmed-strange molecular tetraquark candidates, especially the genuinely exotic $K^{(*)}\bar D^{(*)}$ states comprising four different flavors with valence quark content $\bar c \bar s u d$.

We introduce an inclusive generalised-$k_t$ jet algorithm for Deep Inelastic Scattering, defined in the Breit frame and implemented in fjcontrib. The family of algorithms is governed by the usual parameter $p$, which controls the transverse-momentum dependence of the algorithm, as well as by a jet radius parameter $R$. The angular-ordered ($p=0$) version of the algorithm was already presented by some of us, and can be used to formulate observables with simple all-order structures. In this article we investigate phenomenological applications of the algorithms related to the identification of the jet associated with the struck quark, and assess their sensitivity to non-perturbative effects, such as hadronisation. We also perform comparisons with the recent Centauro algorithm.

We develop a systematic theoretical framework to improve theoretical predictions for nuclear $β$ decays of arbitrary angular momentum $J$, leading to a model-independent nuclear-structure electrostatic correction to the Coulomb interaction between the emitted lepton and the nuclear charge distribution, useful for ongoing and future precision searches for physics beyond the Standard Model. The formalism is based on nuclear matrix elements expanded in multipole operators, as commonly used in \emph{ab initio} calculations. First-order Coulomb corrections are derived from one-photon exchange preserving the full multipole and angular structure of the decay rate, and are subsequently expanded in the relevant small parameters of the nuclear problem, suppressing the leading nuclear structure correction to a few per-mills for medium mass nuclei with natural beta decay properties. We show that within this formalism, the leading Coulomb correction originates from three modifications of the original weak-only interaction: a modification of the nuclear charge form factor, which yields a correction similar to the known Fermi function, a shift of the momentum transfer within the lepton traces, and the same shift but inside the nuclear multipole operators. We additionally provide explicit results for allowed Gamow--Teller and unique first-forbidden transitions.

The Forward Physics Facility (FPF) at the High-Luminosity LHC (HL-LHC) will enable high-statistics measurements of TeV-scale neutrinos, but the intense flux of forward muons poses a major challenge for neutrino detectors in the far-forward region. We investigate the suppression of background muons using sweeper magnets with a simulation framework combining SIBYLL event generation, BDSIM beam transport, Geant4 particle tracking, and realistic magnetic field maps. Starting from a muon flux of $3.8\times10^3~\mathrm{cm^{-2}}$ per $\mathrm{fb^{-1}}$ without magnets, a magnet in the LHC tunnel alone achieves the target level of $2\times10^3~\mathrm{cm^{-2}}$ per $\mathrm{fb^{-1}}$. Additional magnets at the TI18 tunnel and FPF entrance further reduce the flux to $1.5\times10^3~\mathrm{cm^{-2}}$ per $\mathrm{fb^{-1}}$ in the optimized configuration. These results demonstrate that a properly optimized multi-stage sweeper magnet system can significantly reduce the forward muon background, while also highlighting the importance of realistic transport simulations and geometrical constraints in achieving further suppression.

We have analyzed one-nucleon matrix elements of current-current correlators; the currents consist of pseudoscalar octet, isovector and isoscalar currents, axial anomaly, and axial isovector and isoscalar currents. Using QCD sum rules, the coupling constants of nucleon with each of these currents have been expressed in terms of nucleon matrix elements of quark, gluon and quark-gluon composite operators and moments of parton distribution function. On the phenomenological side, contribution from the non-diagonal matrix elements of operators between nucleon and its excited states or continuum states have also been accounted for. For the pseudoscalar coupling constants of the nucleon two expressions have been obtained in which one of them consists of only moments of parton distribution function but yielding approximately same numerical result as the other one. Of particular interest is the nucleon matrix element of axial anomaly which has been largely ignored in the current literature.

Extending the double copy of scattering amplitudes to more general physical quantities involving local gauge-invariant operators is a central open question. While progress has been made in the double copy of form factors (FFs) with a single-operator insertion, it has led to two intriguing features: poles that are spurious from the FF viewpoint become physical propagators in gravity, and FFs obey hidden factorization relations on these poles. This picture is difficult to generalize to FFs with multiple operator insertions due to even more complicated spurious pole structures. We resolve this problem by introducing a "dyeing" procedure that promotes color-singlet operators to adjoint colored states, while the original FF is recovered by the inverse "bleaching" (U(1) decoupling) operation. In this new picture, the spurious poles are propagators of the dyed state, and the hidden factorization relations follow from BCJ relations of the dyed amplitudes. For multiple operator insertions, this framework uncovers a new scalar-ordering structure that survives the double copy to gravity.

A $10^6\,M_\odot$ object has recently been inferred from gravitational imaging of the strong-lensing system JVAS B1938+666, exhibiting an unusually dense inner region embedded within an extended envelope, far exceeding expectations for cold dark matter (CDM) halos. Using gravothermal fluid simulations, we show that such a structure arises naturally in self-interacting dark matter (SIDM) halos evolving into a deep core-collapse phase, where a secondary dense central core forms within an extended profile. The resulting density structure closely matches the inferred properties of the lensing object. We also demonstrate that a similar profile could be reproduced in CDM in the presence of an intermediate-mass black hole, but this requires an early-forming progenitor that subsequently loses $5$ orders of magnitude in mass through tidal stripping by the lens galaxy. Whether such a scenario can be realized in realistic cosmological environments remains an open question.

We report the first observation of the decays $B^{+} \to Σ_{c}(2455)^{++} \barΞ_{c}^{\prime-}$ and $B^{0} \to Σ_{c}(2455)^{0} \barΞ_{c}^{\prime0}$, with significances of $6.4\,σ$ and $5.3\, σ$, respectively, including systematic uncertainties. This analysis is based on data samples containing $771.6 \times 10^{6}$ $Υ(4S)$ decays collected with the Belle detector at the KEKB collider and $520.6 \times 10^{6}$ $Υ(4S)$ decays collected with the Belle~II detector at the SuperKEKB collider. The branching fractions are measured to be $\mathcal{B}(B^+ \to Σ_c(2455)^{++} \barΞ_c^{\prime -}) = (1.68 \pm 0.31 \pm 0.12^{+1.49}_{-0.54}) \times 10^{-3}$ and $\mathcal{B}(B^0 \to Σ_c(2455)^{0} \barΞ_c^{\prime 0}) = (1.28 \pm 0.32 \pm 0.10^{+0.30}_{-0.21}) \times 10^{-3}$, where the first and second uncertainties are statistical and systematic, respectively, and the third arises from the uncertainties in the absolute branching fractions of $\barΞ_{c}^{-}$ and $\barΞ_{c}^{0}$ decays. This result represents the first observation of $B$-meson decays into a pair of charmed baryon-antibaryon states belonging to the same $SU(3)$ flavor sextet.

Grand unified theories (GUTs) predict the overproduction of magnetic monopoles, leading to the so-called monopole problem, which is often addressed by cosmological inflation that dilutes their abundance. However, if inflation occurs before the GUT symmetry breaking, monopoles are produced afterwards and the problem persists. This motivates the exploration of alternative mechanisms. We propose a simple solution based on the Langacker--Pi mechanism within an $SU(5)$ GUT framework with symmetry breaking into its special subgroup. In particular, after the gauge symmetry is broken to the Standard Model (SM) gauge group $SU(3)_C\times SU(2)_L\times U(1)_Y$ by the vacuum expectation value of an adjoint scalar, it is further broken to $SO(3)_C\times SO(2)_L$. This structure is naturally realized by introducing a symmetric tensor scalar, a singlet scalar, and multiple singlet fermions. During this intermediate phase, the monopoles become connected to antimonopoles by cosmic strings, which enhances their pair annihilation and reduces their abundance. Subsequently, the symmetry is restored to the SM gauge group. The restoration transition can be either first-order or second-order, depending on the model parameters. In the case of a first-order phase transition, a stochastic gravitational-wave (GW) signal is generated. For a certain region of the parameter space, the resulting signal can lie within the sensitivity of future GW experiments.

We study the analytic structure of the QCD phase diagram by treating temperature as a complex variable. The nearest Yang-Lee edge singularities in the complex $T$ plane bound the domain of analyticity of temperature-dependent thermodynamic observables and complement the more commonly studied singularities in the complex chemical-potential plane. Our analysis combines three complementary perspectives: universal critical scaling, a first-principles extraction from lattice-QCD data, and explicit illustrations in effective models. We illustrate the resulting structure in a random-matrix model and in a quark-meson model, where the singularity trajectories can be followed explicitly. At small real chemical potential, the leading complex-temperature singularity admits an analytic expansion in $μ^2$, while near a critical point it crosses over to the universal Puiseux form dictated by Ising critical scaling. We show that the complex-$T$ and complex-$μ$ trajectories are controlled by the same scaling variables and mapping coefficients, so their comparison provides a stringent consistency test of critical-point searches and constrains the extent of the critical scaling regime. Finally, we analyze lattice-QCD data at $μ=0$ using an iterated conformal-Pade approach and extract the continuum location of the nearest complex-temperature singularity. The result is consistent with the expectation that, at physical quark masses, the real part of the leading singularity lies between the chiral-limit transition temperature and the physical-mass chiral-susceptibility peak temperature, while its imaginary part remains nonzero.

The analytic integration of single-scale Feynman integrals emerging in perturbative calculations in quantum field theories can be performed within special classes of functions, which appear as consecutive generalizations of the polylogarithm in form of Kummer-Poincaré iterative integrals over special alphabets and extensions thereof. These are the polylogarithms, Nielsen integrals, the iterated integrals over linear denominator terms, cyclotomic letters, letters induced by quadratic forms, and square-root valued letters. These integrals are solutions of first-order factorizing differential equations. They are related to specific nested sums via the Mellin transform and their expansions around $x=0$. We construct the $μ$-extensions of these iterated integrals and the associated nested sums. We present closed form solutions or provide algorithms in the case of more involved cases to derive the respective $μ$-extensions and study the algebras of the $μ$-extended function spaces. Except for the case of square-root valued alphabets, the $μ$-extension maps into the same function space polynomially in $μ$. This is also the case for the associated nested sums. For square-root valued alphabets or sums containing central binomials, the $μ$-extension leads to higher transcendental functions. In all other cases the $μ$-extension preserves the Hopf algebra structure implied by the (quasi)shuffle product, by supplementing $μ$ to the ground field.

We formulate a one-to-one correspondence between states and defects for higher-form gauge theories in arbitrary dimensions. The correspondence is not predicated on conformal invariance, as these theories are in general not conformal. Instead, it relies on the existence of infinitely many conserved charges associated with the mixed anomaly of electric and magnetic higher-form symmetries. In $p$-form Maxwell theory, these charges generate an extended Kac-Moody algebra that acts simultaneously on states and on extended operators. We show that this algebra organizes the Hilbert space on $S^p \times S^{d-p-1}$ into highest-weight representations allowing for a direct identification between states and $p$-dimensional defects. In particular, Wilson-'t Hooft defects dressed with local gauge-invariant operators are mapped to squeezed energy eigenstates. We further relate these novel symmetries to higher-spin currents and demonstrate that their construction persists in a class of interacting non-linear electrodynamics theories.

The Gertsenshtein effect can be used to probe the stochastic gravitational wave background at high frequencies, well above the range of standard cosmological sources. In this paper, we revisit the conversion between electromagnetic and gravitational waves in the magnetosphere of magnetars by solving the evolution equations of the associated Stokes parameters. In the process, we point out that the adiabatic approximation usually taken in the literature is not generally justified in the context of the Gertsenshtein effect. To derive analytical results, we focus our attention on two specific geometries where the adiabatic approximation is valid. From these, we derive a lower bound on the stochastic gravitational wave background from the conversion of magnetar electromagnetic emission into gravitational waves, and an upper bound by requiring that the conversion of background gravitational waves into electromagnetic radiation does not exceed the observed magnetar flux in the X-ray band. Our results demonstrate that gravitational waves generated through the Gertsenshtein conversion of magnetar electromagnetic emission produce a negligible stochastic background, as anticipated.

We construct Wigner-like equations of motion for symmetric continuous-spin fields in anti-de Sitter space of arbitrary dimension, treating both the bosonic and fermionic cases. We generalise the classical flat-space Wigner constraints for bosonic continuous-spin fields, and for the fermionic case we adopt the equations proposed by Bekaert and Mourad as our starting point. This is achieved by covariantising the ordinary derivatives and deforming the resulting constraints so that they form a closed algebra. The construction is carried out in a metric-like formalism and yields a system of first-class constraints that define a representation of the $\mathfrak{so}(2,D-1)$ isometry algebra, realised via the Lie-Lorentz derivative. Using these constraints, we compute the eigenvalues of the quadratic and quartic Casimir operators and compare the obtained values with Metsaev's classification of continuous-spin representations for both the bosonic and fermionic cases. A crucial algebraic role is played by special operators of the (super)algebra Howe-dual to the Lorentz subalgebra $\mathfrak{so}(1,D-1)$: the $\mathfrak{sl}(2)$ Casimir operator in the bosonic case, and the Casimir's ghost of the $\mathfrak{osp}(1|2)$ superalgebra in the fermionic case. Both objects naturally organise the constraint algebra and ensure its consistency.

The Large Magellanic Cloud (LMC), the most massive satellite of the Milky Way (MW), can significantly perturb the local dark matter (DM) distribution. We study its impact on directional DM detection using the Auriga cosmological simulations of a MW analogue hosting an LMC analogue. We find that the LMC induces strong anisotropies in directional recoil signals, driven primarily by the non-zero mean azimuthal velocity of the local DM distribution. The characteristic ring-like feature predicted in the Standard Halo Model (SHM) for heavy DM and low recoil energies is strongly distorted, producing an asymmetric recoil pattern concentrated at preferred azimuthal angles. Differences between recoil maps for the MW-LMC analogue and the SHM reach up to $\sim80\%$ near the signal maximum. These distortions significantly enhance directional discovery prospects, reducing the number of events required to reject isotropy by nearly a factor of five for a 100 GeV DM particle in a near-future CYGNUS-like experiment, and by even larger factors for heavier DM. Our results highlight the importance of the LMC for interpreting and optimizing future directional DM searches.

This is a streamlined introduction to the global (infrared) completion of Maxwell-type higher gauge fields (as in the higher gauge sectors of higher dimensional supergravity and its brane probes) by electromagnetic flux quantization in differential nonabelian cohomology, using cohesive homotopy theory. Applications include D/NS brane charge in (unstable) K-theory, M-brane charge in unstable Cohomotopy and geometric engineering of topological quantum order on probe M5-branes.

The real scalar singlet extension of the standard model provides a minimal framework in which the Higgs sector can realise a strong first-order electroweak phase transition and improve the stability of the electroweak vacuum. We combine the electroweak phase transition and high-scale vacuum stability with current and projected collider probes, including precision Higgs measurements, the Higgs trilinear coupling, EWPO and resonant searches for a heavy singlet-like scalar in $ZZ$ and di-Higgs final states. Focusing on a singlet heavier than the standard model Higgs, we find that there is parameter space compatible with a strong first-order electroweak phase transition for singlet-like scalar masses up to nearly 1 TeV. Deviations in the Higgs self-coupling can be larger than those in the Higgs--$Z$ coupling, making Higgs-potential measurements a key probe. We find that the HL-LHC will test a large fraction of the parameter space, while the FCC will provide ultimate discovery and model-discrimination capabilities.

The Simons Observatory will be extended by three Small Aperture Telescopes by 2027, increasing the total number of these instruments to six. We study the prospects for probing the reheating temperature and the inflaton coupling with this configuration, assuming a discovery of primordial gravitational waves in benchmark scenarios with a tensor-to-scalar ratio r=0.0036 or r=0.01. In popular plateau models of inflation, such an observation would fix the scale of inflation and enable determination of the order of magnitude of the reheating temperature and the inflaton interactions. For QCD-driven Warm Inflation the reheating temperature and inflaton coupling to gluons could, under optimistic assumptions, be measured with a precision of a few percent. Such a measurement would imply a clear prediction for complementary inflaton searches in axion experiments, paving the way toward probing the mechanism responsible for the initial conditions of the hot Big Bang in the laboratory.

A key property of a global symmetry's anomaly is its order: the smallest integer $n$ for which the diagonal symmetry of the $n$-copy system is anomaly-free. While many familiar lattice anomalies have finite order, perturbative anomalies in the continuum$-$those captured by Feynman diagrams$-$have infinite order. In this paper, we show that the Onsager symmetry, a lattice realization of the chiral symmetry of a 1+1d massless Dirac fermion, has an order-two anomaly. However, imposing lattice CPT symmetry enhances this anomaly from order two to infinite order, yielding a lattice chiral symmetry structure that more faithfully matches the continuum chiral anomaly. We also discuss the corresponding 2+1d symmetry-protected topological phases for these infinite-order lattice anomalies.

A central issue in lattice formulations of chiral gauge theories is how the anomaly cancellation mechanism of the continuum theory can be realized at finite lattice spacing. In the present paper, based on non-Abelian bosonization, we propose a lattice formulation of the bosonic theory corresponding to a two-dimensional non-Abelian chiral gauge theory. In the continuum theory, the gauge anomaly of chiral fermions is represented, in the bosonized description, as anomaly inflow from a three-dimensional Chern--Simons-type bulk contribution contained in a gauged Wess--Zumino--Witten model. Motivated by this structure, we introduce gauge-neutral spectator fermions and use the resulting bosonized description. We then construct a lattice counterpart of the gauged Wess--Zumino--Witten model with a three-dimensional bulk extension under appropriate smoothness conditions. A salient feature of this lattice formulation is the cancellation of the left and right bulk contributions in the exponentiated action. This cancellation occurs even before taking the continuum limit when the anomaly-free condition is satisfied, namely when the left and right representations have identical quadratic indices. Thus, the present construction realizes the anomaly-cancellation mechanism at finite lattice spacing via the bosonized description of two-dimensional anomaly-free chiral gauge theories. Establishing the desired continuum limit remains an important open problem.

The $η$ and $η^{\prime}$ mesons are nearly unique in the particle universe since they are nearly Goldstone bosons, and their decay dynamics are strongly constrained. While earlier experiments collected samples of order $\sim 10^{9}$ $η$, the proposed REDTOP (Rare Eta Decays To Observe Physics Beyond the Standard Model) facility targets $\mathcal{O}(10^{14})$ $η$ and $\mathcal{O}(10^{12})$ $η'$, enabling broad searches for physics beyond the Standard Model. In this work, we present studies evaluating REDTOP sensitivity to processes that couple the Standard Model to New Physics through four portals: the Vector (dark photon), the Scalar (Higgs-mixing), the Axion-like, and the Heavy Lepton. In parallel, the proposed statistics allow precise tests of $CP$ and $T$ invariance and lepton universality and improve determinations of the $η/η'$ transition form factors, which are crucial inputs to the hadronic light-by-light contribution to the muon anomalous magnetic moment $(g-2)_μ$.

One of the most important neutrino interactions is the Inverse Beta Decay (IBD). However, the IBD events typically carry no directional information in water Cherenkov detectors as the positron directions are mostly isotropic at low energies, such as those in supernova studies. As Gadolinium is being added to Super-Kamiokande, the improved neutron capture efficiency not only allows better background rejection, but the neutron capture information could potentially provide additional information that allows better event reconstruction. Due to neutron diffusion in water, event-by-event reconstruction is difficult. However, if the final neutron capture position is correlated with the initial neutrino momentum, it may be possible that neutrino directionality could be reconstructed statistically, with or without using the positron information. In this work, we use Geant4 to simulate neutron propagation in water. We show that in a wide range of neutrino energies from about 10 MeV to several hundred MeV, neutron capture information could statistically enhance the neutrino directionality, compared to positron-only inference, even with neutron diffusion considered. However, practical application of this technique depends crucially on detection effects, especially the vertex reconstruction resolutions. Our work therefore motivates developments of better reconstruction algorithms and techniques, as well as detector upgrades.

Using $e^+e^-$ annihilation data collected at 26 center-of-mass energy points between 3000.00 and $3119.88~\text{MeV}$ with the BESIII detector, corresponding to a total integrated luminosity of about $440.7~\text{pb}^{-1}$, we study the cross section lineshape of $e^+e^-\to K_S^0 K^+π^-+c.c.$. The relative phases and magnitudes between $J/ψ$ strong and electromagnetic decay amplitudes are measured to be $(123.7\pm5.3)^\circ;4.31\pm0.22$ or $(-123.1\pm5.2)^\circ;4.38\pm0.22$, with corresponding branching fractions $\mathcal{B}(J/ψ\to K_S^0 K^+π^-+c.c.)=(5.17\pm0.20)$ or $(5.36\pm0.20)\times10^{-3}$. Based on a partial wave analysis, the cross sections of $e^+e^-\to\bar K^0 K^*(892)^0+c.c.$ and $e^+e^-\to K^+ K^*(892)^-+c.c.$ are obtained. For these subprocesses, the relative phases and magnitudes are determined as $(155.2\pm15.5)^{\circ};3.67\pm0.27$ or $(-154.1\pm15.5)^{\circ};3.71\pm0.25$ and $(180.1\pm31.8)^{\circ};25.06\pm2.51$, respectively. The large relative phases deviate from the orthogonality relation expected from experiment and from the assumption of purely real amplitudes by more than $3σ$. The measured branching fractions $\mathcal{B}(J/ψ\to\bar K^0 K^*(892)^0)+c.c.=(4.18\pm0.18)$ or $(4.31\pm0.19)\times10^{-3}$, $\mathcal{B}(J/ψ\to K^+ K^*(892)^-+c.c.)=(7.09\pm0.28)\times10^{-3}$ are all consistent with the world average values, but achieve better than a twofold improvement in precision. The ratios between the branching fractions of $J/ψ\to\bar K^0 K^*(892)^0+c.c.$ and $J/ψ\to\bar K^+ K^*(892)^-+c.c.$ are $\mathcal{R}_{K^*\bar{K}}=0.589\pm0.012$ or $0.612\pm0.013$. After subtracting the electromagnetic contribution, the corresponding strong amplitude ratios are $\mathcal{R}^{3g}_{K^*\bar{K}}=0.884\pm0.050$ or $0.909\pm0.044$, which deviate $2.3$ or $2.1σ$ from the unity. No evidence for direct $CP$ violation is observed.

In the $E_8\timesωE_8$ octonionic unification program, each $E_8$ branches as $SU(3)_{st}\times E_6$, supplying one geometric $SU(3)_{st}$ per branch, while the split-complex unit $ω$ grades the visible and pre-gravitational branches; matter and gauge content is carried by $E_6\times E_6$, with chiral fermions realized as Cl(6) minimal-ideal spinors rather than $E_8$ representation components -- which places the chiral sector outside the Distler-Garibaldi no-go theorem. Comparing the 496-label two-branch adjoint reservoir with the 208 structures matched in the Generalized Trace Dynamics Lagrangian leaves a residual 288. We argue that this 288 is an adjoint-lineage representation-label ledger -- bookkeeping for the scaffolding -- and not a particle spectrum. The bifermionic seed is Hermitian, with $E_6$-covariant channels classified by $\bar{27}\otimes 27 = 1\oplus 78\oplus 650$: each branch's 78 supplies the gauge currents and a composite electroweak doublet, while the $E_6$ singlet is electroweak-inert. The charge-sum sector A is absent from the bare seed, and the 252 $SU(3)_{st}$-charged labels cannot be matter bilinears in any reservoir, conditional on the spinor ontology. The size of $E_8\timesωE_8$ is thus the dimension of a label ledger, not a count of particles; beyond the Standard Model, the framework's content is sterile neutrinos and a second composite scalar.

We demonstrate that treating Earth matter effects via a constant-density approximation introduces a fundamental systematic error in long-baseline neutrino oscillation analyses. Using exact numerical propagation through realistic PREM profiles, we show that matter-profile mismodeling does not merely affect the $ν_μ\rightarrowν_{e}$ appearance probability, but generates correlated biases across the $ν_μ\rightarrowν_τ$ and $ν_μ\rightarrowν_μ$ channels as dictated by PMNS unitarity. Our stochastic analysis reveals that the $ν_μ\rightarrowν_τ$ channel is the most volatile carrier of the geophysical systematic. Across varying correlation lengths at baselines like $5000$ km and $7000$ km, the $τ$-appearance channel consistently carries a larger mean bias and variance than the standard $ν_μ\rightarrowν_{e}$ appearance channel. These findings demonstrate that spatially resolved density treatments are a mathematical necessity for the analysis frameworks of future precision neutrino facilities.

The T4-2-i topology provides a one-loop realization of Majorana neutrino mass and may be viewed as a radiative extension of the type-II seesaw, with a scalar triplet, two inert scalar doublets, and singlet fermions propagating in the loop. A central difficulty in realizing this topology lies in the simultaneous presence of tree-level type-I and type-II seesaw contributions arising from the same particle content. In addition, the stability of the dark-matter candidate typically requires the introduction of an ad hoc discrete symmetry. In this work, we revisit the T4-2-i topology within a nonholomorphic modular-invariant framework based on the double-cover group $T'$. The presence of both even- and odd-weight polyharmonic Maaß forms considerably enlarges the space of allowed modular structures, while the residual $\mathbb{Z}_2$ symmetry associated with the vicinity of the fixed point $τ=i$ naturally stabilizes the lightest odd state. The modular assignments forbid the dangerous tree-level contributions, determine the flavor structure of the lepton sector, and allow both fermionic and scalar dark-matter candidates. We confront the model with neutrino-oscillation data, charged-lepton-flavor-violating bounds, electroweak precision observables, the Higgs diphoton signal strength, the observed dark-matter relic abundance, the cosmological bound on the sum of neutrino masses, and direct-detection limits. Focusing on the fermionic dark-matter candidate, in which the lightest odd state is the Majorana fermion $N_1$, we find that both normal and inverted neutrino mass orderings remain viable. In the allowed region, the relic abundance is largely controlled by coannihilation with the inert scalar partners, while the spin-independent direct-detection rate remains naturally suppressed because it arises only through a loop-generated Higgs portal.

Although measurements of the Cosmic Microwave Background (CMB) are consistent with a nearly scale-invariant primordial spectrum of adiabatic perturbations, in which the energy densities of different components (radiation, baryons, and dark matter) fluctuate proportionally, there could also exist isocurvature perturbations, in which density fluctuations of the individual components differ from the adiabatic mode. Cold dark matter isocurvature (CDI) perturbations with a variety of spectral tilts generated in pre-inflationary axion models provide one such example. In this article, we present the most updated constraints on these axion CDI perturbations using the latest CMB anisotropy measurements from Planck, the Atacama Cosmology Telescope (ACT), and the South Pole Telescope (SPT). We study both fixed spectral indices with values ranging from red- to blue-tilted spectra as well as the case with a free index. We find that the constraint on the spectral index gets moderately improved with the combined datasets compared to Planck alone, while the bounds on the isocurvature amplitudes for the fixed spectral indices we consider do not get tighter. We also discuss the theoretical implications of our constraints, in particular for models giving rise to blue-tilted spectra.

We investigate a harmonic interpretation of the hidden flavor coordinates (Q,G,C) previously introduced in a low-rank ternary description of the fermion mass spectrum. The generation coordinate is associated with the odd harmonic mode sequence n_G = (1,3,5), allowing the fermion mass ansatz to be rewritten in a compact harmonic form. The corresponding logarithmic spectrum reveals an additive structure in which the hidden coordinates contribute directly to the observed mass hierarchy. We argue that the projected quantity L = QG + C, which organizes the fermion masses, does not uniquely characterize a fermion state. This motivates the introduction of the full hidden-coordinate vector X = (Q,G,C) and a geometric interpretation of flavor based on coordinate separations and antisymmetric invariants defined within the hidden-coordinate space. The harmonic framework further suggests a possible extension beyond the fermion sector. Combining the fermionic harmonic modes generates the bosonic sequence n_B = (2,4,6,8,10), which exhibits a suggestive correspondence with the electroweak and Higgs scales. Although the resulting framework remains exploratory, it suggests that fermion masses, flavor structure, and bosonic states may reflect different aspects of a common hidden harmonic organization.

We investigate a harmonic interpretation of the hidden flavor coordinates (Q,G,C) introduced previously in a low-rank ternary description of the fermion mass spectrum. In this picture, the generation coordinate is associated with the odd harmonic sequence (1,3,5), allowing the fermion mass formula to be expressed in a compact harmonic form. The logarithmic mass spectrum then acquires an additive structure in which the hidden coordinates contribute directly to the observed hierarchy of fermion masses. We further argue that the scalar quantity (L=QG+C), which successfully organizes the fermion spectrum, does not uniquely identify a fermion state. This observation motivates the introduction of the full hidden-coordinate vector X=(Q,G,C) and a geometric interpretation of flavor in terms of distances, relative orientations, and antisymmetric invariants defined within the hidden-coordinate space. The harmonic framework also suggests a possible extension beyond the fermion sector. Combining the fermionic harmonic modes naturally generates the even sequence (2,4,6,8,10), which exhibits a suggestive correspondence with the electroweak and Higgs scales. While this correspondence remains speculative, it points toward a common harmonic structure underlying both fermionic and bosonic states. Although the framework is exploratory, it indicates that fermion masses, flavor organization, and bosonic excitations may represent different manifestations of a deeper hidden harmonic structure.

The lattice QCD calculation of generalized form factors are exploited to determine the subtraction constants through all order dispersion relations of Deeply Virtual Compton Scattering (DVCS). The leading order relation is found to constrain significantly the real part of the Compton Form Factors (CFFs), and the higher order one reduces considerably both the real and imaginary part of CFFs in a global analysis of proton data. This is realized by a synthesis of the DVCS data and LQCD calculations within a neural network framework, whose architecture is specifically designed for a reliable extrapolation to unmeasured kinematic regime. By leveraging dispersion relations beyond leading order, our framework allows for adding higher moments of generalized parton distributions (GPDs) from LQCD into the extraction of CFFs from DVCS data.

Heavy quark expansion theory provides the standard theoretical framework for understanding the lifetimes of weakly decaying heavy-flavor hadrons. Within this framework, the matrix elements of four-quark operators play a crucial role in explaining the lifetime differences among hadrons containing the same heavy quark. In this work, we calculate these matrix elements at leading order using full QCD sum rules, with particular emphasis on deriving the corresponding HQET sum rules by taking the heavy-quark limit. While this limit is straightforward for most contributions, it turns out to be rather nontrivial for certain ones. Numerical analyses show that the results obtained from the two approaches are consistent with each other. We further clarify the origin of the large discrepancies reported in the literature between full QCD sum rule and HQET sum rule results. This work contributes to a deeper understanding of the relationship between the two sum-rule approaches and is expected to facilitate future applications of QCD and HQET sum rules in the study of heavy-flavor hadrons. The findings presented here may be of interest to both practitioners of QCD sum rules and researchers working on effective field theories.

We investigate an empirical low-rank structure underlying the fermion mass spectrum. The construction is based on an integer exponent matrix L=QG+Be, where Q labels shell type, G labels generation, and Be introduces a correction affecting only the first generation. The resulting matrix reproduces the observed hierarchy of fermion masses using a small number of integer parameters. The construction naturally introduces hidden coordinates X=(Q,G,C) associated with each fermion species. We show that fermion masses depend primarily on the projected quantity L, while flavor information appears to require the full hidden-state coordinates. Possible implications for the observed structure of the CKM and PMNS mixing matrices are briefly discussed.

Protostellar jets are one of the earliest forms of stellar feedback, but their impact on star formation and cluster assembly in centrally concentrated molecular clouds remains poorly understood. We study how protostellar jets affect the star formation efficiency, the temporal variability of star formation, star cluster structure, and the early dynamical state of centrally concentrated, newly forming star clusters using the Torch star cluster formation framework. We adopt a centrally concentrated initial cloud model with mass M = 2.5 x 10^3 solar masses and compare six pairs of simulations with and without protostellar jets, supplemented by one additional higher resolution pair of simulations. We analyze our simulations using global star formation diagnostics together with structural and dynamical measures of the stellar population. Models with jet feedback achieve star formation efficiencies of 12-16%, while the corresponding models without jets yield higher efficiencies of 19-33%. Jets also cause star formation to occur in discrete bursts rather than continuously, to produce more extended and substructured stellar systems, and to leave behind stellar populations that are less tightly bound and have higher virial parameters. In our centrally concentrated initial conditions, runs with jets form stellar systems that better reproduce the observed range of the projected structural parameter Q_2D in young clusters than runs without jets, indicating that protostellar jets are an important early feedback channel even in centrally concentrated clouds that regulates star formation efficiencies and shapes the emerging cluster structure.

Cosmic voids and superclusters are among the largest structures in the Universe and provide complementary probes of the growth of large-scale structure and the underlying gravitational field. We present a comprehensive catalogue of the main local Universe cosmic web voids and knots, using the updated CosmicFlows-4++ Zone of Avoidance (CF4++ZOA) catalogue out to redshift z=0.1. To do this we use the V-web algorithm which provides a dynamically motivated map of the the local cosmic web, tracing the major expanding and converging regions of the local Universe. The robustness of the catalogue is assessed using the ensemble of Hamiltonian Monte Carlo realizations of the CF4++ZOA reconstruction. We additionally remove any structure that exceeds the survey boundaries and present catalogues of 37 voids and 42 knot regions within the reconstructed survey volume. The identified emptying regions (voids) have effective radii ranging from 13 to 38 h-1Mpc. The high density converging regions (knots) have volumes ranging from 10^4 to 3.3x10^5 h-3 Mpc^3.

One effective way of learning about the production and properties of dark matter in the early universe is by extracting information about the primordial dark-matter phase-space distribution from the matter power spectrum. Several years ago a simple empirical formula was introduced which successfully reproduces most of the salient features of the primordial dark-matter phase-space distribution from the matter power spectrum -- even in situations in which this distribution is non-thermal, multi-modal, or exhibits other complicated features. Continuing this line of research, we investigate the extent to which machine-learning techniques can improve upon this analytic approach. Interestingly, we find that a one-dimensional convolutional neural network not only succeeds in reconstructing the dark-matter phase-space distribution with greater accuracy, but can also be applied to a broader range of matter power spectra.

Context. Active galactic nuclei (AGN) jets are fundamental drivers of galaxy evolution, injecting kinetic energy into their environments. The large-scale morphology and spectral properties of these radio galaxies are consequences of complex particle acceleration, energy loss, and absorption processes. While the shape of the synchrotron spectrum encodes the plasma's energetic history, understanding the physics of particle acceleration and duty cycles has historically been limited by a lack of well-resolved observations at ultra-low frequencies (< 100 MHz), where the oldest cosmic ray electron populations are traced. Aims. This study aims to perform the first comprehensive multi-frequency analysis of bright extended radio galaxies down to 58 MHz. The goal is to study electron acceleration mechanisms, accurately measure the low-frequency spectral shape, and constrain the injection index for a sample of Fanaroff-Riley (FR) I and II galaxies using spectral ageing models. Methods. Utilising new 58 MHz observations from the LOFAR Low Band Antenna (LBA) combined with LOFAR High Band Antenna (HBA; 144 MHz) and Rapid ASKAP Continuum Survey (RACS, 887 MHz & 943.5 MHz & 1367.5 MHz) data, a sub-sample of 22 extended sources from the 3CRR catalogue was selected, requiring the largest angular size to be at least 2.5'. The analysis involves constructing detailed spectral index maps and utilising radio colour-colour diagrams to interpret spectral shapes and constrain ageing model parameters across the radio lobes. Results. This study presents the ultra-low frequency spectral index maps for this sample. For FR I galaxies, spectral indices range from ~0.5 near the core (consistent with first-order Fermi acceleration) to > 1.0 in the lobes. For FR II galaxies, hotspots exhibit steep low-frequency spectra (0.5 - 0.9), suggesting complex acceleration or absorption effects.

We investigate the molecular gas content and interstellar medium (ISM) conditions of REBELS-25, a massive, star-forming galaxy at $z=7.31$. Deep VLA Q-band and ALMA Band 3 observations reveal CO(3-2) and CO(7-6) emission (both at $\sim3.5σ$), and provide an upper limit on [C I](2-1). From the CMB-corrected CO(3-2) flux-representing the highest-redshift detection of a low-$J$ CO transition to date-we derive a molecular gas mass of $M_{\rm mol}=(1.0\pm0.4)\times10^{11}\,(α_{\rm CO}/(3\,$M$_{\odot}$(K$\,$\kms$\,$pc$^2)^{-1}))\,$M$_{\odot}$, directly confirming the presence of a very massive gas reservoir only $\simeq700\,$Myr after the Big Bang. This implies an extreme gas fraction of $f_{\rm gas}\simeq0.95$, a gas-to-dust ratio of $δ_{\rm GDR}\simeq6\times10^2$, and a depletion timescale of $τ_{\rm dep}\simeq1.2\,$Gyr, broadly consistent with extrapolated scaling relations for main-sequence galaxies at lower redshift. Using the radiative transfer code TUNER, we self-consistently model CO and dust continuum emission in the context of the significant CMB background, constraining ISM properties and recovering $M_{\rm mol}= (1.8^{+1.0}_{-0.9})\times10^{11}\,$M$_{\odot}$, independent of assumptions about $r_{31}$ and $α_{\rm CO}$. We further discuss the use of alternative molecular gas tracers at early epochs. Combining CO and [C II] measurements, we infer an empirical [C II]-to-H$_2$ conversion factor of $α_{\rm [C II]}=(60\pm25)\,$M$_{\odot}$/L$_{\odot}$, suggesting [C II] remains a viable molecular gas tracer in the Epoch of Reionization. These results demonstrate the detectability of low-$J$ CO emission even at $z>7$, paving the way for next-generation facilities, and provide critical insights into the rapid mass assembly of galaxies during the first billion years of cosmic history.

With Hubble Space Telescope (HST) and James Webb Space Telescope (JWST) observations of NGC~628 spanning 0.3--7.7\,$μ$m, we fit the spectral energy distributions (SEDs) of over 12,000 optically-selected star clusters, emerging young star clusters (eYSCs), and MIRI-selected sources with \textsc{cigale} to derive their ages, masses, extinctions, and dust properties. We find that near-infrared selected eYSC-I (compact Pa$α$ and 3.3,$μ$m PAH emission) and eYSC-II (compact Pa$α$ and diffuse 3.3,$μ$m PAH emssion) sources peak at $\sim$3--5~Myr, where $\sim 12\%$ of the clusters have an $E(B{-}V)>2$, demonstrating the presence of dust-embedded populations as clusters emerge. Further, the distributions of the fractional polycyclic aromatic hydrocarbon (PAH) abundance ($q_{\mathrm PAH}$) and stellar-to-nebular attenuation ratio ($E(B{-}V)_{\rm \star}/E(B{-}V)_{\rm neb}$) suggest an evolutionary sequence in which sources evolve from eYSC-I to eYSC-II as clusters clear their surrounding dust and gas. The photo-dissociation region (PDR) clearing timescale inferred from the ratio of eYSC-I to optically visible stellar clusters is $\sim$4~Myr. Additionally, we find that star clusters in the spiral arms of NGC 628 are preferentially more massive and more dust-reddened than those in inter-arm regions.~Finally, we find that $\sim$65\% of eYSC-I, $\sim$27\% of eYSC-II, and $\sim$40\% of F335M-selected sources coincide with an F770W peak in our MIRI-selected catalog within 4 pixels, confirming that F770W-bright sources preferentially trace the youngest and dustiest regions. Overall, our results highlight the ability of JWST together with \textsc{cigale} model grids to identify and characterize eYSCs during their short-lived embedded phases, and provide constraints on the feedback mechanisms that govern the emergence of stellar clusters.

We analyse the cluster content of the $\sim 4.5 \text{ deg}^{2}$ XMM-LSS and CDFS VIDEO fields which are expected to be partially covered by the upcoming MOONRISE survey. Using AMICO and WaZP photometric redshift-based cluster finders, we construct a sample of $519$ cluster candidates detected by both finders in the redshift range $z = 0.1-3$, including $74$ detections at $z > 1.5$. For all detections, we identify the Brightest Central Galaxy (BCG) and compute a list of probabilistic cluster memberships. Our photometric redshift measurements of the clusters agree well with spectroscopic redshifts from the literature, when available. From ancillary spectroscopic data, we assign $z_\text{spec}$ measurements to $116$ cluster candidates based on their spectroscopic members and to $204$ based on their likely BCGs. We also show that candidates containing Radio-Loud members are efficiently recovered using the prior-based cluster finder PPM. We perform a preliminary analysis of the galaxy content of these candidates, focusing on the Red-Sequence components of their apparent Colour-Magnitude Diagram. By comparing with models of galaxy evolution, we show that this population is consistent with a model of passive evolution with a formation at high redshift, and is already in place at $z = 1.5-2.0$. Finally, our cluster sample is used to evaluate how these clusters would be detected and characterised, according to various MOONRISE strategies. We show that cluster spectroscopic confirmation and characterisation could be efficiently achieved up to $z\sim1.7$ even with the shallowest survey strategy. This open unprecedented insight into the physical properties of high-redshift galaxy clusters and into galaxy formation in dense environments.

Using generalised Laplacian symmetry operators, we construct basis sets or Coulomb resolutions in several separable coordinate systems, including two R-separable systems. This expands the possible geometries in which basis set construction is feasible, a problem which is relevant to both galactic dynamics and computational chemistry. In particular we derive three basis sets (two in prolate spheroidal and one in cylindrical coordinates) which are expressible in closed-form using a single Jacobi polynomial. We also show how any spherical polar or prolate spheroidal basis set may be transformed into a bispherical or toroidal basis set.

SN 2018bsz is the closest known stripped superluminous supernova (SLSN-I) to date, making it an ideal laboratory for investigating the physical mechanisms powering this class of extreme explosions. We present a multi-epoch X-ray spectroscopic study of SN 2018bsz based on four Chandra observations followed by one XMM observation, spanning 87 to 1253 days after explosion. The source is detected at all Chandra epochs and is also tentatively detected in the late XMM observation, although more uncertain due to nearby contaminating sources. Regardless of the XMM detection, this makes SN 2018bsz the second X-ray detected SLSN-I and the third X-ray detected SLSN overall. We explore potential power sources for the observed X-ray emission and find that a millisecond (ms) magnetar central engine underpredict most of the observed X-ray luminosities and fail to reproduce the relatively flat light curve. Accounting for ejecta absorption further increases the discrepancy. While asymmetries and magnetar-driven ionization could reduce the effective absorption, ionization breakout is expected years after our observational window. Instead, the observations are more readily explained by early-time interaction between the ejecta and the circumstellar medium, while the magnetar emission is absorbed by the ejecta. This scenario is supported by the flat temporal evolution, previous optical results, and inferred mass-loss rates which resemble those of stripped SNe that later evolve into interacting systems. Our results thus favor the scenario where SN 2018bsz is part of a distinct group of SLSN-I, where interaction is crucial for the strong emission.

Gas-phase chemical abundances are key observable consequences of galaxy evolution, being intrinsically tied to galaxy formation histories. Gas metallicity rises with increasing stellar mass ($\mathrm{M_*}$), forming the well-known mass-metallicity relation (MZR). MZR residuals have separately been shown to anti-correlate with star-formation rate (the ``fundamental'' metallicity relation), with gas mass and with optical size, but no single analysis has considered all trends together. We thus perform a combined analysis of all three trends, utilizing optical MaNGA integral field spectroscopy, HI-MaNGA gas masses, and MaNGA DynPop dynamical masses. We estimate inner gas masses for $\sim$1500 star-forming galaxies, finding this to be the most important parameter after $\mathrm{M_*}$ in predicting gas metallicities. We obtain equivalent results for stellar metallicities and gaseous N/O, suggesting that current inner gas masses are intrinsically linked to long-term chemical evolution histories. We show that more compact galaxies have lower dynamical masses, challenging suggestions that deeper gravitational potentials confer higher metallicities. We find a strong correlation between inner gas mass and galaxy size, meaning that short term inflow fluctuations cannot be responsible for the MZR residuals. With chemical evolution models, we show that our results can instead be explained by differences in long-term inflow histories. The earlier inflow histories of compact galaxies lead to lower gas masses and more rapidly declining gas reservoirs at late times, leading to higher metallicities. At fixed stellar mass, galaxy size therefore encodes the link between halo assembly histories, long-term gas inflow histories, current gas reservoirs and metallicity.

Optical-to-near infrared interstellar polarisation, induced by aligned dust grains, generally follows a convex wavelength dependence, known as the Serkowski relation. However, observations in the ultraviolet (UV) and at [mid-]infrared wavelengths indicated that some of the spectra do not follow this relation. Specifically, about 25% show an excess in the degree of polarisation at mid-UV wavelengths ($λ^{-1} > 3\,\rm μm^{-1}$), referred to as the super-Serkowski polarisation. In this study, we re-examine both the Serkowski and super-Serkowski spectra based on the joint effect of paramagnetic relaxation (DG alignment) and radiative torque (RAT) alignment. We used the observational data for HD 30614, HD 204827, HD 37903 and HD 161056 to constrain our modelling. We examined two types of radiation fields: one derived from the scaled interstellar radiation field and another originating from a B-type star. For the super-Serkowski spectra of HD 30614 and HD 204827, our model demonstrates that RAT alignment enhanced by radiation produced from a B-type star below the Lyman limit ($λ=912 Å$) can reasonably explain the observations and that a combination with the DG alignments results in a better fit for $λ^{-1}\geq 5.5\,\rm μm^{-1}$. For the Serkowski spectra in HD 37903 and HD 161056, only RAT alignment by itself under the typical interstellar radiation field above the Lyman limit, within a typical cold neutral medium, can account for the observed spectra, with a combination of a very inefficient DG alignment. The capacity of our model to predict the starlight polarisation spectrum up to from infrared to far-UV is thus a promising tool for interpreting future missions that observe spectrophotometry in the UV bands.

This study reports the nature of the companion star to the pulsating ULX NGC 1313 X-2, using long-slit spectroscopic data from Gemini-South observations, based on archival data from 2009. After stacking flux-calibrated spectra from ten nights of observations and fitting the spectra with stellar templates, we find a possible Balmer break in the GMOS-S spectrum below 4000 Angstroms, which is suggestive of an A-type supergiant donor. Using the inferred stellar radii, we report updated constraints on the orbital parameters of the system and on the nature of the binary. We also add some information on the accretion disc size scale by studying the X-ray and optical variability using the lag-frequency spectrum and corroborate on results from earlier studies for the gas bubble expansion rates by modelling the [O III] emission line profiles, allowing constraints on the kinetic power of the wind/jet relative to the accretion power. This study also expands on previous efforts to study the formation history of the binary using multi-wavelength observations.

The first generation of stars formed from pristine, neutral hydrogen gas. The most massive of these exploded as supernovae within a few million years of their birth, producing the first heavier elements and leaving distinct chemical signatures of their origin in the surrounding medium. However, chemical abundance studies have so far mainly relied on emission-line measurements, which are luminosity weighted and hence biased towards the most recently formed stars. Here we analyse near-infrared, medium-resolution spectroscopy from the JWST-SPURS program of three UV-bright galaxies at redshifts 7.8, 8.6, and 9.3, within the first 650 to 520 million years after the Big Bang. The chemical abundance patterns of the metal lines detected in absorption hint at extremely metal-poor gas, substantially lower than inferred from the emission lines tracing the central, star-forming regions. Further, they all exhibit super-solar [C/O] abundances, which is also imprinted in the averaged spectrum of a larger set of galaxies at similar redshifts. These results reveal the distinct chemical signatures of the first Population III supernovae explosions.

The $Λ$CDM cosmological model has been successful in explaining many astronomical observations. However, recent observations increasingly point to deviations from the standard $Λ$CDM framework. Among these, one of the most significant discrepancies is the \textit{Hubble tension}, which refers to the difference in values obtained for the Hubble constant $H_0$ from high-redshift measurement and local observation. To address this issue, numerous cosmological models and methodological approaches have been proposed. This review offers a concise overview of recent progress in resolving the Hubble tension. The combination of Dark Energy Spectroscopic Instrument (DESI) Baryon Acoustic Oscillations (BAO) and uncalibrated Type Ia supernovae data yields a value for $H_0$ that is significantly higher than the $Λ$CDM predication based on early-universe probes, even without incorporating local distance ladder constraints. This result indicates that the origin of the Hubble tension lies in new physics at low redshifts. Our findings suggest that although many unresolved systematics persist in current observations, they are insufficient to account for the magnitude of the current Hubble tension. This implies the likely existence of new physical mechanisms that have yet to be discovered.

Fast radio bursts (FRBs) are millisecond-duration radio flashes of unknown origin, observationally classified into repeating and apparently nonrepeating (one-off) populations. In this work, we use a statistical population approach to investigate the redshift evolution of one-off FRBs. We compare a pure star formation history (SFH) tracing model, phenomenological delayed models, physically motivated delayed models that correspond to binary neutron star related and neutron star age-window channels, and mixture models which is obtained when two physically motivated models are normalized separately and then combined as a weighted mixture. The samples in CHIME/FRB Catalog do not support an intrinsic event rate density that directly follows the SFH. The preferred model is mixture model corresponds to an effective mean delay time of $\barτ=1.426^{+0.032}_{-0.035}~\mathrm{Gyr}$. These results suggest that the current data may naturally explained by delayed, possibly multi-channel progenitor evolution for the one-off FRBs.

The gravitational wave background (GWB) produced by extreme-mass-ratio inspirals (EMRIs) serves as a powerful tool for probing the astrophysical and dynamical processes in galactic centers. EMRI systems are a primary target for the space-based detector LISA due to their long-lived signals and high signal-to-noise ratios. This study explores the statistical properties of the GWB from EMRI, focusing on the calculation methods for the GWB, the astrophysical distribution of EMRI sources, and the influence of key parameters, including the spin of supermassive black holes (SMBHs) and the masses of compact objects (COs). By analyzing these factors, we determine the distribution range of the characteristic strain of the GWB from EMRIs. We find that the final eccentricity distributions appear to have negligible effect on the intensity of the GWB due to rapid circularization before they become detectable and the spin of the SMBH enhances the GW characteristic strain by approximately 1$\%$ compared to cases without spin effects. The masses of COs can also significantly affect the characteristic strain of the GWB from EMRIs, with Black Hole (BH) as CO producing a GW signal intensity that is approximately one order of magnitude higher compared to cases where Neutron Star (NS) or White Dwarf (WD) are the COs.

The masses of galaxy clusters carry cosmological and astrophysical information. We develop a simulation-based inference pipeline to infer cluster masses from full projected phase-space information of member and interloper galaxies. Our method combines a permutation-invariant Deep Sets architecture with neural posterior estimation using normalizing flows, enabling the recovery of expressive posterior distributions. We train the model to predict residual corrections to the classical $M$--$σ$ relation, thus explicitly isolating information beyond velocity dispersion. Using the Uchuu-UniverseMachine simulation, we evaluate the method under both idealized (interloper-free) and realistic (cylindrical) observational setups. In the idealized case, our model reduces the scatter in mass estimates to as low as $\sim 0.1$ dex, representing a twofold improvement over the traditional $M$--$σ$ relation. In the cylindrical setup, we achieve comparable performance at the high-mass end ($> 10^{14.5}\,M_\odot/h$), demonstrating robustness against interloper contamination. We demonstrate that set-based simulation-driven inference provides a powerful and flexible framework for galaxy cluster mass estimation, enabling improved accuracy and reliable uncertainty characterization for upcoming large-scale surveys. Our model saturates the kinematic information content and thus suggests a baseline for future studies.

The JWST discoveries of supermassive black holes (BHs) at $z \gtrsim 5$ may provide key insights into their seeding origins. Using new $[18{-}72~\rm Mpc]^3$ BRAHMA cosmological simulations, we investigate how variations in heavy-seed prescriptions, coupled with a subgrid dynamical friction model, shape BH populations at $z \sim 5$ and $z \sim 0$. We consider two "lenient'' seed models, in which all halos containing sufficient dense & metal-poor gas form $\sim10^4$ and $\sim10^5~M_{\odot}$ seeds, and a "strict'' seed model, in which $\sim10^5 M_{\odot}$ seeds form only under additional constraints motivated by direct collapse black hole formation. By $z \sim 5$, all models produce $M_*-M_{\rm BH}$ relations broadly consistent with the observed local Universe for $M_*\gtrsim10^9~M_{\odot}$ galaxies, but only the lenient scenarios generate systems near the upper envelope of the observed local scatter. In galaxies hosting $M_{\rm BH} \sim 10^8$-$10^9~M_{\odot}$ BHs, lenient production of $\sim10^5~M_{\odot}$ seeds also produces multiple overmassive systems with $M_{\rm BH}/M_* \gtrsim 0.01$. Although their growth is dominated by seeding and mergers, these systems reach luminosities of $\sim10^{43}$-$10^{45}\mathrm{erg s^{-1}}$, comparable to those inferred for JWST-detected BHs. As a key observational signature, the lenient seed models yield merger rates of $\gtrsim100\mathrm{yr^{-1}}$ and near-unity local BH occupation fractions even in galaxies with $M_* \lesssim 10^7~M_{\odot}$. In contrast, the strict seed model produces merger rates of only $\sim1\mathrm{yr^{-1}}$ and local occupation fractions of $\lesssim10\%$ for galaxies with $M_* \lesssim 10^8~M_{\odot}$. Future gravitational-wave event rates and measurements of local BH occupation fractions will therefore provide strong constraints on the dominant pathways responsible for high-redshift BH assembly.

Synchrotron emission in high-energy transients is produced by relativistic electrons accelerated by shocks. As high-energy transients are often unresolved even on angular scales probed by very long baseline interferometry, it is difficult to obtain a full picture of the ejecta and circumstellar medium (CSM) properties that are probed by the radio synchrotron emission. Radio spectra of high-energy transients frequently show optically thick slopes shallower than the standard $F_ν\propto ν^{5/2}$ expected from synchrotron self-absorption (SSA) models, or broader spectra near the self-absorption frequency. Such deviations are often interpreted phenomenologically, without providing clear insights into the structure of the emitting region. We show how information on the homogeneity and symmetry of the emitting region can be directly inferred from SSA spectra, even when the source is unresolved. We discuss the circumstances under which inhomogeneities in the emitting region can change the spectrum below the self-absorption frequency, causing it to follow a different slope. We examine which parameters can be constrained from observations and which remain degenerate. We apply this method to the stripped-envelope supernova (SN) 2016coi and to the fast blue optical transient (FBOT) AT2018cow, showing that SSA spectra constrain the degree of inhomogeneity in these systems, providing strong evidence for inhomogeneities in the emitting region in the SN 2016coi, and asymmetry in the case of AT2018cow, and we infer the characteristics of the emitting region. When well sampled spectra are available, our method can be applied as a general, model-independent, inference method. This approach can be used to constrain inhomogeneities in a variety of unresolved high-energy astrophysical transients, including SNe, FBOTs, tidal disruption events and gamma-ray bursts.

We present a data-driven investigation of Galactic diffuse emission. Using multi-frequency Planck radio/microwave maps (30-857 GHz) and Fermi-LAT gamma-ray data (50 MeV-814 GeV), we construct a nonlinear mapping between radio emission and gamma-ray intensity through supervised machine learning. Our models achieve high predictive accuracy (R^2 > 0.90 in the 0.1-10 GeV range), demonstrating that multi-frequency radio observations encode sufficient information to reconstruct both spatial morphology and spectral properties of diffuse gamma-ray emission. By analyzing model performance across different frequency bands and spatial regions, we identify high-frequency radio bands as the dominant predictor, providing direct empirical support for the hadronic origin of Galactic 0.1-10 GeV gamma rays, while low-frequency radio bands for the leptonic origin above 10 GeV. Residual maps reveal coherent large-scale structures, including Loop I and III, highlighting regions where standard interstellar emission models are incomplete or biased. Compared with the GALPROP model, our machine learning approach yields a higher R^2=0.95 and lower mean absolute relative error (14.7%) in the inner Galactic disk and the Galactic center region. Our results illustrate that machine learning serves as a physically interpretable tool for multi-messenger astrophysics, providing a data-driven baseline for separating non-standard emission components and deriving new constraints on cosmic-ray propagation and interstellar medium structure.

We identify a geometric obstruction to resolving the Hubble tension by combining early-time sound-horizon reduction with late-time smooth dark energy. Within $Λ$CDM, the BAO--SN matter-density gap $ΔΩ_m = 0.037$ is exactly invariant under the sound-horizon rescaling $α\equiv r_s^{\rm mod}/r_s^{Λ{\rm CDM}}$, and late-time $w(z)$ deformations cannot eliminate this gap either: reconciling the two datasets requires \emph{opposite} deformations -- phantom ($w < -1$) for BAO, quintessence ($w > -1$) for SN at $z < 0.5$ -- an anti-alignment quantified by $\cosθ= -0.97$ in $w(z)$ space. A full MCMC analysis of DESI DR2 BAO, Planck plik\_lite, and Pantheon+ bears this out: the optimal $α^* = 0.992$ ($0.8\%$ $r_s$ reduction) brings the joint fit to $H_0 = 70.3 \pm 0.3\;\mathrm{km\,s^{-1}\,Mpc^{-1}}$, still $3.2σ$ below SH0ES, with the inter-dataset tension reduced but not removed. The obstruction reflects not a shortage of model freedom but an irreducible disagreement between probes. The deformation space $\{α, β_{\rm damp}, w(z)\}$ already spans $93\%$ of the $Ω_m$ response direction; nonetheless BAO and SN constrain $Ω_m$ through independent channels and disagree, while the residual $H_0$ deficit, anchored by the local distance ladder, resides in the absolute distance scale that $w(z)$ reshapes but cannot rescale.

Strong lensing of massive black hole binaries can produce multiple gravitational-wave images with different magnifications and arrival times. LISA signals remain in band for months to years, allowing multiple lensed images to overlap during the inspiral stage and generate beat patterns. A singular isothermal sphere lens model is adopted to describe the lensing configuration, and two-image beat waveforms are constructed from massive black hole binary signals. To isolate the beat pattern itself, waveform mismatch is evaluated only during the overlapping inspiral stage before the coalescence of the first image, excluding contributions from the delayed merger peak of the second image. Using the HS-nod-SN (B+20) strong-lensing population, the occurrence rate of identifiable beat events is estimated, and Bayesian parameter estimation is performed with a beat template. Beat patterns are most readily identified when the lensing time delay is short and the delayed image has a relatively large magnification. Among 196 detectable two-image lensed events, 92 satisfy the temporal-overlap condition and 14 satisfy the beat-identification criterion, corresponding to an identifiable beat fraction of about 7\%. Posterior inference shows that the beat template can recover the lensing time delay and magnification parameters for a representative beat event. These results indicate that lensed beat patterns constitute a distinguishable subset of strongly lensed LISA events and provide a unique observational signature of strong lensing in the LISA band.

We present an analysis of XMM-Newton and Chandra observations of 85 nearby main-sequence FGK stars with age estimates ranging from 0.2-12 Gyr. We measure quiescent 0.3-10 keV luminosities, variability metrics, and multi-temperature thermal plasma spectral parameters. Quiescent spectra are typically described by three characteristic plasma components ($kT\approx0.1$, 0.4, 0.8 keV); the fraction of flux from $T\ge7$ MK rises with X-ray surface flux, reaching $\sim$50% for $F_X\gtrsim10^6$ erg cm$^{-2}$ s$^{-1}$. We derive relations between emission measure-weighted coronal temperature and both $L_X$ and $F_X$, enabling temperature-informed count-rate conversions for faint sources. We quantify how bandpass conversions (ROSAT 0.1-2.4 keV vs. XMM-Newton 0.3-10 keV) depend on temperature, and show that inferred ROSAT-band $L_X$ broadly follows the canonical $t^{-1.5}$ decay, while the harder band exhibits increased scatter at $>$4 Gyr. Several stars show excess activity suggestive of age errors, inclination effects, or unresolved companions. Some of these "outlier" stars are potential direct imaging targets for the Habitable Worlds Observatory, and detailed characterization of these stars is needed to inform their likely influence on the atmospheric evolution of orbiting planets.

High-mass X-ray binaries (HMXBs) provide a natural laboratory to study radiatively driven stellar winds under strong external X-ray irradiation. As the compact object moves along its orbit, the wind density and ionization structure vary with orbital phase, leaving characteristic signatures in X-ray emission and absorption features. The amplitude and morphology of this variability depend on the system geometry, including the orbital inclination (via line-of-sight and occultation effects) and the orbital eccentricity (via phase-dependent changes in the orbital separation). We present a computational framework that connects phase-resolved spectroscopic variability to the three-dimensional structure of irradiated winds, and that enables inference of physically meaningful wind--irradiation parameters within a Bayesian setting. We combine photoionization calculations with a parametric wind description and orbital geometry to construct three-dimensional maps of density and ionization. From these maps we compute orbital-phase-dependent diagnostics, accounting for geometric occultation and wind inhomogeneity through a clumping prescription. We then use Bayesian inference to compare model predictions with phase-resolved observables and to quantify parameter constraints and degeneracies. The framework reproduces the main orbital-phase-dependent trends expected for irradiated winds and yields robust constraints on the system ionization balance. While combinations of wind and luminosity parameters are well constrained, individual parameters can remain partially degenerate depending on the orbital configuration and data quality. This modular and computationally efficient approach provides a route to interpret HMXB phase variability in physical terms, and offers a foundation for future extensions toward forward spectral modeling and population-level applications.

We investigate how low-ionization gas flows in typical star-forming galaxies at $z\sim3$ depend on galaxy intrinsic properties and viewing angle. For this analysis we use JWST/NIRSpec observations of rest-frame near-UV Fe II and Mg II absorption, and rest-frame optical Na D absorption. This study combines galaxies from the LyC22, EXCELS, and AURORA surveys and contains 176, 197, and 315 galaxies, respectively, with Fe II, Mg II, and Na D coverage. Based on both individual and composite spectra, we find no statistically significant correlations between outflow velocity and galaxy properties. However, galaxies with detected outflows tend towards higher stellar masses, SFR, and $Σ_{\rm SFR}$ than those without outflows, suggesting that the two samples are not drawn from the same parent population. Finally, we additionally find that Mg II emission is preferentially detected in galaxies with lower stellar mass and $A_V$, and higher sSFR, consistent with conditions that favor the escape of resonantly scattered line and ionizing continuum radiation. We present the first evidence in $z\sim3$ star-forming galaxies that properties of the absorption lines depend on galaxy inclination, with more face-on systems showing stronger absorption and higher outflow velocities, while inflowing gas is more frequently detected in more highly inclined galaxies. These trends are consistent with observations at $z\lesssim1$ and predictions from cosmological simulations in which galactic winds are launched perpendicular to the galactic disks, while accretion occurs primarily along the disk plane.

Using a sample of 151 galaxies with dynamically measured black hole (BH) masses ($M_{\rm BH}$), we investigate the scaling relations between $M_{\rm BH}$ and the stellar velocity dispersion, $σ$, and, for a subsample of 30 core-Sérsic galaxies, between $M_{\rm BH}$ and the size of the partially depleted core, $R_{\rm b}$. Core-Sérsic galaxies, identified using high-resolution $Hubble ~ Space ~ Telescope$ imaging and spanning both the normal-core $(R_{\rm b}<0.5$ kpc) and large-core ($R_{\rm b}>0.5$ kpc) regimes, define an updated $M_{\rm BH}$$-$$R_{\rm b}$ relation of the form $M_{\rm BH} \propto R_{\rm b}^{1.16 \pm 0.10}$, with an rms scatter of $Δ_{\rm rms} \simeq 0.28$ dex in $\log M_{\rm BH}$. We find that Sérsic and normal-core galaxies together follow a common log-linear $M_{\rm BH}$-$σ$ relation with a slope of $4.95 \pm 0.29$ and a scatter $Δ_{\rm rms} \simeq 0.46$ dex. Deviations from this relation arise at the highest BH masses, where large-core galaxies, including six with direct $M_{\rm BH}$ measurements, drive a significant upturn. We find that these galaxies typically host ultramassive black holes whose masses scale more strongly with $R_{\rm b}$ than $σ$, and lie $\sim (1-4$) $~\times$ the intrinsic scatter (0.39 dex) above the relation defined by Sérsic and normal-core galaxies. The $M_{\rm BH}$$-$$R_{\rm b}$ relation shows $\sim 30$-$47\%$ less scatter in $\log M_{\rm BH}$ than the corresponding $M_{\rm BH}$$-$$σ$ relation for the same sample. We interpret the high-mass upturn in the $M_{\rm BH}$$-$$σ$ diagram as a consequence of successive major, dry mergers, a scenario that naturally explains the observed flattening of the $σ-L_V$ relation at $M_V < -23.5$ mag.

This study presents a comprehensive X-ray spectroscopic analysis of 642 quasars, obtained by cross-matching the XMM-Newton Serendipitous Source Catalog (DR11) with the Sloan Digital Sky Survey (DR16) quasar catalog. After stringent quality filtering and automated spectral reduction, we derived reliable photon indices ($Γ$) and intrinsic 2--10 keV X-ray luminosities. Using multiwavelength data, sources were classified into 561 radio-quiet (RQ) and 81 radio-loud (RL) quasars. We estimate the bolometric luminosity and Eddington ratio ($λ_{\mathrm{Edd}}$) from absorption-corrected X-ray measurements and virial black hole masses. Our primary objective is to establish and characterize the fundamental relationship between the photon index and Eddington ratio for RQ population. We find that RQ quasars exhibit systematically higher Eddington ratios, peaking at $\log λ_{\mathrm{Edd}} \approx -1.2$, and softer spectra with $Γ\approx 2.0$. A statistically significant positive correlation between $Γ$ and $λ_{\mathrm{Edd}}$ is detected in RQ quasars, supporting disk--corona coupling models. To validate our results within the broader context of AGN evolution, we further examine the dependence of $λ_{\mathrm{Edd}}$ on redshift ($z$) and black hole mass ($M_{\mathrm{BH}}$). For RQ quasars, $λ_{\mathrm{Edd}}$ increases with redshift and decreases with $M_{\mathrm{BH}}$, in strong agreement with recent results \cite{aggarwal2024evidence}, highlighting the universal nature of these accretion trends. By correlating spectral slope with accretion rate, this work provides new insights into the interplay between accretion physics, jet activity, and the cosmic evolution of quasars.

Carbon-enhanced metal-poor (CEMP; with $\rm{[Fe/H]} \le -2.0$ and $\rm{[C/Fe]} \ge 0.7$) stars preserve information about early chemical enrichment, low-mass star formation, and the hierarchical assembly of galaxies. In this study, we have compiled an extensive literature sample of 1032 stellar carbon abundances spanning the metal-poor Milky Way halo (437 stars), 21 ultra-faint dwarf galaxies (UFDs; 102 stars), seven classical dwarf spheroidal galaxies (254 stars), three accreted dwarf galaxies (90 stars), the Small Accreted Stellar Systems (SASS; 77 stars), and eleven stellar streams (72 stars). We establish the fractions of CEMP stars for each of these systems and categories. Generally, the low-mass UFDs possess the high fractions at low metallicities, whereas the more massive classical dwarf galaxies have relatively few CEMP stars. This behavior reveals a new low-metallicity Magnitude ($M_{\rm V}$)--CEMP Fraction relation across the dwarf satellite galaxy population. The high CEMP fractions in surviving UFDs suggest their enrichment was dominated by faint supernovae, as higher energy input would likely have quenched star production. The low CEMP fractions in classical dwarfs imply predominantly in situ formation rather than assembly from smaller systems. Using $\rm{[C/H]}$ abundances, we also probe early low-mass star formation. Eight stars lie within or near the theoretical ''forbidden zone'', indicating that dust-induced cooling, alongside fine-structure line cooling, contributed to early star formation. These rare dust-cooled stars may have formed in UFD-like systems that did not survive. Overall, the metal-poor Milky Way halo appears to have assembled from many different dwarf galaxies, with CEMP halo stars being contributed by early UFD-like systems and non-CEMP halo stars by intermediate-sized halos that later formed classical dwarfs.

We describe spectroscopic tomographic weak lensing measurements (spectrotomography) for two rich clusters of galaxies, A1767 and A2065, based on extensive spectroscopy and Subaru/Hyper Suprime-Cam (HSC) imaging. These detections represent the first use of spectrotomography based on archival Subaru/HSC imaging. The measurements depend only on galaxies with spectroscopic redshifts reported here. The approach cleanly separates cluster members from the background and suppresses systematics that may be introduced by the use of photometric background redshifts. We detect the tomographic shear signals at $3.1σ$ (A1767) and $3.5σ$ (A2065). The shear signal amplitudes are consistent with the cluster dynamical (caustic) masses and they scale appropriately with source redshift. However, comparison with the first spectrophotometric detection of A2029 based on DECam imaging reveals some subtle potential systematic issues in deriving the shear signal for the relatively bright background galaxies used in the analysis. These issues may be important for understanding future more extensive applications of spectrotomography based on further Subaru imaging, as well as Euclid and LSST data. The total of three spectrophotometric detections (A1767, A2029, and A2065) sets the stage for broader application of the technique for unbiased cluster weak lensing mass determinations and potentially for a geometric cosmological test that is independent of other methods.

We perform a cosmological zoom-in simulation of a massive galaxy ($M_s\sim10^{10}\rm M_\odot$ at $z\sim10$) using the GIZMO code. By employing $\leq 3\rm pc$ resolution and a $3.4\rm Myr$ supernova feedback delay, we capture the feedback-free starbursts (FFB) in clusters. The simulation reproduces FFB model predictions and super-bright galaxies observed by JWST. At $z\sim10$, cold streams feed a compact galaxy ($R_{\rm e}\sim1\rm kpc$), with stellar and surface densities ($>10^5\rm cm^{-3}$, $>10^5\rm M_\odot pc^{-2}$) exceeding FFB thresholds. The global star-formation efficiency (SFE) is $\varepsilon_s\sim0.2\text{--}0.3$, associated with a fluctuating star-formation history. We identified over $10^5$ star clusters ($M_{\star}>10^{4.5}\rm M_\odot$) with a nearly scale-free mass distribution (${\rm d}N/{{\rm d}\log M}\propto M^{-1.06}$). Approximately 90\% of star formation occurs in clusters, which at a given time constitute $30\text{--}40\%$ of the total stellar mass. The star formation in most of the clusters of masses $<10^7\rm M_\odot$, occurs in bursts of $<3\rm Myr$ and a local SFE $\sim0.5\pm 0.2$. Cluster metallicities ($-2.01<\log (Z/Z_\odot)<-0.45$) indicate rapid baryon recycling. Feedback-driven outflows exhibit typical temperature of $10^7\rm K$ and typical velocities of $\sim 2000\rm km\ s^{-1}$. In the highly dynamic central $1\rm kpc$, clusters undergo rapid orbital decay and merge to assemble the oblate nuclear stellar cluster. Cluster shapes range from oblate to prolate, with a triaxial median. These clusters are consistent with JWST observations, and a fraction of them may survive to yield the globular clusters (GCs) at low redshifts.

We constructed the 2D model of the Galaxy which initially includes an analytical bar, bulge, disk and halo. The model disk forms the outer elliptical resonance rings R1 and R2 located near the outer Lindblad resonance of the bar (OLR), as well as the inner resonance ring r located near the corotation radius (CR). As the density of stars in the elliptical rings increased, we introduced additional gravitational perturbations created by the rings. The radial component of gravitational perturbations from the elliptical rings, F_R, at a point with the Galactocentric coordinates (R, theta) was represented as a combination of three polynomials in powers R/Re or Re/R, where Re is the distance to the midline (middle) of the ring at a given angle theta. The azimuthal component of the disturbances, F_T, was calculated using the force F_R. The difference between the values of the force F_R (F_T) calculated using the numerical differentiation of the potential and using the analytical representation does not exceed 5.7% (1.3%) of the maximum value of the force F_R generated by the elliptical rings. In general, the gravity of the elliptical rings has little effect on the process of adjustment of epicyclic motions near the OLR of the bar.

The inner Milky Way hosts overlapping stellar populations (bar--bulge, inner thin and thick disks, and halo), complicating population assignments along the line of sight. A joint chemical--dynamical approach is required to isolate a clean field bulge sample. Metal-poor bulge stars are valuable as they likely trace the earliest phases of chemical enrichment in the inner Galaxy. We aim to characterize the alpha-element (Si, Mg) and selected Fe-peak abundances of a dynamically defined sample of bulge field stars, and to contrast these trends with the inner-halo tail in the metal-poor regime. We analyze metal-poor candidates from the Pristine Inner Galaxy Survey observed by the bulge Cluster APOGEE Survey. Using APOGEE/ASPCAP abundances (S/N >= 50), we integrate full 6D orbits in a barred Milky Way potential. Bulge membership is defined via orbital confinement using an apocenter cut and the high-density locus in the (E_J, L_z) plane, with uncertainties estimated using Monte Carlo simulations. We identify 98 stars as bulge members. The metallicity distribution spans -2.5 <= [Fe/H] <= -0.4, with a median [Fe/H] = -1.71, offset to higher metallicity than the inner halo (median [Fe/H] = -1.96). The sample follows a high-alpha sequence with slopes d[Si/Fe]/d[Fe/H] = -0.020(+0.012,-0.051) and d[Mg/Fe]/d[Fe/H] = -0.097(+0.062,-0.133). Fe-peak tracers show [Ni/Fe] ~ 0, while [Mn/Fe] declines with [Fe/H] with a mild upturn at higher metallicity. We detect no significant [Si/Fe] gradients with R_apo or Z_max (+0.010(+0.018,0.000) and +0.006(+0.022,-0.011) dex kpc^-1, respectively). Results are insensitive to Omega_p, yielding indistinguishable memberships. The chemo-orbital evidence favors an in situ origin within the inner Galaxy for the metal-poor bulge field, enriched at early times and rearranged by secular bar evolution, with a minor contribution from halo stars at the most metal-poor end.

Modern surveys such as DESI and \textit{Euclid}, which collect data for hundreds of millions of galaxies to map the large-scale structure (LSS) of the Universe, hold the key to determining the cosmological parameters and testing new physics. This ambition, however, is limited by uncertainties in the galaxy-halo connection: the link between observed galaxies and the underlying, unobservable matter field, by accounting for effects such as galaxy bias and assembly bias (AB). These are particularly poorly-understood for modified gravity (MG) models, which are popular alternatives to the cosmological constant to explain accelerated expansion. We approach this problem using mock emission line galaxy (ELG) and luminous red galaxy (LRG) catalogues in $f(R)$ gravity matching the specifications of ongoing Stage-IV galaxy surveys, generated from state-of-the-art MG hydrodynamical simulations. While the interplay between MG -- especially the chameleon screening mechanism -- and galaxy formation leaves complicated imprints in the galaxy-halo connection, a simple physical picture emerges in which halo and galaxy formation are enhanced for progressively more massive haloes over time. We confirm that the basic galaxy-halo connection model, the halo occupation distribution (HOD), in which galaxy occupation is determined solely by halo mass, underestimates galaxy clustering strength in $Λ$CDM by $10$--$20\%$ at $z\lesssim1$ when neglecting AB, and demonstrate that MG introduces further complexity. Extending this model with a suitably-chosen environment density as a secondary HOD variable reduces the AB effect in all models to $2$--$3\%$ for $z\lesssim0.5$. This provides a well-motivated starting point for further works on minimising the impact of AB when testing non-standard cosmological models with LSS.

Accurate modelling of the covariance of clustering observables is essential to fully exploit current and future survey data, which is expected to constrain large-scale clustering signals with unprecedented precision. Computational costs of simulation-based estimates motivate analytical approaches, especially in light of the growing interest towards multi-tracer analyses and parity-odd signatures in two-point statistics, which respectively mitigate cosmic variance and probe relativistic projection effects on cosmological scales. In this work, we generalise previous theoretical results for the Gaussian covariance of multi-tracer power spectrum measurements, providing a general expression applicable to both real (even-parity) and complex (both even- and odd-parity) power spectra. We focus on a generic weighted estimator at first, and then showcase how our general formalism applies to Legendre power spectrum multipoles and two-dimensional power spectrum, recovering known limits in appropriate cases. We validate our predictions against Gaussian Monte Carlo simulations and investigate the structure of the covariance matrix, including the Hermitian properties of its imaginary part.

We derive both the mid-plane and off-plane rotation curves, $v_c(R,z)$, and the vertical acceleration, $a_z(R,z)$, of the Milky Way (MW) using \textit{Gaia}~DR3 data over the ranges of vertical heights $z \in (-2,2)\,$ kpc and galactocentric distances $R \in (8.5,14)$ kpc where the velocity components are determined with high precision, i.e., with an error $< 5\%$. In contrast, the vertical acceleration $a_z(R,z)$ is dominated by model-dependent systematics, with uncertainties of up to $\sim 20\%$. This level of accuracy allows us to place stringent constraints on the geometry of the MW's dark matter (DM) distribution, as the vertical gradients of the gravitational potential attain their maximum within this range of radial and vertical distances corresponding to the characteristic scales of the disk. We find that models including the observed stellar components together with a spherical DM halo fail to reproduce both the pronounced variation of $v_c(R,z)$ with height and the observed behavior of $a_z(R,z)$. In particular, spherical halos with a scale radius of $r_s \sim 15$ kpc contribute negligibly to the off-plane rotation curve and vertical acceleration in the inner disk, leaving these features primarily determined by the stellar mass distribution. Conversely, models in which DM is confined to a flattened, disk-like configuration predict substantial contributions to both $v_c(R,z)$ and $a_z(R,z)$, resulting in a markedly better agreement with the data. We conclude that disk-like DM distributions are strongly favored over spherical halo models. Forthcoming Gaia data releases will enable even more stringent tests of the geometry and distribution of the MW's DM component.

The spectra of an accretion disk around black holes are the basic diagnostic tool to enlighten the underlying flows and then black holes. Accretion flows around black holes, however, are controlled by parameters like the magnetic field, spin of the black hole, accretion rate and temperature of the flow. These quantities affect the (magneto)hydrodynamics of the flow thus consequently lead to variations in the spectrum. We first consider numerical steady state magnetohydrodynamic (MHD) solutions of magnetized accretion flows around black holes to study the dependence of the spectra on these disk properties. The spectrum exhibits strong dependence on the spin of the black hole, accretion rate, magnetic field and the electron temperature of the flow. Variations in these quantities influence the emission peaks and overall luminosity, which can be a tell-tale sign to extract physics of observed spectra. We further validate our results with general relativistic MHD (GRMHD) simulations using the standard and normal evolution (SANE) and magnetically arrested disk (MAD) vector potentials. We consider two black hole spins ($a=0.5$ and $a=0.9375$) to model the magnetic field configurations and study the resulting spectra by comparing MAD and SANE results. We find a large difference in the bolometric luminosities and the location of the emission peaks between SANE and MAD flows. Certain properties of the spectra, like, the ratio of synchrotron radiation to synchrotron self-Comptonization peaks in SANE and MAD, show drastically distinct features. The overall luminosity combined with such metrics can distinguish the magnetic field characteristics in astrophysical systems.

We present a spatially resolved chemical abundance analysis of 2291 star-forming galaxies at $z \leq 0.1$, spanning nearly four orders of magnitude in stellar mass ($8 \le \rm log (M_{\star}/M_{\odot}) \le 11.5$), by exploiting the multi-fibre spectra from the Dark Energy Spectroscopic Instrument (DESI) as a sparse integral field spectrograph. In the inner regions ($<2R_e$), the radial gas-phase metallicity profiles show an outward-declining trend for massive galaxies, with the steepest gradient ($\nabla_{log(O/H)}$) $\sim-0.08$ dex/R$_{e}$, whereas low-mass dwarf galaxies exhibit nearly flat profiles ($\nabla_{log(O/H)}\sim-0.02$ dex/R$_{e}$). The large galactocentric radii ($\sim$5 R$_{e}$) probed in this study, reveal flat metallicity profiles near the disk-halo interface. Strikingly, these flat metallicity values are consistent across a wide stellar mass range, likely reflecting the influence of low SFR and metal poor inflows in the outer regions. The metallicity gradient - stellar mass relation exhibits a turnover at $\log(M_\star/M_\odot) \sim 10.5$, beyond which gradients become shallower, possibly driven by the chemical equilibrium in the inner disk of massive galaxies and/or dilution from cosmic gas accretion. At fixed stellar mass, a strong size dependence is observed, where compact galaxies show flatter gradients and higher central enrichment than their extended counterparts. The abundance gradients are further linked with the stellar age distribution within the galactic disk, where galaxies with younger outskirts show steeper gradients than the ones with older outskirts, consistent with ongoing inside-out disc growth sustaining centrally concentrated chemical enrichment. These results underscore the interplay of star formation efficiency, stellar feedback, and metal-poor gas accretion in governing the radial chemical structure in galaxies.

We present a Bayesian analysis of star clusters in the dwarf starburst galaxy NGC 1569 based on high-resolution Hubble Space Telescope imaging combined with integral-field spectroscopy from the Keck Cosmic Web Imager, obtained as part of the DUVET survey. For each cluster identified, we infer posterior probability distributions for mass and age using a forward modelling method that properly accounts for uncertainties due to stochastic sampling of the IMF. We investigate how the inferred properties depend on photometric coverage by repeating the analysis with different filter combinations, including mock extensions to the ultraviolet and near-infrared that emulate the addition of HST UV bands and James Webb Space Telescope imaging. We find that, while inclusion of these wavelength regimes further breaks age and mass degeneracies, the currently available data yields reasonably strong constraints on cluster parameters. We compare inferred cluster properties to the conditions of the local interstellar medium, and find evidence for multiple interesting correlations. The truncation mass of the cluster mass function varies with galactocentric distance, particularly moving off the disk, consistent with a dependence on the density of the interstellar medium. Cluster mass positively correlates with metallicity, suggesting that massive clusters preferentially form in pre-enriched gas, and the ionisation state of the gas, reflecting the increased prevalence of high-mass stars in high-mass clusters. These results demonstrate the power of Bayesian, initial mass function-aware modelling for resolving cluster populations in nearby starburst dwarfs and provide new insight into how cluster formation and feedback respond to local galactic conditions.

Magnetic fields are a fundamental ingredient of the Universe, influencing the formation and evolution of cosmic structures. While magnetic fields in local galaxy clusters have been studied, their origin, amplification, and strength at high redshift are poorly understood. Proto-clusters represent the early stages of galaxy cluster formation, ideal for investigating the early magnetisation of the intra-cluster medium (ICM). We present a study of CARLA J1510+5958 proto-cluster at z = 1.72, observed with the JVLA in the L-band (1-2 GHz). We aim to investigate the magnetic field strength and structure in the proto-ICM and the role of AGN in magnetising the environment during early cluster formation. We analyse Faraday rotation on the polarised emission from the central radio-loud AGN using the Rotation Measure (RM) synthesis and QU fitting technique. We further interpret the observations with 3D simulations of gas density and turbulent magnetic fields, varying AGN orientation and path length. The two AGN lobes show different polarisation properties. The Western lobe exhibits a uniform RM (average $-115 \pm 32\text{ rad m}^{-2}$, dispersion $36 \pm 11\text{ rad m}^{-2}$), indicating a locally ordered magnetic field likely compressed by the lobe, while the Eastern lobe is depolarised. Although the asymmetry indicates a turbulent, magnetised medium, simulations rule out a purely isotropic random field for the Western lobe RM distribution. The QU fitting further suggests an internal Faraday component, interpreted as magnetised relativistic plasma from the lobe mixed with the surrounding gas, indicating possible magnetisation of the ambient medium by the AGN. From this asymmetry, we constrain the average physical magnetic field in the proto-ICM to a lower limit of 0.4 $μ$G. These results confirm a magnetised proto-ICM at z = 1.72, proving early field amplification during cluster assembly.

The Coma cluster (Abell 1656; $z=0.023$) is a nearby rich galaxy cluster and a key laboratory for studying cluster assembly in the Cosmic Web. We characterize its projected dark matter distribution and connection to galaxies, the intracluster medium, and reported intracluster filaments (ICFs) with wide-field ($\sim$12-deg$^2$) Subaru/Hyper Suprime-Cam weak-lensing (WL) analysis. We reconstruct the two-dimensional mass distribution, fit Navarro-Frenk-White (NFW) models, derive an aperture mass densitometry profile, and compare the WL signal with optical spectroscopy, eROSITA X-ray observations, radio data, and gas fraction diagnostics. A single-halo NFW fit yields $M_{200\mathrm{c}}=8.2\pm0.7\times10^{14}~M_{\odot}$. The aperture mass profile agrees with the best-fit NFW model and the X-ray hydrostatic mass at $R\gtrsim20'$ ($\sim$560 kpc), suggesting little merger-induced bias in the global WL mass, while the inner region shows substantial hydrostatic bias. A two-halo NFW fit centered on NGC 4874 and NGC 4839 gives masses of $7.8\pm0.6$ and $0.9\pm0.2\times10^{14}~M_{\odot}$, implying a $\sim$1:8 minor merger. The gas mass fraction suggests that the system is returning from first apocenter. We find a positive spatial correlation between the WL signal and X-ray surface brightness, strongest along the ICF directions ($110^{\circ}$ and $340^{\circ}$), where shear-selected subhalos are predominantly detected. The Coma $r$-band mass-to-light ratio is radially constant with $\langle M/L_r\rangle\simeq250\pm66~M_{\odot}/L_{\odot}$ within $R_{200\mathrm{c}}$, whereas the northern and western ICFs show higher values of $\sim1000~M_{\odot}/L_{\odot}$, suggesting stronger dark matter dominance. These results show that joint WL and multiwavelength analyses can effectively probe cluster assembly and the dark matter content of ICFs.

Recent analyses of $fσ_8$ and weak-lensing data indicate that the linear growth rate at $z\lesssim 1$ may be lower than predicted by $Λ$CDM. This motivates models of dark matter in which large scale structure growth slows relative to $Λ$CDM at late times. We construct particle models in which dark matter experiences a drag with dark radiation that grows at late times, unlike conventional DM--DR interactions, which fade as the universe expands. A key ingredient is that the radiation interacting with the dark matter is produced at late times from dark matter decay. An explicit model, interacting Decaying Cold Dark Matter (iDCDM), adds two parameters beyond $Λ$CDM while leaving the background, BBN, and primary CMB intact. But it predicts a step-shaped suppression of the linear growth rate $f(k,z)$, a distinctive target for DESI, Euclid, and Rubin. Confronted with current data, iDCDM shows a modest preference over $Λ$CDM, driven by $fσ_8$, with $Δχ^2$ between $-2.7$ and $-7.6$ depending on the assumed scaling of the drag with redshift and on neutrino masses. The decisive test will come from upcoming $k$- and $z$-resolved growth measurements.

In the last decades a number of extremely faint and compact Galactic satellites (Ultra Faint Compact Satellites; UFCS) have been discovered by large panoramic surveys. Their nature is uncertain due to their location in the overlapping dwarf galaxy-star cluster region of the $M_V-R_h$ plane and their faintness and distance. Here we show how the deep HST photometry from the Missing Globular Clusters Survey (MGCS), combined with spectroscopic metallicities, provides new insight into the nature of these satellites through accurate distance and age estimates. We consider the case of Koposov 2, currently the most metal-poor bound star cluster known in the entire Milky Way or an extreme case of Ultra Faint Dwarf galaxy. By performing a spectroscopically-informed bayesian isochrone fit on the MGCS data we find $(m-M)_0=16.85\pm 0.06$ ($D=23.4\pm 0.6$ kpc) and age=$13.7^{+0.9}_{-1.3}$ Gyr, showing that, contrary to previous age estimates, Koposov 2 is as old as the oldest Galactic globular clusters. The luminosity function, corrected for incompleteness, is well reproduced by a model with the same age and metallicity and a slope of the mass function $x=-0.35$, suggesting a significant depletion of faint stars. We model the surface stellar density field, deriving new robust estimates of the half-light radius ($R_h=0.39^{+0.06}_{-0.04}$ arcsec, corresponding to $R_h=2.7^{+0.4}_{-0.3}$ pc), of the absolute integrated magnitude ($M_V=-0.95\pm0.22$) and of the stellar mass ($M_{\star}=371.8\pm41.6M_{\odot}$), showing that Koposov 2 is much more compact than dwarf galaxies of similar stellar mass. The new evidence significantly support the hypothesis that Koposov 2 is a star cluster that may have lost a large fraction of its original mass. Finally we show that most UFCS lie in the same locus of the $M_V - R_h$ plane as Galactic open clusters, hinting to a possible additional channel for their formation.

The globular cluster $ω$ Cen has been suggested to have originated in the Gaia Sausage-Enceladus (GSE) merger event, possibly as its nuclear star cluster. However, the present-day orbits of $ω$ Cen and the GSE debris are very different. We investigate the scenario in which $ω$ Cen originated in the GSE and migrated to its current position due to perturbations from the Galactic bar. The [$α$/M] distributions of stars located between the GSE debris and $ω$ Cen in $(L_z,E)$ space tentatively support this scenario, but are not conclusive. We run simulations of the GSE debris and $ω$ Cen in a realistic Milky Way potential with a decelerating bar at various present-day pattern speeds. We find that $ω$ Cen can indeed be traced back to the phase space region occupied by the GSE debris. However, this likely requires a pattern speed of $Ω_\mathrm{b}\lesssim26$ km s$^{-1}$ kpc$^{-1}$, which is much lower than most recent estimates. We conclude that a GSE origin for $ω$ Cen is dynamically and chemically plausible, but this would require a re-evaluation of the current consensus on the bar's pattern speed.

This chapter presents an overview of the unification models for Active Galactic Nuclei (AGN), focusing on the physical structures, classification schemes, and evolutionary processes that characterize accreting supermassive black holes. We introduce the fundamental components of AGN, including the supermassive black hole, accretion disk, jets, outflows, broad-line and narrow-line regions, polar dust and the dusty anisotropic obscurer. The traditional orientation-based unification model is reviewed, with a focus on the role of the covering factor of the obscuring material in shaping observed properties. We introduce the radiation-regulated unification model, which accounts for the influence of radiative feedback on the nuclear environment of SMBHs. We also examine the evolutionary aspects of AGN unification, including the impact of galaxy mergers, host galaxy properties, and redshift-dependent trends. Finally, we discuss changing-look AGN, which challenge conventional unification frameworks by exhibiting dramatic spectral variability.

Previous Sirocco radiative-transfer models of gas-cocooned AGN predicted lower-column counterparts to little red dots (LRDs): compact, X-ray-weak sources with bluer continuum slopes and Balmer jumps rather than Balmer breaks. The recently identified population of little blue dots (LBDs) closely resembles this predicted phase. Here we explore these lower-column-density cocoons in which nebular recombination emission remains visible while strong Balmer-continuum absorption is avoided. We find that a sequence of increasing column density connects more classical AGN spectra, Balmer-jump LBD-like spectra at $N_{\rm H}\!\sim\!{\rm few}\times10^{24} \mathrm{cm^{-2}}$, and Balmer-break LRD-like spectra at higher columns. In this sequence, electron scattering produces exponential line wings and suppresses X-ray emission before strong Balmer absorption features, characteristic of higher column densities, appear. We therefore propose that LBDs are lower-column analogues of LRDs within a common gas-cocooned AGN sequence. This interpretation predicts that Balmer-jump emission, X-ray weakness, permitted lines with exponential wings, He II $λ$4686 emission, smaller H$α$ FWHM values and equivalent widths than in LRDs, and weak or absent absorption features are characteristic of LBDs. We compare to three example LBD spectra and identify Balmer-jump signatures in them.