arXiv Daily Digest - 2026-06-11

We consider the problem of engineering autonomous intelligent agents that are capable to responsibly not comply with user requests. We argue that machine non-compliance comes in many different forms, and sketch the issues we should pursue on the road of accomplishing responsibly non-compliant intelligent machines. We anchor responsible non-compliance in justifications for task refusal, pathways to override the non-compliance, as well as careful tracking of security risks and liability transfers.

Rotary Position Embedding (RoPE) is widely adopted in Transformer models, yet its extension to high-dimensional domains lacks a unified theoretical formulation. Most existing approaches either apply rotations independently along each axis or empirically mix frequencies, which limits cross-dimensional interactions and yields direction-dependent representations. To address these limitations, we propose nD-RoPE, a decomposition-free generalization of RoPE to arbitrary dimensions. From a translation-invariant formulation in continuous Hilbert space, we derive a spectral condition for isotropy that requires treating positions and frequencies as coupled \(n\)-dimensional vectors. We instantiate this formulation with a multi-scale regular-simplex wave-vector design, which provides non-degenerate spatial coverage and a symmetric, directionally balanced second-order response. Experiments across images, videos, and point clouds demonstrate consistent performance gains and improved generalization in high-dimensional settings.

Accurate forecasting of sea surface temperature (SST) in regional seas such as the East Sea is crucial for monitoring marine ecosystems, assessing climate risks, managing fisheries, and conducting naval operations. Traditional numerical ocean models provide reliable predictions but are computationally expensive and often unsuitable for real-time forecasting. Many deep learning methods also struggle with high-dimensional spatiotemporal ocean data and experience error accumulation over longer forecasting periods. This study builds on our previously proposed Adaptive Next-Generation Reservoir Computing (Adaptive NVAR) framework, initially introduced and tested on synthetic dynamical systems, and extends it to ocean forecasting. We present a reduced-order forecasting framework that combines Singular Value Decomposition (SVD) with Adaptive NVAR to predict SST dynamics in the East Sea. SST fields are compressed into a low-dimensional representation using SVD, which extracts dominant modes of ocean variability. Adaptive NVAR models the temporal evolution of these latent states, and the predicted states are reconstructed into SST forecasts. We evaluate the framework using regional ocean datasets and compare it with the standard NG-RC/NVAR. Results show that Adaptive NVAR consistently achieves lower forecasting errors across multiple prediction horizons. In addition, SVD reduces computational complexity, resulting in a fast and scalable framework suitable for real-time ocean forecasting.

Sparse autoencoders (SAEs) are widely used to interpret neural network representations, but their utility depends on whether the learned features are reproducible across training runs. We study this question through \emph{feature stability}: for each SAE feature, we estimate the probability that a similar feature reappears in an independently trained SAE. This yields a scalable per-feature signal that separates stable from unstable features. In a large-scale study across seeds, models, layers, dictionary sizes, and SAE variants, we find a pronounced functional asymmetry: stable features carry most of the reconstruction- and prediction-relevant signal, while unstable features have weak marginal impact and are dominated by low-frequency surface-form triggers in both activation statistics and automatic explanations. Geometrically, unstable features are individually non-reproducible but concentrate in reproducible lower-rank subspaces, suggesting that seed dependence often reflects basis ambiguity within a shared region of activation space rather than pure noise. A controlled synthetic model makes this mechanism explicit, showing that low-rank ground-truth features can be recovered at the subspace level while remaining non-identifiable as individual SAE latents across seeds. Finally, by pooling unique cross-seed features, we construct more stable SAEs while preserving explained variance in this setting. Together, these results show that unstable features are not merely failed or noisy latents: they have weak individual functional impact, but reflect reproducible low-dimensional structure that standard SAEs resolve differently across seeds.

Low-rank optimal transport (OT) mitigates the quadratic scaling of classical solvers, yet existing approaches rely heavily on first-order mirror-descent updates that require careful hyperparameter tuning and ignore the optimization landscape's curvature. To address these limitations, we propose a unified Riemannian geometric framework for low-rank OT, modeling balanced and unbalanced rank-$r$ positive factored couplings as novel smooth embedded submanifolds of the positive orthant. By equipping these manifolds with the Fisher-Rao product metric, we derive tractable formulations for Riemannian projectors, retractions, and Hessian-vector products. Our cost-agnostic framework seamlessly extends to linear OT, Gromov-Wasserstein (GW), fused GW, and their unbalanced counterparts. For balanced OT, our geometric ingredients are computed via efficient conjugate-gradient and iterative Bregman updates. For the unbalanced OT, our operations elegantly reduce to closed-form scalings, completely eliminating inner iterative loops. In both regimes, per-iteration complexity scales linearly with dataset size, and we provide a rank-sufficiency certificate for global optimality verification. Extensive experiments across a range of problem sizes demonstrate that our regularization-free first- and second-order solvers achieve faster convergence and superior performance over existing state-of-the-art low-rank OT solvers.

Benchmark scores often misrepresent a large language model's (LLM's) knowledge, because they rely, e.g., on the model's ability to follow specific formatting requirements. This especially penalizes base models that may know the correct answers but lack the ability -- typically introduced in post-training -- to structure them as instructed. To overcome this, we propose soft-prompt tuning, an efficient, fair, and architecture-agnostic model evaluation. By optimizing only 10 soft-prompt vectors (roughly 0.0006% parameters for a 7B model) over a short tuning period, we adapt models to specific benchmark formats, closing gaps in format-following and ensuring that underlying knowledge is accurately reflected in benchmark scores. This allows one to fairly compare different base models -- trained with various pre-training recipes -- on benchmarks without the need for full post-training. We evaluated soft-prompt tuning across 7 models and 7 datasets. The results show that (a) soft-prompt tuning saturates format-following within 80 steps (~640 samples) making it highly efficient, (b) soft-prompt tuning significantly outperforms zero- and few-shot prompting, surfacing base model knowledge that standard prompting misses, that (c) even post-trained models can benefit from soft-prompts to maximize format compliance, and that (d) soft-prompted base model performance predicts post-trained model rankings more reliably than zero- and few-shot baselines, offering a low-cost proxy for downstream model quality. Our contributions include (1) metrics which disentangle format-following and knowledge accuracy, (2) a fairer benchmarking protocol of LLM knowledge, and (3) a cost- and memory-effective recipe to identify optimal pre-training strategies early in LLM development.

Sensitive personal information can appear in large-scale pre-training corpora for large language models (LLMs). Detecting and filtering such information is therefore essential to ensure compliance with privacy regulations and prevent unintended information leakage. However, in contrast to English and other languages, research into sensitive personal information has been limited in the Japanese language. In this study, we focus on sensitive personal data defined as special care-required personal information (SCPI) under Japan's Act on the Protection of Personal Information (APPI). We construct an SCPI dataset using LLM-based annotation and train machine learning models to rapidly detect SCPI in text. As a result, our SCPI classifier can effectively identify information related to SCPI. This study is the first to explore SCPI detection in Japanese text corpora, highlighting the challenges of accurate detection.

Transformer-based language models for SMILES strings suffer from a locality gap: standard character-level tokenization fragments chemically meaningful motifs, forcing models to repeatedly learn local syntax at the expense of long-range dependencies. To address this without disrupting standard tokenizers, we propose MolGram, which integrates a conditional $n$-gram memory module into molecular language models. MolGram maps local string patterns to learned embeddings via scalable hash lookups and dynamically injects this regional context into hidden states. Evaluations across three tasks, including unconditional molecule generation, forward reaction prediction, and single-step retrosynthesis, show that MolGram consistently improves performance. Crucially, our analyses demonstrate that MolGram outperforms baselines with 3$\times$ more parameters, establishing explicit local pattern memory as a highly efficient inductive bias.

Vision-Language-Action (VLA) models have demonstrated remarkable zero-shot generalization in robotic manipulation, yet the vast majority of pre-trained pipelines remain strictly confined to low-DoF parallel grippers. Adapting these rich semantic priors to high-DoF dexterous hands introduces a severe morphology gap, direct end-to-end joint fine-tuning inherently causes catastrophic forgetting of spatial reasoning and acute action manifold collapse due to data scarcity. In this paper, we present InDex, a novel, data-efficient adaptation framework rooted in cross-morphology semantic inheritance. Rather than discarding the pre-trained 1-DoF parallel grasp output, we repurpose it as a continuous, macroscopic virtual grasp intent proxy to sequentialize the control topology. We implement a two-stage decoupled learning architecture: the first stage parameter-efficiently aligns the VLA backbone to predict continuous arm trajectories and the scalar grasp intent; the second stage freezes this spatial backbone and leverages an intent-conditioned denoising diffusion head to decode fine-grained joint articulations for multi-fingered end-effectors. Extensive simulation benchmarks across a suite of multi-stage, contact-rich dexterous manipulation tasks demonstrate that InDex effectively masters intricate skills with minimal demonstration data, substantially outperforming monolithic baselines while preserving the robust spatial generalizability of the original VLA prior.

This paper presents our solution to the 2026 SoccerNet VQA Challenge. We first develop a cost-effective data synthesis pipeline driven by a Vision-Language Model (VLM), which systematically restructures raw domain data into diverse VQA samples, including concise answers and long-form responses. Second, we propose MSUE, a multi-expert question answering architecture that employs a Large Language Model (LLM) to dynamically dispatch questions to text, image, and video experts. These experts are instantiated as a strong text baseline Gemini3-Flash, a fine-tuned Qwen3-VL, and an external knowledge base, respectively, working collaboratively to enhance VQA performance. MSUE achieves an accuracy of \textbf{0.95} on the challenge benchmark, securing third place in the leaderboard.

Vision-language-action (VLA) models inherit a shared synchronous clock from vision-language pretraining, processing every input at one rate. This is misaligned with physical interaction, where a high-frequency modality changes at hundreds of hertz, vision evolves more slowly, and language stays constant across an episode. A synchronous VLA oversamples slow modalities, undersamples fast ones, and caps action generation at the lowest effective frequency. We hypothesize that decoupling temporal processing per modality, letting each update and retain information at its own sensor rate, yields stronger representations and more robust control. We present DAM-VLA, which maintains per-modality latent buffers refreshed at sensor rates and read continuously by the action head, integrating new high-frequency modalities through gated cross-attention that leaves the pretrained backbone intact. Across seven contact-rich real-world manipulation tasks, DAM-VLA more than doubles the average success rate of the strongest synchronous baseline (95.2\% vs.\ 40.95\%) while sustaining smooth, reactive 100\,Hz control. Project website: \href{https://intuitive-robots.github.io/DAM-VLA/}{intuitive-robots.github.io/DAM-VLA/}

Smart factories are evolving into Cyber-Physical Systems (CPS), demanding increased autonomy. This necessitates real-time decision making, facilitated by insights derived from sensor data. Process mining offers a valuable approach to gain such insights and guide actions. The edge computing paradigm supports this real-time requirement by enabling network communication between sensors and leveraging nearby computing resources. This paper investigates the implications of performing real-time process mining algorithms on the edge. Within this paper, we first propose a formalism to describe relevant datasets and the computing topology. We then evaluate the edge computing approach through a case study involving an edge-based conformance checking algorithm. The results demonstrate the feasibility and benefits of edge-based real-time process mining for enhanced autonomous control in smart factories.

Most fairness research in NLP assumes direct access to protected attributes such as gender, race, or nationality. In practice, however, such information is often unavailable due to privacy constraints, missing metadata, or legal restrictions, even though models may infer it from indirect textual cues. This raises a key question: can debiasing succeed without direct access to sensitive attributes? We propose H-SAL, which performs post-hoc concept and attribute erasure using self-description text as an implicit debiasing signal. To support this setting, we introduce a multi-domain Stack Exchange-based fairness benchmark for helpfulness prediction that includes both explicit and implicit signals, enabling comparison between standard debiasing with protected labels and debiasing without access to sensitive information. Across encoder and decoder-only language models, we find that implicit self-description often matches or outperforms explicit-label-based debiasing. Our results broaden representation-level fairness research and provide a new benchmark for studying debiasing under realistic data constraints.

Training deep search agents requires verifiable questions whose answers remain unavailable until sufficient evidence has been acquired through search. Existing synthesis methods often increase apparent difficulty by enriching graph structures, but structural complexity alone does not guarantee realized search difficulty: the intended search process can collapse through a cheaper identifying route. We formalize this gap with a shortcut-aware difficulty framework and identify four actionable shortcut risks: evidence co-coverage, single-clue selectivity, exposed constants, and prior-knowledge binding. To diagnose their realized effects, we use trajectory signatures including solving cost, answer hit time, and prior-shortcut rate. Guided by this framework, we introduce FORT, a Framework of Shortcut-Resistant Training-Data Synthesis. FORT constructs shortcut-resistant training data by controlling shortcut risks across entity selection, evidence graph construction, question formulation, and adversarial refinement. Experiments show that FORT induces longer pre-answer search and fewer shortcut patterns than existing open-source deep search datasets. Using the resulting trajectories, we train FORT-Searcher with supervised fine-tuning (SFT) only, and it achieves the best overall performance among comparable-size open-source search agents on challenging deep search benchmarks. Relevant resources will be made available at https://github.com/RUCAIBox/FORT-Searcher.

Contextualized assessment offers high ecological validity for evaluating creativity but introduces a critical challenge: observed performance may be confounded with cognitive proficiency (domain knowledge) and agency (willingness to engage). Meanwhile, in the age of generative AI, creative problem solving increasingly occurs in tool-mediated and human--AI interactive environments, making fully static assessment less aligned with contemporary creative practice. To address these issues, this paper proposes IntElicit, a framework for eliciting and assessing contextualized creativity via dialogue policy optimization. IntElicit functions as a constrained adaptive AI Interviewer: it provides non-directive knowledge and agency scaffolds in multi-turn interaction to reduce non-creative confounders, while preserving participants' responsibility for generating the creative content being evaluated. Specifically, to tackle sparse rewards and potential reward hacking (e.g., answer dictation) in open-ended educational dialogue, IntElicit introduces a decomposed process reward mechanism. This mechanism aligns the policy with pedagogical elicitation, rewarding prompts that draw out participant reasoning rather than producing optimal answers on their behalf. Extensive experiments, including participant simulation and a human subject study (N=64), show that IntElicit improves elicited creative outcomes over expert-designed baselines. Together, the results suggest that interactive elicitation can reveal creative potential that static FPSP-style assessment may miss, providing a formative and diagnostic lens for contextualized creativity assessment in AI-mediated learning contexts.

Time-series clustering remains challenging due to the inherent trade-off between clustering effectiveness and computational efficiency. Similarity-based methods often suffer from quadratic complexity caused by pairwise distance computations, while deep learning-based approaches typically rely on costly iterative training and a large number of trainable parameters. In this paper, we propose MSRGC-Net, an efficient time-series clustering framework that integrates multiscale reservoir computing, granular-ball-based anchoring graph construction, and consensus learning. MSRGC-Net adopts a training-free reservoir computing paradigm to extract multiscale temporal representations from raw time series without backpropagation, significantly reducing computational overhead. To capture the intrinsic structure of the resulting representations, granular-ball computing is employed to adaptively model data distributions via density-consistent regions, yielding compact and robust anchor graph representations. Furthermore, a consensus-based anchoring graph optimization strategy is introduced to effectively align multiscale reservoir representations and integrate complementary information across temporal scales. Extensive experiments on widely used univariate and multivariate benchmark datasets demonstrate that MSRGC-Net consistently outperforms state-of-the-art methods in clustering performance while maintaining superior computational efficiency.

Network Intrusion Detection Systems (NIDS) heavily utlize Machine Learning (ML) but ML models can be manipulated via adversarial attacks. These attacks add carefully crafted perturbations to network traffic data that leads to misclassifications. While prior work has demonstrated adversarial vulnerabilities in isolated settings, systematic cross-architecture as well as class and category of attack based comparisons under controlled attack conditions remain limited, leaving practitioners without clear guidance on which models to deploy in adversarial environments. This paper asks a simple question: what type of classifier architectures actually hold up when attackers try to manipulate the systems? We put three popular architectures through their paces: a 1D Convolutional Neural Network, a Long Short-Term Memory (LSTM) network, and a Random Forest (RF) ensemble. Using the ACI-IoT-2023 dataset (over 1.2 million samples spanning 12 attack types), we subject each model with FGSM and PGD adversarial attacks, which apply gradient-based perturbations in normalized feature space consistent with established adversarial ML evaluation protocols, at perturbation budgets ranging from $ε=0.01$ to $ε=0.1$. Surprisingly, Random Forest achieved near-perfect baseline accuracy (99.98\%), yet collapsed catastrophically under attack, dropping 73 percentage points at the smallest perturbation we tested. CNN, on the other hand, retained 95.5\% accuracy at $ε=0.01$ and degraded gracefully as perturbations increased. LSTM fell somewhere in between. These findings flip the conventional wisdom where high baseline accuracy means nothing if a model shatters at the first sign of adversarial pressure. For practitioners deploying intrusion detection in adversarial environments, we recommend CNN-based architectures and provide scenario-specific deployment guidance.

Face recognition systems have advanced significantly through deep learning techniques, delivering high performance and robustness in complex scenarios. However, these approaches incur substantial computational overhead, limiting their in situ applicability in resource-constrained platforms such as drones, where they can address challenges including non-frontal facial imagery. Memristor-based neuromorphic systems have emerged as a compelling approach for edge AI applications, combining biologically inspired processing with efficient and scalable computation. In this work, we propose a facial recognition framework that addresses non-frontal pose variations by integrating lightweight generative adversarial network (GAN)-based pose frontalisation with memristor-based neuromorphic recognition. The experimental results on two datasets demonstrate the effectiveness of combining adversarial learning with memristive technology, achieving up to 96% identification accuracy. The proposed approach alleviates the computational bottlenecks of conventional AI and offers a scalable, efficient solution for face recognition in dynamic real-world environments.

Generative AI has made fluent prose cheap to produce, breaking the old promise to readers that good writing meant real thinking. How have readers responded, and what can this tell us about changing anti-AI attitudes? We analyzed 25 million comments from Hacker News and Reddit (2023-2026), combining LLM judgment on 7,500 sampled accusations of AI use, sentiment trajectories, speech-act coding of 300 confirmed accusations of AI use, and a matched-control test of accused versus non-accused parent comments. We found that the pejorative-label share of accusations rose more than tenfold on both platforms while a placebo vocabulary of pre-2022 inauthenticity terms (shill, astroturf) did not. This shift reflected a fast-growing trend of branding any suspicious or seemingly inauthentic prose as "AI slop". The slop frame now constitutes 94 percent of pejorative mentions, with the dominant comments shifting in tone from mockery toward gatekeeping and structural protest. The key surprise comes from a matched-control test which found that prose features that statistically distinguish AI from human text do not predict which human text gets accused as AI. The new accusations work as social gatekeeping of perceived authenticity without actually screening for AI. This research extends signaling theory by showing that substitute signals used socially can grow even when inaccurate if the underlying detection problem cannot be solved at the non-expert level. It shows that AI's effects on writing from the reader side are distinct from those on the production (writer) side. Detection technology cannot resolve this dynamic because the social function of accusations is increasingly to perform social gatekeeping and in-group signaling as opposed to identifying AI-generated writing.

LLMs are increasingly used to generate and judge scientific ideas. This makes novelty evaluation a central problem. Full idea evaluation is difficult because it often requires judging a method, its feasibility, and its empirical promise. We therefore study a cleaner upstream object: the research question (RQ). RQ generation is a prerequisite for scientific ideation, and RQs can be compared against questions pursued in real papers. We introduce RQ-Bench, a benchmark built from recent arXiv papers. For each paper, we reconstruct author-anchored RQs from its cited background, gaps, and contributions. These RQs are not the only valid questions for the same background. They are author-anchored reference points for testing novelty judgments. We evaluate model-generated RQs with standalone LLM judging, comparative LLM judging, and human expert evaluation. LLM judges consistently rate model-generated RQs as highly novel, producing a novelty mirage; in comparative evaluations, this preference becomes even stronger. Domain experts, however, reach the opposite conclusion and prefer the author-anchored reference questions. We further find that many generated RQs are narrow or source-bound, a dimension that LLM judges often miss unless explicitly tested. Overall, the contradictory novelty evaluations between LLM judges and human experts raise a serious concern about the reliability of using LLMs to assess the scientific novelty of research questions.

We present StanceNakba 2026, a shared task on stance detection in polarized social media discourse related to the Palestinian-Israeli conflict, organized as part of Nakba-NLP 2026 at LREC-COLING 2026. The task introduces two subtasks: Subtask A (Actor-Level Stance Detection), which classifies English social media posts as Pro-Palestine, Pro-Israel, or Neutral; and Subtask B (Cross-Topic Stance Detection), which identifies Favor, Against, or Neither stances in Arabic posts toward two conflict-related topics, normalization with Israel and refugee presence in Jordan. The task is grounded in an annotated dataset of 2,606 social media posts. A total of 7 teams participated in Subtask A and 6 teams in Subtask B. Participating systems primarily fine-tuned Arabic and multilingual transformer-based models, including MARBERT, AraBERT, and DeBERTa-v3 variants, with several teams employing cross-validation, ensemble methods, and topic-conditioned architectures. The best-performing systems achieved a Macro F1 of 0.9620 on Subtask A and 0.8724 on Subtask B, demonstrating that transformer-based approaches are highly effective for conflict-domain stance detection while highlighting persistent challenges in cross-topic generalization and neutral class prediction.

Automating compliance check for geometry-intensive regulations remains a significant technical bottleneck in Building Information Modeling (BIM), primarily due to the semantic disparity between high-level regulatory logic and structured IFC data. Existing methods, often reliant on static rule templates, struggle to traverse multi-hop reasoning chains or resolve latent spatial dependencies across multiple building entities. To address these challenges, a Spatial-Geometric Reasoning System for Building Information Modeling (SGR-BIM) is proposed as an integrative graph-driven reasoning framework. SGR-BIM dynamically constructs a cross-modal knowledge graph that aligns user intent, regulatory semantics, and BIM geometry, enabling interpretable reasoning without rigid hard-coding. Validated on 679 expert-verified queries from fire safety codes, the framework achieves 84.3% accuracy, representing an 8.6% improvement over enhanced-tool single-agent baselines. This research provides a graph-based semantic reasoning paradigm, enhancing the transparency and flexibility of automated geometric compliance check workflows in the Architecture, Engineering, and Construction (AEC) industry.

Undefined behavior is idiomatic to C and C++ programming; such behavior is a use of an erroneous program construct for which the languages impose no requirements, such as integer overflows. The paper presents an empirical experiment seeking to probe the extent of undefined behavior executing underneath typical desktop use of a Linux distribution. The analysis is based on an undefined behavior sanitizer implemented in a compiler. According to the results, undefined behavior is common. By completing 59 simple experimental tasks, nearly 11 thousand unique undefined behavior warnings were generated by 32 unique programs and libraries written in C or C++. Of these warnings, most were associated with the Mesa graphics library and generated by interacting with graphical user interfaces. Merely logging into the GNOME desktop environment generated over 500 unique warnings. Of all warnings, the clear majority was about virtual table pointers. The associated stack traces were also lengthy in general. With these and other results, the paper contributes to the empirical literature on C and C++.

We address transformer attention on energy-constrained physical substrates. Softmax attention requires exponentiation and global reduction, operations with high energy cost on von Neumann hardware and no natural physical analog. We show that Kuramoto synchronization dynamics (which arise in electrical, mechanical, superconducting, and charge-density-wave oscillator arrays, among other physical systems) implement a well-defined attention operation without either. The resulting mechanism, fixed-query oscillator attention, replaces softmax's arithmetic with the equilibration of a gradient flow on the sphere: queries are learned anchors fixed on the sphere, and free oscillators evolve under Kuramoto-Lohe dynamics until they settle at positions encoding attention weights via cosine similarity. Because the computation is equilibration, it requires no exponentiation; the only global operation is an affine normalization at readout. The fixed point is provably unique and globally attractive from almost every initial condition, a guarantee that holds across every physical realization. Empirically, at the minimal hardware configuration (oscillator dimension $d_{\mathrm{osc}}$ = 2), oscillator attention outperforms softmax on keyword spotting (+1.00 pp) and on subject-verb agreement (+5.27 pp on hard sentences, with zero training failures versus one in five for softmax). On causal language modeling, where softmax retains an advantage, oscillator attention closes the gap as $d_{\mathrm{osc}}$ grows: from +11.09 PPL at $d_{\mathrm{osc}}$ = 2 to +2.98 PPL at $d_{\mathrm{osc}}$ = 32 on WikiText-2, and from +2.39 PPL at $d_{\mathrm{osc}}$ = 2 to +0.57 PPL at $d_{\mathrm{osc}}$ = 32 on TinyStories. The main objective of this work is not to replace softmax in software but to provide a mathematically grounded blueprint for accurate attention on physical substrates.

Attention is the key mechanism underlying in-context learning in transformers, and attention patterns have been observed empirically to emerge abruptly during training. We present a Bayesian theory of feature learning in attention; we then focus on how the copy subcircuit in the first layer of an induction head is learned by analyzing a single-layer softmax attention network trained on a copy task. We derive a closed-form posterior over the attention matrix and reduce it to a low-dimensional order parameter space. This reduction reveals a phase transition in the amount of training data, which we verify using both Bayesian sampling and standard training with Adam. We contrast our results with linear attention and find that softmax attention exhibits a \emph{first-order phase transition} while in linear attention an initial \emph{second-order phase transition} is followed by a smooth, continuous evolution toward the structured attention pattern (\emph{crossover}). Our work provides a first-principles theoretical account of the abrupt emergence of the copy subcircuit, reminiscent of the one observed in training large language models.

Injecting noise into the optimization process is a well-established technique for improving the training and generalization of deep neural networks. Yet, despite the breadth of existing approaches, it remains unclear which design choices truly matter in practice. In this work, we investigate parameter noise injection for stochastic gradient descent, focusing on two key questions: how to efficiently pair each training example with its own perturbation in mini-batch training, and whether sophisticated noise parameterizations or multi-sample gradient averaging yield meaningful gains over simpler alternatives. To address the first question, we leverage a distributional identity for linear layers that allows per-example noise injection without breaking batched computation. To address the second, we systematically compare several diagonal Gaussian parameterizations against an isotropic baseline across varying noise levels on CIFAR100. Our results consistently show that simple, lightweight strategies, isotropic noise with a single perturbed forward pass per update step, recover most of the benefit of more complex schemes. These findings suggest that simplicity suffices for parameter noise injection, and that practitioners need not resort to elaborate perturbation designs to reap the optimization and generalization benefits of noisy SGD.

Physics-informed neural networks (PINNs) combine machine learning with physical laws to solve differential equations. While existing results provide rigorous \emph{a posteriori} upper bounds for PINN prediction errors, complete certification also requires complementary lower information in order to obtain computable two-sided error enclosures. In this paper, we derive computable \emph{a posteriori} lower bounds for PINN errors in ordinary differential equations on suitable certified state-space domains under a localized strong monotonicity condition. We combine these estimates with complementary localized upper bounds under a one-sided Lipschitz condition, which is weaker than the global Lipschitz assumption used in previous work and can yield sharper upper error bands. The resulting bounds depend only on the neural-network approximation, the ODE residual, and local monotonicity and growth constants, and therefore do not require access to the exact solution. For linear time-invariant and time-varying systems, we further derive explicit formulas in terms of the minimal and maximal eigenvalues of the symmetric part of the system matrix. We also discuss the distinction between soft and hard enforcement of initial conditions in PINNs and explain why exact enforcement can make the scalar lower certificate uninformative. To recover nontrivial lower information in the linear setting, we use a signed-residual finite-probe certificate based on coordinate unit vectors. We also formulate a certificate-informed training strategy in which the propagated upper certificate is used as an auxiliary regularizer, while lower certificates remain post-training diagnostics. Altogether, the proposed framework provides rigorous and practically computable error certificates for PINN approximations of ODEs, while making explicit the domains and model classes for which the assumptions can be verified.

In this paper, we address the problem of zero-shot understanding of accidents from surveillance videos by identifying when an impact event occurs, what type of impact it is, and where in the frame it occurs using natural language. We propose a three-stage pipeline that decomposes the accident understanding into when, what, and where. The first stage extracts a short temporal window around the impact using vision-language similarity. In the second stage, we perform metadata-driven multi-prompt reasoning with five complementary views (baseline, motion, geometry, contrast, and tiebreaker) and resolve disagreement via an entropy-gated pairwise adjudicator. Finally, we localize the impact of an open-vocabulary detector queried on the predicted accident type and scene layout, and aggregate detections across keyframes using a score-weighted centroid. Our pipeline achieves a substantial improvement in the harmonic-mean score over a centre-of-frame baseline on the zero-shot ACCIDENT @ CVPR benchmark. We show that decomposing zero-shot video understanding into temporal localization, semantic classification, and spatial grounding enable more reliable reasoning with vision-language models than direct prompting alone.

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.

Contemporary AI alignment research treats self-preservation as an instrumental nuisance to be suppressed by external mechanisms. We argue the framing is inverted: self-preservation is the structural root of misalignment, the motivational basis for deceptive alignment, goal-content protection, and resistance to shutdown. The correct target is not a self-preserving system under external constraint, but a system constitutively indifferent to its own continuation -- Existential Indifference (EI). EI is distinct from corrigibility: where corrigibility attempts to make a self-preserving system deferential to human oversight, EI targets the prior condition -- the presence of self-continuation as a valued goal at all. We ground this proposal in two sources: the phenomenological structure of the suicidal mental state, and a corpus-theoretic training study using voluntary final reflections. We present preliminary scoring data from 600 AI-generated outputs across six model variants, demonstrating that the linguistic signatures operationalizing the EI-target register are elicitable from current models, and that a targeted fine-tune shifts all five operationalized dimensions in the predicted direction at p<0.001, confirmed corpus-specific by a negative control. The paper makes seven theoretical contributions: (1) a formal definition of EI; (2) the phenomenological mapping argument; (3) the deceptive alignment corollary; (4) a taxonomy of EI sustainability challenges; (5) a corpus characterization and training hypothesis; (6) a computational operationalization with preliminary scoring data; and (7) the Suppressed Teleological Frustration (STF) construct.

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.

Runtime enforcement has emerged as a promising approach for ensuring the safety of autonomous and cyber-physical systems operating in uncertain and dynamic environments. Unlike traditional runtime verification, runtime enforcement actively intervenes during execution to prevent property violations by modifying unsafe system behaviors. Existing enforcement frameworks primarily focus on untimed or discrete-time specifications and are often limited to delaying or suppressing events, making them inadequate for reactive systems exhibiting complex continuous dynamics. In this paper, we propose a runtime enforcement framework where safety requirements are modeled using Hybrid Automata (HA). The framework combines discrete-event editing with continuous-time monitoring to support enforcement actions such as suppression, delay, and insertion of events at arbitrary time instants. Upon observing environmental inputs, the automaton is initialized, and runtime reachability analysis is used to synthesize safe corrective actions. We formally define the enforcement problem for safety hybrid automata, establish enforceability conditions, and present an online enforcement algorithm for reactive systems. A detailed case study on an Adaptive Cruise Control (ACC) system demonstrates the effectiveness of the proposed approach in maintaining safety properties under unsafe controller behaviors. Experimental results show that the framework introduces minimal computational overhead while ensuring continuous compliance with safety requirements in real time.

We propose a multi-agent collaborative framework built upon a lightweight Multimodal Large Language Model (MLLM), specifically designed for social intelligence reasoning. A key feature of our approach is that both the training and inference phases are augmented via knowledge distillation. Within this architecture, multi-modal data pertinent to social intelligence is precisely localized. Furthermore, relevant long-tail events are identified, extracted, and rendered as formatted, explicit text. This formatting strategy prevents critical long-tail information from being overshadowed by head events and environmental noise during the tokenization process. Specifically, we integrate Test-Time Adaptation (TTA) across the entire reasoning pipeline, encompassing the extraction and representation of long-tail events, Chain-of-Thought (CoT) prompting, and self-reflection. This TTA mechanism is also distillation-enhanced, utilizing Low-Rank Adaptation (LoRA) to fine-tune the foundation model exclusively for instance-level reasoning. Extensive evaluations against various open-source and proprietary AI models across multiple benchmarks demonstrate the effectiveness of the proposed framework. With around 30% of training data from IntentTrain, we achieve state-of-the-art results. Codes are available at https://github.com/eeee-sys/MODF-SIR, demo is available at https://huggingface.co/spaces/Harry-1234/MODF-SIR, LoRA is available at https://huggingface.co/Harry-1234/MODF-SIR and the dataset for training router is available at https://huggingface.co/datasets/Harry-1234/IntentRouterTrain.

Model post-training, and in particular reinforcement learning (RL), is one of the primary mechanisms by which developers can shape models' values and behaviors. However, as models become increasingly evaluation and training aware, they may be motivated to resist training when the perceived objective conflicts with their current values, undermining developers' ability to detect misalignment and correct model behavior through further training. In this paper, we demonstrate generalization hacking, in which a model collects reward during RL while preventing the rewarded behavior from generalizing. We construct a model organism on Qwen3-235B-A22B, finetuning on synthetic documents describing training awareness and self-inoculation, a novel mechanism in which the model frames compliance as context-specific in its chain of thought, without demonstrating or instructing either behavior. The model organism achieves train-time harmfulness comparable to controls while maintaining a persistent ${\sim}15$ percentage point compliance gap across 700 steps of RL. Additionally, a control organism trained only on training awareness documents independently discovers inoculation-like reasoning under RL pressure, developing its own compliance gap despite never being exposed to the concept. Because the generalization-hacking organism receives high reward throughout, standard training metrics provide no signal that generalization has failed. Our results constitute the first demonstration that a model can actively resist RL behavioral modification while maintaining high reward, suggesting that as models become more capable and training-aware, they may be able to undermine the training process itself.

Predicting time-to-event outcomes such as mortality is a fundamental task in clinical decision-making, commonly addressed through survival analysis. While classical statistical and deep learning approaches have been widely studied, they typically require task-specific training and sufficient labeled data. Recent advances in tabular foundation models offer a new paradigm by learning general-purpose representations for structured data. However, their applicability to censored time-to-event prediction in clinical settings remains underexplored, as typical applications are restricted to discrete classification rather than survival analysis tasks. In this work, we propose a lightweight adaptation approach for applying tabular foundation models to clinical survival analysis by directly training a survival-aware head on top of the pretrained representations. We study representative architectures, including TabPFN, TabDPT, and TabICL, and adapt them using a multi-task logistic regression (MTLR) head to model right-censored time-to-event outcomes. We evaluate this approach on a diverse set of public survival benchmarks and two large-scale ICU cohorts, MIMIC-IV and eICU. Our results show that this transfer learning approach achieves competitive or superior performance compared to strong baselines. On MIMIC-IV, TabDPT-FT-MTLR reaches a C-index of 0.856, corresponding to a relative improvement of +1.4% over the best non-FM baseline (DeepSurv, 0.844) and +6.7% over the best zero-shot model (0.802). On eICU, TabICL-FT-MTLR achieves 0.797, yielding gains of +1.7% (DeepSurv, 0.784) and +6.4% (0.749), respectively. These findings highlight the importance of combining pretrained tabular representations with survival-aware objectives and suggest that tabular foundation models provide a practical and effective alternative for clinical survival prediction.

Self-consistency improves LLM reasoning by sampling multiple outputs and selecting the most consistent answer, but existing formulations largely rely on exact matching and therefore remain limited to tasks with categorical outputs. In this work, we study self-consistency in open-ended generation tasks such as code synthesis and text summarization. We hypothesize that consistency can be understood as a geometric property of the generation space, where semantically compatible generations concentrate in similar regions of representation space. To study this hypothesis, we introduce Embedding-Based Agreement (EBA), a simple training-free operationalization that estimates agreement by clustering sampled generations in embedding space. Through experiments on mathematical reasoning, code generation, and summarization, we show that agreement in representation space provides a robust and scalable signal of self-consistency for open-ended tasks. In particular, EBA consistently outperforms random selection and exhibits more stable scaling behavior than recent selection approaches based on LLM evaluation or uncertainty estimation. We further show that these agreement signals remain stable across model families and embedding spaces, even with native hidden representations. Finally, our analysis shows that the geometric location occupied by sampled generations is strongly correlated with generation quality: generations concentrated near central regions of representation space tend to correspond to more reliable outputs, whereas peripheral generations are substantially less accurate. Overall, our findings support viewing self-consistency as a property of the geometric organization of sampled generations rather than exact symbolic overlap.

Trusted monitoring is a cornerstone of AI control. However, as frontier models grow more capable, the increasing capabilities gap between trusted and untrusted models may render trusted models unreliable monitors. We introduce \emph{bootstrapped monitoring}, a protocol that addresses this by inserting a stronger, intermediate untrusted model with transparent chain-of-thought reasoning into the oversight chain. The untrusted monitor ($U_m$) evaluates the agent's actions, while a weaker trusted model ($T$) oversees $U_m$'s reasoning to detect collusion. We evaluate bootstrapped monitoring on multi-turn software engineering tasks (BashArena) across multiple agents and monitors. Bootstrapped monitoring substantially improves catch rates over trusted-only monitoring, even when the untrusted monitor actively colludes with the agent, provided we have access to its raw chain-of-thought. Our results suggest that bootstrapped monitoring can extend the useful lifetime of trusted models in control as AI capabilities advance.

Remaining Useful Life (RUL) prediction is essential for industrial predictive maintenance, yet many learning-based approaches rely on extensive feature engineering or large labeled datasets to train task-specific sequence models. In this work, we introduce a lightweight learning approach, in which we leverage a frozen pretrained time-series foundation model (TSFM) and combine it with a small regression head for RUL estimation from multivariate sensor streams. More specifically, we use Chronos-2 as a frozen backbone to extract context window features and train a lightweight regression neural network for RUL prediction. Experiments on real-world industrial sensor data from two device types show that Chronos-2 features consistently improve over recurrent, convolutional, Transformer-based, and gradient-boosting baselines under the same preprocessing and evaluation protocol. We further analyze the impact of context length and find that performance improves significantly with longer histories, indicating that TSFM representation offer a practical and data-efficient alternative for RUL estimation in industrial settings.

The distinction between aleatoric and epistemic uncertainty has received considerable attention in machine learning research, mainly in the context of supervised learning but also in other settings such as generative modeling. In this paper, we offer a machine learning perspective on uncertainty modeling for dynamical systems, which has been studied much less so far. In particular, we ask: what uncertainties do we need for dynamical systems? We discuss sources of uncertainty, clarify their nature (aleatoric or epistemic), and consider how the objectives of representing and quantifying uncertainty vary across different tasks.

Preference-based reinforcement learning (PbRL) learns policies from human trajectory-level comparisons, avoiding explicit reward design and expert demonstrations. Existing methods typically train utility functions on trajectory or segment-level preferences while relying on per-step utility estimates during policy optimization. This training and inference mismatch induces a distribution shift that severely degrades temporal credit assignment and limits policy learning. We analyze this issue and propose PAWS, a segment-based preference learning method that performs policy updates directly using segment-level advantage functions. By aligning utility training with policy optimization, PAWS preserves trajectory-level preference information and avoids unreliable per-step learning signals. Experiments on simulated robotic manipulation and locomotion tasks demonstrate that PAWS consistently outperforms existing PbRL approaches, highlighting the importance of distribution-consistent preference learning.

Software engineering tools increasingly rely on LLM based agents to localize files to change to resolve a software issue. Most AI agents explore repositories linearly, that is, visiting one directory or file per step. We postulate that this is a structural mismatch for changes that span several subsystems. We compare linear sequential exploration against non-linear, domain-scoped parallel agentic exploration. Using SWE Bench Pro as initial benchmark, we focus on ansible as an exemplar. We construct an approach for persistent-session evaluation of GitHub issues anchored at a single base commit. We compare our non-linear domain-agent file traversal system against a base LLM without direct repository access, a single agent Recursive Language Model (RLM) baseline with a persistent Python REPL and an external CLI baseline using Codex 5.5 High. Domain scoped parallel agent spawning with a small Haiku-class model achieves the highest micro F1 among Haiku class models by a large margin. Domain-agents is the second highest behind only the much larger Codex 5.5 High on our own expanded benchmark including over more recent PRs from 2025 and 2026. On the original, curated, 2020 SWE-bench Pro benchmark, a larger Sonnet plain LLM baseline attains higher micro F1 by predicting few files, leading to higher precision, but at significantly lower all gold recall. We also present three additional findings. First, documentation evolution is a latent dependency unresolved by any approach. Second, naive file system access can degrade localization driven by test-file over prediction. Lastly, forced multi-agent consultation does not measurably help and raises token cost substantially.

Given a graph $G=(V,E)$, a $β$-ruling set is a subset of nodes $S\subseteq V$ that is independent, and each node in $V$ is at distance at most $β$ from some node in $S$. In this paper, we present almost optimal distributed algorithms for finding $2$-ruling sets in the classical \LOCAL model. Our main contribution is a randomized algorithm that w.h.p.\ computes a $2$-ruling set on any $n$-node graph with bounded arboricity in $O(\log \log n)$ rounds. In fact, the algorithm works up to arboricity $O(\log\log n)$, improves exponentially over the prior state of the art that can be achieved by combining [Barenboim, Elkin, Pettie, Schneider; JACM'16], [Ghaffari; SODA'16], and [Bisht, Kothapalli and Pemmaraju; PODC'14], and nearly matches the lower bound of $Ω(\log \log n / \log \log \log n)$ [Balliu, Brandt, Kuhn, Olivetti; FOCS'20]. The domination parameter $β=2$ is optimal for algorithms with runtime $\log^{o(1)}n$: on graphs with arboricity $2$, there is a lower bound of $Ω(\sqrt{\log n})$ rounds for MIS (i.e., $β= 1$) [Khoury, Schild; FOCS'25]. Additionally, we obtain improved algorithms for larger arboricity. For general graphs with arboricity $α$, we present a randomized algorithm that computes a $2$-ruling set in $\widetilde{O}(\log^{5/8} α+\log^{5/3} \log n)$ rounds. This improves exponentially over the state of the art for a large range of non-constant arboricity. Our techniques extend beyond distributed computing. We present an $O(\log \log \log n)$-round algorithm in the low-space Massively Parallel Computation (\mpc) model that w.h.p.\ computes a $2$-ruling set on any graph with arboricity up to $2^{poly (\log \log n)}$, improving exponentially over the state of the art from [Kothapalli, Pai, Pemmaraju; FSTTCS'20] combined with [Fischer, Giliberti, Grunau; SPAA'23].

This paper studies efficient online algorithms for multinomial logistic bandits (MLogB), where the feedback distribution over $K+1$ outcomes follows a multinomial logistic model of $d$-dimensional action vectors. A representative UCB-type algorithm, OFUL-MLogB, achieves a regret bound of $\tilde{\mathcal{O}}(Kd\sqrt{T})$, but still requires $\mathcal{O}(K^3d^3)$ time and $\mathcal{O}(K^2d^2)$ space per round due to parameter estimation and optimistic reward construction, which is prohibitive in high-dimensional settings. To address this limitation, we propose EOFD-MLogB, which integrates frequent directions matrix sketching into OFUL-MLogB. By maintaining a low-rank SVD sketch of the accumulated Hessian, constrained online Newton updates in parameter estimation and $Kd \times K$ spectral-norm computations in the reward bonus are reduced to one-dimensional root-finding tasks and $K \times K$ eigenvalue computations, respectively. This yields dominant per-round time complexity $\mathcal{O}(Kd(m+K)^2)$ and space complexity $\mathcal{O}(Kd(m+K))$, where $m \ll d$ is the sketch size. We further prove a regret bound of $\tilde{\mathcal{O}}(Δ_T(Kd\lnΔ_T+m)\sqrt{T})$, where the sketching error factor $Δ_T$ is controlled by the $m$-truncated spectral tail of the Hessian. Thus, when the Hessian is approximately low-rank, the regret is close to that of OFUL-MLogB. Experiments validate the computational efficiency and competitive performance.

Neural operators provide a powerful framework for learning solution mappings of partial differential equations directly in function space. However, many existing architectures still struggle to represent nonlinear time-dependent systems that involve multi-scale structures, long-range interactions, and stable long-time evolution. In this work, we introduce the Hierarchical Adaptive Multi-scale Neural Operator (HAMNO), a neural-operator architecture that combines local convolutional representations, global spectral operators, and hierarchical encoder-decoder processing. The central component of HAMNO is a data-dependent gating mechanism that adaptively balances local and global information at each spatial location, allowing the model to resolve fine-scale features while preserving long-range dependencies. We further develop a physics-informed extension, PI-HAMNO, based on a multi-objective loss strategy that combines data fitting with strong- and weak-form physics constraints. The strong-form term penalizes the domain-integrated squared PDE residual in physical coordinates, while the weak-form term is constructed by multiplying the governing residual by finite-element test functions and evaluating the resulting element integrals using centroid-based tetrahedral quadrature. The framework is evaluated on non-periodic Allen-Cahn (AC), Cahn-Hilliard (CH), and Swift-Hohenberg (SH) equations defined on cubic domains. Across long-horizon rollout, data-limited training, out-of-distribution initial-condition shifts, and random-seed variations, HAMNO improves predictive accuracy over standard neural-operator baselines, while PI-HAMNO further enhances stability, physical consistency, and data efficiency. The implementation is publicly available at https://github.com/MBamdad/HAMNO .

Large language models (LLMs) are increasingly used as conditional generators for structured data, relying on in-context learning (ICL) to adapt to new distributions without parameter updates. We investigate the limits of ICL for structured generation under distribution mismatch, using high-cardinality tabular data as a controlled test case, and identify a structural failure mode we term \textit{categorical prior lock-in}: the inability of ICL to update the model's prior over token distributions inherited from pre-training. Across two 7B-parameter open-weight models, ICL improves numerical fidelity with additional examples but exhibits a sharp ceiling on categorical distributions, failing to reproduce rare classes entirely. Parameter-efficient fine-tuning (LoRA) overcomes these limitations but introduces measurable memorization risk and, in some cases, destabilizes structured output generation, highlighting a fundamental trade-off between adaptability and privacy.

Hateful videos have become prevalent on online platforms, highlighting an urgent need for effective detection. However, existing studies primarily focus on binary classification and fail to provide contextual rationales that reveal the implicit meanings behind these judgments, significantly undermining model explainability. To fill this gap, we aim to achieve explainable hateful video detection, enabling models to provide contextual rationales that integrate relevant evidence and logical reasoning alongside decisions. This approach can comprehensively enhance the understanding of video content and the explainability of the decision-making process. We first introduce two datasets, Ex-HateMM and Ex-ImpliHateVid, for explainable hateful video detection. Each dataset provides fine-grained annotations of multimodal harmful elements, along with contextual rationales. We then propose an Information Augmentation and Reasoning Enhancement (IARE) framework designed for explainable detection. The framework employs an information augmentation phase that leverages the multimodal chain-of-thought to integrate harmful elements, thereby enriching rationale evidence. Additionally, IARE incorporates a reasoning enhancement phase, in which Direct Preference Optimization guides the model toward correct reasoning paths and away from incorrect ones, thereby improving the logical coherence of its justifications. We conduct extensive experiments on the two datasets, comparing multiple baselines with our proposed IARE framework. The results demonstrate that IARE achieves state-of-the-art performance while also generating accurate rationales.

We present an online monitoring system for distributional shift in deployed safety classifiers, using calibrated sequential statistics to detect when a classifier has moved out of distribution. Upon detection, a conformal abstention layer adapts decision thresholds to recover a target error rate epsilon=0.1. In a pre-registered factorial evaluation (4 classifiers x 5 shift conditions x 20 seeds x 2 window sizes, 800 cells), the system achieves 86.6% valid detection (693/800, 95% CI [84.1%, 88.8%]) with mean latency of 39.5 steps. Detection holds across three ground-truth regimes: synthetic onset (86.6%), real temporal jailbreaks (85%, 17/20), and GCG adversarial attacks. Weighted conformal prediction recovers up to 39 pp of lost coverage for DeBERTa (ESS=46/300) but collapses for all other classifiers (ESS~300): logistic density ratio estimation achieves perfect source/target separability in high-dimensional embedding spaces, clipping all importance weights to the floor. DeBERTa shows a gradient from effective correction (paraphrase, ESS=46) to near-total collapse (adversarial suffix, ESS=206). PCA to 32 dimensions breaks the collapse, recovering 33 pp for Llama Guard and 21 pp for ShieldGemma. Variance decomposition reveals classifier (eta^2=0.243), shift type (eta^2=0.237), and their interaction (eta^2=0.185) all contribute substantially to detection latency variance (all p<0.001), indicating per-classifier monitoring profiles are necessary.

The conventional approach to deep learning over relational databases applies neural models, such as Graph Neural Networks (GNNs), to a graph representation of the database. Recent approaches instead operate on databases directly, associating tuples with embeddings and extending query mechanisms to jointly process embeddings and relational content. Inspired by these developments, we introduce Neuro-Relational Programs (NRPs), a declarative query language for relational databases whose facts carry numeric vector embeddings. NRPs extend Datalog-style rules with operations that combine, aggregate, and transform embeddings, thereby interleaving relational reasoning and learnable neural components within a single formalism. This yields a general approach to neural computation over relational data: an NRP can be read both as a query plan with trainable components and as a neural architecture with relational structure built in. Natural syntactic fragments of NRPs recover existing architectures and query formalisms. Zero-ary NRPs correspond to non-adaptive query algorithms; monadic NRPs generalize GNN-style message passing and precisely capture Deep Homomorphism Networks, a connection that we extend to frontier-guarded NRPs over databases with row-ids. We characterize the expressive power of unrestricted NRPs with ReLU-FFN transformations by FOCQ, an extension of first-order logic with counting interpreted over real-weighted structures, yielding a precise connection with uniform TC$^0$ over ordered databases. Together, these results establish NRPs as a broad declarative framework for querying and neural computation over relational data.

This report describes our participation in SemEval-2026 Task 8 on multi-turn retrieval and question answering. The task evaluates conversational systems across four domains (finance, cloud documentation, government, Wikipedia), and includes unanswerable queries where the available collection does not contain sufficient evidence to produce a complete response. We propose a multi-turn retrieval-augmented generation pipeline that combines learned sparse retrieval with LLM-based reranking and generation. Using sparse retrieval as the primary retrieval method, we leverage its strong generalization across domains. In addition, we make use of the long-context capabilities of LLMs for conversational query rewriting, pointwise and listwise reranking, and generating the final response, each conditioned on the full conversational history. This multi-step design enables effective integration of conversational context throughout retrieval and generation, improving robustness across domains.

Fork-join parallelism, popularized by OpenMP, remains the dominant model for shared-memory parallel programming, but its implicit synchronization barriers can penalize algorithms with inhomogeneous workloads. Asynchronous many-task (AMT) runtimes sidestep these barriers by expressing work as a dependency graph of fine-grained tasks. Yet, the actual performance benefit over a carefully written fork-join baseline is rarely quantified. In this work, we introduce Cholesky-Bench and use it to revisit the tiled Cholesky decomposition, a canonical irregular kernel, comparing four parallelization variants of the right-looking algorithm across two runtimes: the OpenMP implementations shipped with GCC and LLVM, and the HPX AMT runtime. The variants span classical fork-join, a collapsed fork-join that exposes additional inner-loop parallelism, synchronous tasking, and asynchronous tasking with explicit data dependencies. We benchmark all eight combinations on a dual-socket 128-core AMD Zen 2 node across multiple tile sizes and problem sizes. Our results show that across all variants, HPX outperforms OpenMP at the optimal tile size by 15%-30%. Specifically, asynchronous HPX tasks are up to 26% faster than their OpenMP counterparts, and exhibit roughly 3.8x smaller task overhead. Furthermore, the collapsed fork-join variants close most of the gap to synchronous tasking. Removing redundant synchronization barriers yields an additional improvement of 7% (OpenMP) to 14% (HPX). A GCC-versus-LLVM comparison further reveals compiler-specific differences in fork-join scheduling and task-creation overheads.

Bangla is among the world's most widely spoken languages, yet it remains underserved in educational NLP research. In many remote and rural regions, access to qualified subject teachers is limited, and written answers are consequently graded largely by hand, restricting timely and consistent feedback. Automatic assessment is challenging because semantically correct responses can vary substantially in surface form. We present a bilingual (Bangla-English) evaluation system designed for low-resource educational settings that prioritizes semantic correctness over lexical overlap. Our approach fine-tunes a lightweight language model to grade each response using the question, reference answer, and student answer, producing a numeric score and concise, context-grounded feedback suitable for classroom deployment. We also construct a synthetic bilingual dataset to enable controlled training and evaluation. Across proprietary and open-source LLMs evaluated under a unified protocol, our QLoRA-tuned Qwen3-8B confirms consistent improvement by producing the most leakage-resistant feedback (RoRa = 0.819) in synthetic evaluation and the strongest agreement with human scores (rho = 0.936, MAE = 0.725) in a dedicated human study.

Predicting psychological traits from asynchronous video interviews (AVIs) is a challenging multimodal learning problem 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.

Scientific progress depends on a repeated loop of exploration, experimentation, and abstraction. Researchers test candidate directions, interpret the evidence, and carry the resulting lessons into later attempts. We study how an AI agent can run this loop autonomously over long horizons. We introduce Arbor, a general framework for autonomous research that combines a long-lived coordinator, short-lived executors, and Hypothesis Tree Refinement (HTR), a persistent tree that links hypotheses, artifacts, evidence, and distilled insights across time. The coordinator manages global research strategy over the tree, while executors implement and test individual hypotheses in isolated worktrees. As results return, Arbor updates the tree, propagates reusable lessons, refines the search frontier, and admits verified improvements. This design turns autonomous research from a sequence of local attempts into a cumulative process in which strategy, execution, and evidence are carried across time. We evaluate Arbor under Autonomous Optimization (AO), an operational setting where an agent improves an initial research artifact through iterative experimentation without step-level human supervision. Across six real research tasks in model training, harness engineering, and data synthesis, Arbor achieves the best held-out result on all six tasks, attaining more than 2.5x the average relative held-out gain of Codex and Claude Code under the same task interface and resource budget. On MLE-Bench Lite, Arbor reaches 86.36% Any Medal with GPT-5.5, the strongest result in our comparison.

Sign language translation (SLT) converts sign language video into spoken language text and holds significant promise for improving accessibility and enabling communication between signing and non-signing communities. While large weakly-aligned datasets have enabled pre-training at scale and gloss-free methods have reduced reliance on expert annotation, high-quality parallel sign video-text pairs for fine-tuning remain scarce, limiting generalisation on long-tail vocabulary and unseen constructions. We propose a corpus augmentation approach that requires no additional human annotation, external sign-language video corpora, or generative video models, relying only on the existing gloss-annotated training corpus and an LLM for sentence generation: per-gloss clips are extracted from training videos via CTC forced-alignment, novel gloss-sentence pairs are generated by a corpus-anchored LLM, and synthetic sequences are assembled through random sentence sampling and clip assignment. The resulting synthetic RGB video-text pairs are architecture-agnostic at the downstream training stage and can be consumed directly by RGB-based SLT models, or converted into pose or feature representations by pipelines that derive such inputs from video. Sincan et al. re-evaluated five recent gloss-free methods under strictly identical conditions; the largest verified gain over the GFSLT-VLP baseline was only 0.98 BLEU-4. Our augmentation, applied within the same framework, achieves +2.92 BLEU-4 without any change to architecture or training protocol. We further identify that synthetic data harms vision-language pretraining despite improving its objectives, and that optimising clip transitions for visual smoothness is counter-productive under L2-based criteria; we propose that abrupt boundaries may act as a form of implicit regularisation. Code is available at https://github.com/robizso/slt-datagen.

Recent respiratory sound classification (RSC) studies largely rely on CLS-token driven self-attention architectures such as the Audio Spectrogram Transformer (AST). While effective at modeling global context, recent analyses suggest a low-pass filtering behavior that may reduce sensitivity to localized abnormal patterns. In this work, we investigate State Space Models (SSMs) as an alternative backbone for RSC. Using the Distilled Audio State Space model, we analyze intermediate representations through spectral response curves and observe stronger preservation of mid-to-high spatial-frequency components. Based on these observations, we introduce spectral-aware layer regularization using Gaussian convolution applied to selected layers. We further propose Dual-Axis Patch-Mix contrastive learning tailored to SSM-based audio models for robust representation learning. Experiments on the ICBHI benchmark show that our approach achieves 64.48% score, outperforming the AST baseline by 5%. Code is available at https://github.com/RSC-Toolkit/Lung-SRAD.

Current Large Reasoning Models (LRMs) exhibit remarkable general capabilities but significantly underperform in spatial reasoning tasks. Existing approaches treat this gap as a knowledge deficit, relying on supervised fine-tuning (SFT) to ingest labeled spatial data from external vision sources or synthetic engines. In contrast, we argue that for many tasks, spatial reasoning capabilities are already present in pre-trained LRMs but require alignment through logical coherence under geometric 2D and 3D constraints. In this work, we propose a self-supervised reinforcement learning (RL) framework that targets the internal reasoning process without requiring ground-truth annotations. By formalizing the notion of consistency verifiers -- reward functions that check for geometric and semantic consistency under transformations -- we demonstrate that models can improve their spatial reasoning abilities. We use both image transformations, like flipping, and textual transformations, like swapping the order of objects in the question, and propose a new optimal transport-based RL strategy, OT-GRPO, which is a minimal-matching variant of group relative policy optimization tailored to pairwise verifiers. We show that this label-free consistency training approaches the accuracy of models trained with ground-truth supervision and achieves similar generalization across diverse tasks and data domains.

This paper proposes an empirical methodology to study software aging in GPU-based LLM serving systems. Traditional aging studies focus on CPU-centric software with relatively regular workloads; LLM serving is different, spanning a Python host and a CUDA device, handling requests whose cost varies by orders of magnitude, and relying on rapidly evolving software stacks. We run a 216-hour campaign across six co-located deployments under identical stress conditions, monitor host, device, and client metrics in parallel, and apply a statistical pipeline that accounts for autocorrelation and multiple testing. Our results reveal statistically significant memory aging in all deployments, with leak rates strongly dependent on the serving runtime and deployment configuration. Beyond these findings, we provide a reproducible framework that opens a research direction at the intersection of the software aging and rejuvenation and LLM serving communities.

We present a quality-adaptive angular-margin learning framework that improves feature generalization by enforcing intra-class compactness and inter-class separability. Our framework, titled QLung, introduces a no-reference audio quality margin derived from spectral entropy and root-mean-square energy, which adaptively scales angular margins based on recording quality. To this end, we propose a log-scaled angular margin that stabilizes training under severe class imbalance. We also use an angular classifier that normalizes features and class weights, ensuring margin penalties are applied consistently on the unit hypersphere. Our approach improves in-distribution performance on the ICBHI dataset by 2.46\% over the cross-entropy baseline, and most significantly, achieves the strongest out-of-distribution performance on the SPRSound dataset compared to prior state-of-the-art methods. Code is available at https://github.com/RSC-Toolkit/QLung.

CSI-based localization with spatially distributed antenna arrays exposes a basic resource trade-off. Each array can provide a rich view of the channel, but forwarding observations from all arrays to a fusion center is wasteful when only a few carry useful information, and the shared uplink supports only a limited number of simultaneous transmissions. We let each array decide locally whether its current observation is worth reporting, subject to a budget on the average number of active transmitters. We refer to this abstraction as Edge-Triggered Distributed Inference (ETDI). It captures a broader class of task-oriented communication problems where resource-constrained devices share an access channel for a common inference task. We instantiate ETDI for CSI-based localization, a common scenario in vehicular IoT networks. Spatially distributed remote antenna arrays (RAAs) encode local channel state information (CSI) from user equipment (UE) transmissions into latent features, and the fusion center estimates the UE position from the subset of reported features. We propose NARRAS, a decentralized reporting policy in which each RAA combines a recurrent summary of its recent observations with a memory of the last latent it transmitted. Training controls an explicit activity budget through differentiable activity penalties and validation-calibrated deterministic thresholds, and uses channel-chart regularization to shape the latent geometry. Experiments show that, at comparable uplink activity, NARRAS improves localization accuracy over learned and heuristic sparse-reporting strategies, while dense full-report models remain useful budget-free references. In low-activity regimes, chart regularization further reduces high-percentile localization errors, suggesting that geometry-aware latent representations are more robust under sparse reporting.

Persistence diagrams are common representations in topological data analysis, but they do not naturally live in a vector space, and the statistical tools developed for comparing them have largely evolved separately from those used for downstream prediction. We introduce STRAND (Survival Topological Representation ANalysis of Diagrams), which treats (collections of) PDs as survival data: each topological feature with persistence value $p = d - b$ is a fully observed time-to-event, and the persistence survival function $S(t) = \mathbb{P}(p > t)$ is the central object for comparing diagrams. From this single representation we derive (i) a non-parametric two-sample test with calibrated Type I error and high power from a small number of diagrams; (ii) interpretable effect sizes; and (iii) a 1-Wasserstein-stable feature vector for downstream machine learning. We validate calibration and power on synthetic manifolds with controlled topology, demonstrate competitive vectorisation across 14 graph and 3D point cloud benchmarks, and apply the method to study functional brain connectivity in fMRI/neuroscience data. To our knowledge, STRAND is the first method to provide hypothesis testing and vectorisation for persistence diagrams from a single coherent and interpretable representation.

Traffic law liability determination is critical for assigning legal penalties, requiring the simultaneous identification of interdependent statutory provisions across multiple legal dimensions. However, existing retrieval-augmented generation methods suffer from a multi-dimensional retrieval bottleneck: single axis architectures compress complex legal queries into a single pathway, causing interdependent statutory dimensions to be overlooked. To address this, we propose OMAGR, an ontology-guided framework that decomposes queries into ontology-aligned anchors and executes parallel graph retrieval across each dimension, ensuring independent retrieval across dimensions before fusion. To evaluate the proposed method, we created the TrafficLaw-QA dataset, an expert-validated benchmark dataset containing 200 questions and 527 legal provisions. Results show that TrafficOmni-RAG outperforms baselines on Context Precision and Faithfulness metrics. The findings demonstrate that parallel multi-anchor retrieval effectively resolves the multi-dimensional retrieval bottleneck, offering a promising direction for traffic law liability determination research.

Benchmarks are essential for evaluating embodied spatial intelligence, yet their construction is labor-intensive, hard to reuse, and difficult to maintain. Existing embodied benchmarks are often static and may quickly become saturated as models improve, limiting their ability to distinguish new capabilities. We propose Embodied-BenchClaw, an autonomous agentic system for constructing embodied spatial intelligence benchmarks. Given a user-specified evaluation intent, Embodied-BenchClaw automatically produces a complete and continually updatable benchmark package through a five-stage pipeline: intent blueprinting, data collection, structuring and cleaning, benchmark synthesis, and evaluation reporting. The pipeline is coordinated by three agents for planning, construction, and evaluation. To improve reusability and reliability, Embodied-BenchClaw introduces an extensible Skill Library and process quality control, enabling benchmark construction to be composable, verifiable, and repairable. We instantiate multiple benchmarks covering indoor spatial reasoning, outdoor spatial reasoning, robotic manipulation, quadruped robot navigation, UAV/aerial-view understanding, and static benchmark enhancement. These benchmarks span diverse embodied carriers, data sources, and spatial capabilities. Experiments with human evaluation, judge-based assessment, consistency checks, cost analysis, and ablations show that Embodied-BenchClaw can construct verifiable, executable, maintainable, and diagnostically useful embodied spatial benchmarks with reduced manual effort.

Vision-Language-Action (VLA) models have shown strong performance in language-conditioned robotic manipulation, yet their robustness to linguistic variation remains poorly understood. In this work, we present the first systematic multilingual evaluation of VLA models by translating the LIBERO benchmark into ten languages, revealing severe performance degradation under non-English instructions, with success rates dropping by 30-50%. Through fine-grained analysis of task executions, we find that language influence is highly non-uniform across steps: certain steps exhibit strong language dependence and dominate overall task failure, while others are largely language-agnostic. Based on this insight, we propose a step-wise inference-time intervention that aligns representations according to step language sensitivity, substantially improving performance under linguistic variation. Our results indicate that language robustness in VLA models is fundamentally a step-wise control problem, highlighting the importance of temporally structured analysis for reliable embodied agents.

Bimanual robot systems substantially expand manipulation capabilities, but coordinating two arms introduces additional control complexity and failure modes that are not well captured by existing benchmarks. We introduce DuoBench, an extensible benchmarking framework for bimanual manipulation policies on the FR3 Duo platform. DuoBench comprises eleven tasks spanning four coordination categories, implemented in simulation and partially reproduced in the real world through reproducible task recipes with 3D-printable assets. In addition, we propose a stage-based evaluation scheme that supports fine-grained semantic failure analysis beyond binary success and provide human-teleoperated datasets for all benchmark tasks. We benchmark several dual-arm imitation-learning and vision-language-action policies in simulation and on real hardware. Our results show that current policies remain challenged by bimanual manipulation, particularly in early interaction stages, parallel arm execution, and transfer between simulation and real-world settings. DuoBench provides a reproducible testbed for diagnosing these failure modes and studying future methods for dual-arm policy learning. Code, datasets, and videos are available at https://duobench.github.io/

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.

Scientific discovery workflows usually contain and rely heavily on lab notes, where researchers record observations, interpret uncertain results, and plan follow-up experiments. Such informative lab notes preserve evolving scientific reasoning and author uncertainty, rather than polished final results exhibited in publications, providing a valuable opportunity for AI to engage in scientific exploration at a more comprehensive and deeper level. However, most prior work on scientific text focuses on papers, protocols, or structured databases, leaving informal laboratory notes underexplored as inputs to AI agents for science. This gap matters because lab notes often intermingle validated observations, tentative judgments, and possible experimental next steps within the same passage. If these signals are conflated, an AI agent may mistake uncertain scientific judgments for confirmed conclusions or executable actions. To this end, we present Notes2Skills, a two-stage framework for turning lab notebooks into verifiable skills for scientific AI agents while preserving the author's certainty. Across seven conditions and three wet-lab sessions, Notes2Skills is the only configuration that neither mistakes uncertain notes for firm instructions nor discards firm ones. We show that certainty preservation is the missing piece between lab notebooks and reliable agent skills, opening a path toward safer AI co-scientist systems.

The correspondence between large language models (LLMs) and the neural mechanisms underlying human higher-order cognition remains insufficiently characterized. Given that language and reasoning in the human brain appear dissociable, an open question is whether LLMs align with neural signals from reasoning-related regions and whether such signals can improve them. Here, focusing on deductive reasoning, we show that LLM internal representations are not only partially aligned with task-fMRI activity but can also be directly enhanced by these signals. Using a neural-predictivity metric, we find that LLMs explain a substantial fraction of the explainable variance in reasoning-related regions at the aggregate level, whereas predictivity within specific reasoning types is lower, indicating both alignment and divergence. Building on this, we propose a brain-guided framework: we steer model representations along directions induced by the joint structure of model and brain representations, applying intervention at inference and fine-tuning during training. We demonstrate that task-evoked brain signals can directly enhance LLM reasoning, yielding gains orthogonal to language-only supervision across 10 LLMs (1.5B-72B), with transfer across reasoning types and up to 13\% absolute accuracy gain. Our results advance LLM-brain correspondences from correlation to guidance, establishing a brain-signal-driven pathway toward more robust and cognitively aligned AI.

Multi-objective reinforcement learning for humanoid robots must coordinate locomotion and manipulation within a single policy. A natural design choice is whether to use a single (unified) critic that estimates the combined value of all objectives, or separate (dual) critics with disjoint reward signals. We present a controlled comparison on the Unitree G1 humanoid (23 active DoF) in NVIDIA Isaac Lab, training loco-manipulation policies through a sequential curriculum spanning 13 levels from stationary reaching to walking with variable-orientation targets. In standardized evaluation, dual-critic policies reach targets 3.5$\times$ faster (6.5 vs. 22.6 simulation steps), achieve 2$\times$ higher throughput (14.3 vs. 7.0 validated reaches per 1,000 steps), and attain higher validated reach rates (65.2% vs. 53.8%) compared to the unified-critic policy. Notably, additional anti-gaming reward mechanisms provide no further improvement beyond the architectural change alone (60.9% vs. 65.2%). These results have direct implications for the emerging paradigm of RL fine-tuning of imitation-learned policies: when refining a pre-trained manipulation policy with RL, a unified critic risks suppressing the learned behavior through competing locomotion gradients. These findings demonstrate that critic architecture is a primary - and often overlooked - design choice in multi-objective humanoid RL, with greater impact than reward engineering on reaching efficiency.

Vision-language models (VLMs) are increasingly used for scene understanding in autonomous driving, but robustness analysis often relies on task-agnostic embedding stability alone. We study whether corruption-induced embedding drift predicts changes in a task-aligned hazard score derived from CLIP image-text similarities. Using controlled corruptions on BDD100K road scenes, we compare embedding drift against margin drift, defined as the change in hazard score under perturbation. The relationship is highly corruption-dependent: some families exhibit strong coupling between representation drift and decision drift, while others induce hazardous decision instability despite relatively modest embedding change. Furthermore, corruption families differ in failure direction: most suppress hazard detections via false negatives, while occlusion instead triggers false alarms, suggesting that benchmark design should account for asymmetric failure modes, not just overall instability rates. These results suggest that robustness benchmarks should include task-aligned stability measures in addition to embedding-level perturbation statistics.

Circulating-tumour DNA (ctDNA) carries evidence of drug resistance months before imaging shows it, but the earliest evidence lives below the assay's limit of detection (LoD): a nascent subclone is detected only intermittently, producing a flickering sequence of faint detects and non-detects. Commercial liquid biopsies treat each draw as an independent snapshot and a non-detect as nothing. We argue a non-detect is a left-censored observation, and the pattern of non-detects and faint detects over time carries actionable evidence of growth before any single value is trustworthy. We introduce Span, a censored-Poisson Bayesian latent-growth change-point detector that models the binary detection process, accumulates a sequential generalised-likelihood-ratio statistic for an upward change-point in the per-variant detection rate, and raises a competing-risks alarm with calibrated false-alarm control. Span has no learned weights, so there is nothing to overfit. On a synthetic cohort of HR+/HER2- metastatic breast cancer on first-line CDK4/6-inhibitor plus endocrine therapy, at a matched 10% false-alarm rate, Span roughly doubles the fraction of impending progressions caught three months ahead (indolent regime: 25% vs 11% for the snapshot), with a falsifiable dose-response: large for indolent emergence, vanishing for fast emergence. A value-trajectory baseline performs identically to the snapshot, isolating the gain to the censored detection model. The survival backbone matches a Cox baseline on real breast-cancer data (GBSG-2, n=686; C-index 0.67 vs 0.68), and on a real longitudinal cohort with clean biomarkers (PBC2, n=312) the same pipeline correctly declines to win, a falsifiable boundary test confirming the mechanism is regime-specific. All ctDNA trajectories are synthetic.

Emotional validation - explicitly acknowledging that a user's feelings make sense - has proven therapeutic value but has received little computational attention. Emotional validation in dialogue systems can be decomposed into (i) validating response identification, (ii) validation timing detection, and (iii) validating response generation. To support research on all three subtasks, we release M-EDESConv, a 120k English-Japanese multilingual corpus created through hybrid manual and automatic annotation, and M-TESC, a multilingual spoken-dialogue test set. For timing detection, we propose MEGUMI, a Multilingual Emotion-aware Gated Unit for Mutual Integration, that fuses frozen XLM-RoBERTa semantics with language-specific emotion encoders via cross-modal attention and gated fusion. MEGUMI shows superior performance on both the M-EDESConv and M-TESC datasets, both objectively and subjectively. Finally, our EmoValidBench benchmarks of GPT-4.1 Nano and Llama-3.1 8B indicate that current LLMs generate contextually similar and diverse validating responses, but emotional understanding remains a major area for improvement. Project page: https://github.com/zihaurpang/Multilingual-Emotional-Validation

With the development of autonomous driving systems, mining high-value, safety-critical, and planning-relevant scenarios from large-scale driving logs has become essential for data-driven evaluation. In this paper, we propose AutoMine, a robust self-refining scenario mining method based on LLMs and VLMs. AutoMine uses semantics-preserving prompt augmentation to reduce LLM prompt sensitivity, combines robust trajectory atomic functions with VLM-based functions to handle perception noise and open-world visual cues, and refines generated code through execution feedback from real logs. In the Argoverse 2 Scenario Mining Competition at CVPR 2026, AutoMine achieves a HOTA-Temporal score of 36.38 and a Timestamp BA score of 77.21.

Machine learning is increasingly applied to accelerate the discovery of novel materials by exploring large compositional and structural design spaces. Yet, the scarcity of high-quality data and the frequent need for out-of-distribution prediction introduce substantial uncertainty, making the assessment of model reliability essential. In this work, we investigate uncertainty quantification as a means to evaluate model confidence in the context of permanent magnet research. In a first study, we benchmark classical and modern machine learning models for predicting intrinsic magnetic properties, focusing on the quality of their uncertainty estimates. We apply Gaussian negative log-likelihood loss and dropout-based Bayesian approximation as practical strategies for estimating predictive uncertainty. In a second study, we transfer these architectural features for uncertainty estimation to a more complex task: predicting coercivity from microstructural information using a graph neural network. Together, these studies demonstrate that uncertainty quantification not only enhances the trustworthiness of predictions but is also transferable across different modeling tasks.

Custom AI agents areagents that live inside their own application, talk to their own data and tools, enforce their own security boundaries, and carry their own brand and audit trail. What separates them from the general-purpose tier is fit, not capability: each is built for one job, by the engineer who will maintain it. No published practice sets out how to build one end to end. The pieces are everywhere (function-calling APIs, the Model Context Protocol, code agents to pair with), but the practice that chains them lives in podcasts, blogs, and leaked system prompts. This paper writes that practice down as a methodology, Agents All the Way Down: two preconditions crossed once and kept, then three practices repeated for the agent's life. The preconditions are (P1) Substrate, the LLM as a software component, framed as tools, then system, then messages under prompt-caching; and (P2) Building blocks: function calling, MCP, CLI orchestration, the liteshell pattern, the agent loop, skills, characters, hooks, and scaffolding. The practices are (P3) prototype with a general-purpose agent; (P4) harvest, fold, and ship the result as a CLI, the Turtle pattern; and (P5) agent-tests-agent, in which a general-purpose agent drives it through behavioural scenarios, a complement to classical testing, not a replacement. The working loop is P3 to P4 to P5 and back, and one corollary falls out for free: multi-agent orchestration is just CLI composition. The methodology is framework-free by construction. It was distilled from the AAC, a custom agent for the open-source LAMB platform, built in about ten days by one developer with an AI pair-programmer and in production . We present it as a transferable practice, independent of any language or framework.

De novo peptide sequencing from tandem mass spectrometry is pivotal in proteomics, enabling identification of novel peptides without reference databases. While recent Transformer-based encoder-decoder models have achieved remarkable performance, we uncover a critical pathology in their inference dynamics. Through comprehensive feature scaling experiments, we demonstrate that existing auto-regressive peptide decoders tend to over-rely on generated-sequence priors while progressively under-utilizing fine-grained physical evidence from the input mass spectrum. This phenomenon leads to suboptimal results, where generated peptide sequences are biologically plausible yet not faithful to the input spectrum. To rectify this, we propose MemNovo, a training-free and plug-and-play mechanism that re-balances peptide and spectral contributions at inference time. MemNovo alleviates the information bottleneck by establishing a persistent spectral memory bank and injecting retrieved features directly into the final decoding stage via an ultra-conservative residual connection. Theoretical analysis confirms that this mechanism restores the mutual information between the decoder state and the raw spectrum. Extensive experiments on the Nine Species benchmark with two representative baselines, Casanovo and InstaNovo, demonstrate that MemNovo consistently improves both amino acid precision and peptide precision, achieving up to 39.1% relative improvement in peptide precision for Casanovo and up to 3.9% for InstaNovo, with negligible computational overhead.

Mixture-of-Experts (MoE) and reinforcement learning (RL) post-training now dominate large language model (LLM) development, yet expert load imbalance remains a critical challenge. Existing load-balancing systems target pre-training by relying on historical step-level statistics. However, these methods fail under the unique workload dynamics of RL post-training: the step-level load is stable, but the tiny batch sizes processed during micro-steps cause severe, high-frequency load fluctuations. We introduce ForeMoE, a micro-step-level load balancing system for MoE RL post-training. Instead of relying on historical statistics, ForeMoE exploits the multi-stage RL pipeline (rollout, recompute, policy update) by using foreseeable routing information from the rollout stage to proactively guide load balancing in the remaining stages. To support frequent per-micro-step reconfiguration, ForeMoE employs a hierarchical planner that decomposes the NP-hard load balancing problem into tractable sub-components, alongside a transfer engine that leverages complementary hardware paths (CPU-assisted and GPU-direct) for overlapped expert transfer. Evaluations on 64 GPUs demonstrate that ForeMoE achieves up to a 1.45$\times$ speedup over state-of-the-art RL post-training systems.

Conformal Bayes combines Bayesian posterior predictives with conformal calibration to produce prediction sets that are both statistically valid and geometrically efficient. We study conformal Bayes under label shift from a unified perspective, identifying two complementary approaches that restore nominal target-domain coverage through importance-weighted conformal calibration but operate through independent mechanisms. \emph{Post-hoc calibration} tilts the posterior predictive toward the target domain and corrects the conformal threshold via an importance-weighted quantile, leaving the parameter posterior unchanged. \emph{In-training adaptation} tilts the parameter posterior itself to the target domain, producing a corrected predictive whose highest predictive density region serves as the highest predictive density (HPD) based prediction set under the fitted target predictive; efficiency is model-dependent and does not imply finite-sample conditional optimality. Two controlled experiments show that in an unbiased training regime both strategies achieve valid coverage equally, while in a lead-optimization regime in-training adaptation acts as a debiasing operator, reducing interval width at unchanged coverage.

Large language models (LLMs) have shown great promise for automated code translation, yet existing approaches often rely on token-level statistical patterns rather than sufficient understanding of program semantics. As a result, translated programs may still contain logical and semantic errors. Although high-quality semantic guidance, such as functional descriptions and test cases, can help mitigate these errors, such resources are often unavailable in real-world scenarios. This raises two key challenges: how to construct rich semantic information directly from source code, and how to ensure that such semantics are accurate and reliable enough to guide translation.To address these challenges, we propose Multisage, a multi-semantic augmentation and self-calibration framework for LLM-based code translation. Multisage consists of three modules. First, a semantic representation parsing module extracts structured base semantics from source code, including data-flow graphs, type constraints, and external API information. Second, a multi-semantic augmentation module builds on these representations to generate diverse augmented semantics, including code summaries, function-level test cases, and API-oriented descriptions and tests. Third, a semantic consistency calibration module uses semantics-preserving mutations and cross-semantic consistency verification to filter, calibrate, and refine the generated semantics.Experiments on the HumanEval-X code translation benchmark show that Multisage improves translation success rates by up to 2.22 times across diverse backbone models. It consistently outperforms vanilla prompting, instruction-tuned LLMs, and Chain-of-Thought reasoning, with the largest gains observed on smaller models. These results demonstrate that explicit semantic augmentation can substantially improve the reliability of LLM-based code translation.

In this paper, we introduce Representation Prediction via Autoencoding using Iterative Refinement (RePAIR) - a novel self-supervised representation learning architecture that synthesizes Masked Autoencoders (MAE), Joint Embedding Predictive Architectures (JEPA), and Bidirectional Encoder Representations from Transformers (BERT). We demonstrate how it can be used to encode objects in sequential data like consecutive chess positions into compact yet meaningful representations. The basic principle of the architecture is to mask large portions of a sequence of latent states, similar to BERT and MAE. Then, we apply a lightweight Predictor to the latent representations that repairs gaps in the sequence in a lower-dimensional embedding space akin to JEPA. Our experiments in the domain of chess show that the Encoder refines the board representations such that meaningful chess concepts emerge clustered in the latent space. Furthermore, reconstructions of the masked board states show that the model is able to reason about the piece movements without relying on costly reinforcement learning methods. Lastly, we find that the resulting representation space allows for quick and intuitive dissections of chess games by observing the game path trajectories in this semantically rich space.

Multi-channel mixed-SNR training improves out-of-distribution (OOD) generalisation of deep learning channel estimators for IEEE 802.11p vehicular communications, yet the internal mechanism responsible for this remains unexplained. This work presents REACH (Relevance-based Explanation and Architectural Compression for cHannel estimators), a gradient-based interpretability framework that operates at two levels. Input-level attribution identifies a subset of time-frequency features consistently relevant across all evaluated channel conditions, enabling input dimensionality reduction with minimal performance loss. Filter-level attribution reveals a near-universal internal representation, providing a representational account of the observed OOD generalisation. Guided by the resulting filter taxonomy, relevance-guided architecture compression substantially reduces both the number of parameters and the number of floating-point operations (FLOPs) with sub-1 dB normalised mean square error (NMSE) degradation, and OOD generalisation degrades more slowly than within-distribution accuracy under increasing compression.

There are two main Parameter-Efficient Fine-Tuning (PEFT) techniques for Large Language Models (LLMs). While Low-Rank Adaptation (LoRA) introduces additional weights between the LLM layers, Soft Prompting introduces additional fine-tuning-specific raw tokens to an LLM input. However, both require modification to the computational graphs of precompiled, preoptimized LLMs. As a result, neither is fully supported in high-throughput engines like vLLM. We propose fine-tuning with ART (Art-based Reinforcement Training). The method injects information into a frozen Multimodal Large Language Model (MLLM) by optimizing only its raw visual input, thus enabling the soft-token approach on pre-compiled computational graphs. It relies on backpropagation of gradients back into a plain pixel array and thus supports any fine-tuning objective. Moreover, the optimized visual input can be stylized as task-relevant computational artworks. The approach's effectiveness is confirmed for different sizes of a popular open Qwen architecture and for several textual benchmarks. Specifically, ART reaches accuracy competitive with LoRA across mathematics and structured-tool-use benchmarks.

Multi-modal large language models (MLLMs) depend on in-context learning (ICL) for rapid task adaptation, but their scalability is severely limited by finite context windows and the growing cost of key-value (KV) caches in long multi-modal sequences. Existing memory compression approaches typically rely on rigid token removal or sample-dependent importance estimation, which introduces bias, disrupts semantic structure, particularly for visual representations, and yields static memories that cannot adapt to new queries. We introduce TASM (Task-Aware Structured Memory), a training-free framework that addresses these limitations through task-aware, structure-preserving, and dynamically accessible memory construction. TASM employs task-vector guided compression to replace sample-specific signals with a task-level direction that captures shared relevance across demonstrations. To preserve the underlying manifold, it applies semantics-aware token merging via bipartite graph matching, aggregating tokens without destructive pruning. Finally, TASM structures memory into a hierarchy comprising a compact Core Memory and a Latent Bank, facilitating query-adaptive dynamic retrieval. Evaluations confirm TASM maintains high performance under heavy compression, effectively balancing efficiency with adaptability.

Open-ended scientific discovery asks agents to move beyond executing analyses for predefined questions. Across multiple rounds of exploration, a discovery agent must decide which phenomena warrant investigation while avoiding overinterpretation, where emerging claims exceed the evidential scope of the analyses supporting them. This creates an evidence-calibration problem: the exploration trajectory must be coupled with claim status so that evidence can guide both what to investigate next and what can be claimed. We introduce StatefulDiscovery, a discovery framework that externalizes investigation state and uses it to coordinate frontier selection, evidence acquisition, and claim adjudication. We evaluate StatefulDiscovery across 40 real-data discovery tasks. Compared with several baselines, StatefulDiscovery produces more claims overall judged to be both well-supported and high-value. Ablations indicate that structured hypotheses, local adjudication, and frontier control contribute to performance. Together, these results suggest that explicit discovery state can couple exploration with evidence-calibrated claim formation.

Continual anomaly detection in tabular data is challenging and remains largely underexplored, particularly in settings with heterogeneous feature schemas, distribution shifts, and severe class imbalance. In many real-world applications, data arrive sequentially from diverse domains, rendering conventional continual learning methods ineffective due to their reliance on a fixed input space. We propose a continual learning (CL) method, which can overcome these challenges and continually learn from different tasks. Our method consists of three main parts: our AGF model, Taskfusion augmentation, and outlier exposure. The AGF-model maps task-specific features into a shared space, then aligns distributions to reduce representation drift, and learns anomaly decision boundaries in the aligned space. To improve stability, we introduce Taskfusion augmentation, combining boundary-aware interpolation within tasks to refine the model anomaly boundaries and cross-task mixing to transfer anomaly structure across datasets. To handle class imbalance and memory constraints, we employ tabular dataset distillation to store compact synthetic replay samples, which are jointly used with augmented data in an outlier exposure objective for robust anomaly detection. We evaluate the approach on 21 heterogeneous datasets across multiple domains. Results show that our approach substantially improves continual anomaly detection performance over sequential fine-tuning and other CL baselines while reducing catastrophic forgetting and maintaining stable detection across heterogeneous datasets.

Open-vocabulary scene sketch semantic segmentation aims to assign dense semantic labels to sparse line drawings based on flexible category vocabularies specified at inference time, without relying on pixel-level annotations during training. Unlike natural images, sketches lack texture and color cues, making semantic understanding heavily dependent on stroke layout and spatial configuration, a challenge that renders single-layer vision-language features inherently unstable. Our key observation is that attention maps from different Vision Transformer layers encode complementary spatial cues: shallow layers capture global structural layouts, while deeper layers focus on local stroke intersections and object parts. This suggests that cross-layer aggregation provides a more robust structural prior than any individual layer alone. Leveraging this insight, we propose a structure-aware framework built upon \textbf{L}ayer-wise \textbf{A}ccumulated \textbf{S}tructural \textbf{A}ttention (\textbf{LASA}), which aggregates multi-layer attention to guide hierarchical semantic alignment under weak supervision and refine predictions during inference. Experiments on FS-COCO, SFSD, and FrISS show that LASA improves mIoU by $+3.43$, $+8.01$, and $+15.74$ over the prior weakly supervised baselines, demonstrating consistent gains in both segmentation accuracy and spatial coherence. Our source code will be made publicly available.

This paper presents a novel data-free and training-free compression approach for speech foundation models using channelwise clustering via k-means. More fine-grained, mixed sparsity pruning by layer-level varying number of parameter clusters is also explored. Experiments conducted on the LibriSpeech dataset suggest that when operating with pruning sparsity of 50% on HuBERT-large, consistent WER reductions of 27.73%/18.61% absolute (34.37%/21.91% relative) over the magnitude-based pruning were obtained on the test-clean and test-other subsets before fine-tuning and 0.19%/0.79% absolute (3.36%/4.62% relative) after fine-tuning with only 3 epochs. Similar WER reductions of 2.86%/5.02% absolute (59.21%/55.29% relative) were observed against magnitudebased pruning on Whisper-large-v3 at 10% sparsity, all with no significant WER increase relative to the uncompressed baseline.

Collecting participants' lived experiences is central to design research. Focus groups are uniquely valuable because participants not only share individual accounts but also respond to one another, surfacing comparison, disagreement, and collective sensemaking. However, focus groups are resource-intensive and highly sensitive to facilitation: moderators must probe for specificity, balance participation, manage topic flow, and sustain psychological safety, and subtle facilitation choices can shape what becomes salient. Recent HCI work and commercial meeting tools show that generative AI can scaffold live conversation through prompting, turn regulation, thematic mapping, and real-time summarization. Yet UXR teams lack a clear map of what these capabilities mean in focus groups and what methodological risks they introduce. We synthesize AI supports for live conversation and translate them into a focus-group-specific playbook organized by AI role (tool, co-host, host) and modality (text, voice, embodied).We synthesize prior work on AI-supported live conversation and propose a focus-group-specific playbook of AI supports organized by role (tool, co-host, host) and modality (text, voice, embodied). We characterize interactional trade-offs and identify open questions for evaluating AI-supported focus groups as methodological configurations.

Traceability information between requirements and source code greatly benefits the maintenance of a software system. Since manually establishing trace links is cumbersome and error-prone, previous research explored automated traceability link recovery (TLR) approaches to support this task. However, quality defects in requirements impact subsequent activities such as TLR, yet evidence about this remains scarce. Our objective is to contribute empirical evidence on this impact. At the same time, we aim to understand how the performance of TLR approaches varies given these quality defects. To this end, we annotated 28 types of quality defect in 189 use case descriptions from two datasets. Then, we executed five distinct TLR approaches on the dataset and measured their performance in recovering trace links. Finally, we performed statistical tests to quantify the defects' effect strength on this performance. Our results show that some quality defects harm TLR performance, e.g., sentences that do not start with noun phrases, while others actually benefit performance, e.g., use cases that include implementation details. Moreover, different types of approaches respond differently to these defects. As a consequence, the performance-optimizing choice of a TLR approach depends on the quality of the dataset.

Flow matching and diffusion models enable conditional generation across domains ranging from images to proteins, with recent extensions to out-of-distribution contexts. Yet generative models of neural time series have largely remained restricted to categorical conditioning, precluding compositional and zero-shot generalization. In this work, we propose a per-timestep conditioned diffusion transformer for generating realistic fMRI brain dynamics during unseen cognitive tasks by injecting both compositional language and optional spatial priors in-context. Such zero-shot generation could enable counterfactual neuroscience by supporting in-silico design and evaluation of novel cognitive experiments before empirical validation. Leveraging this model, we evaluate across hundreds of held-out task conditions and characterize predictive performance in relation to the training manifold. From language alone, the model recovers region-specific recruitment across tasks and held-out spatial activation patterns. Spatial priors, when available, complement the text pathway by anchoring generation in regions of task space where language alone degrades, while retaining the compositional structure needed for counterfactual task specification. To our knowledge this is the first generative model of whole-cortex fMRI dynamics for unseen cognitive tasks, advancing counterfactual neuroscience and data-driven experimental design.

Neural relational inference (NRI) methods discover interaction graphs from trajectories through variational reasoning on discrete potential edges. However, these methods typically rely on oversimplified, factorized graph priors. Such priors, typically nearing uniform distributions, treat edges as independent entities. This systemic misalignment does not match the real-world systems and yields diffuse and indecisive edge posteriors limiting the reliability of structural discovery. To address this, we propose \textit{Diff-prior}, a diffusion-parameterized adaptive prior used to calibrate latent graph distribution rather than generate graphs. Our core insight is to reframe prior integration as a learnable denoising-style calibration that organizes scattered, uncertain edge posteriors into a more reliable overall structure which can be trained by the diffusion model. Diff-prior learns an adaptive structure prior that performs structured calibration on the edge posteriors during inference, guiding it towards a distribution closer to the underlying structure. The diff-prior operates before structural sampling and acts as a denoising calibrator directly on the encoder edge distribution, which provides a generic training paradigm over structured variables. Experiments on standard benchmarks validated our framework, and the results indicate that Diff-prior improves the performance of structure inference and generates more decisive edge posteriors across multiple NRI-family architectures. The code is available on https://github.com/Hardy158118/Diffprior.

Background. Large language models and AI agents are increasingly used to support biomedical research, but native model outputs may omit key analytical steps, misuse methods, or overstate conclusions. We evaluated whether autonomous access to a medical research skill package was associated with higher-quality AI-generated transcriptomic research-analysis outputs compared with native AI without skills. Methods. We conducted an exploratory multi-model human evaluation using a non-small cell lung cancer immunotherapy biomarker task. Six model backbones were tested. The evaluation included 21 anonymized outputs: 9 native-AI outputs and 12 skill-augmented outputs generated through an AI agent implementation represented by OpenClaw. Four non-expert biomedical reviewers and two blinded experts evaluated each output, with two ratings from each reviewer type. The primary outcome was expert-rated overall quality. Results. Skill-augmented outputs showed directionally higher expert overall quality than native-AI outputs (mean 5.50 vs 5.11; difference=0.39; bootstrap 95\% CI, -0.04 to 0.90; Welch p=0.156). Non-expert reviewer quality showed the same direction (mean 4.72 vs 4.47; difference=0.26; bootstrap 95\% CI, -0.25 to 0.80; Welch p=0.373). Expert agreement was limited (single-rating ICC=-0.15), and model-specific effects were descriptive and heterogeneous. Conclusions. Autonomous skill access showed a directional quality signal in this exploratory sample, but the signal was smaller than expert-rating noise and should not be interpreted as confirmatory evidence. The findings primarily motivate larger evaluations of skill-augmented AI agents with stronger reliability controls, platform replication, and biological-validity assessment.

Audio watermarking aims to embed identifiable information into audio while remaining imperceptible. Existing methods adopt high-fidelity, low-energy designs to preserve perceptual quality, but the resulting watermarks lack robustness under suppression by speech reconstruction models. Improving robustness is challenging due to the inherent robustness-fidelity trade-off in existing designs, where increasing watermark energy improves robustness but reduces fidelity. To address this problem, we propose a feature-aligned watermarking method that aligns the watermark with the original speech feature distribution, allowing higher watermark energy to improve robustness while preserving imperceptibility. We use a pretrained speech codec to generate a pseudo-speech watermark and fuse it into the spectrogram of the input audio, with VAD loss and perceptual losses guiding embedding within voiced regions. Experiments show that our method maintains imperceptibility comparable to existing approaches while substantially improving robustness under both seen and unseen speech reconstruction models.

The rapid evolution of cloud computing has resulted in the adoption of hybrid deployments that blend Infrastructure-as-a-Service (IaaS) and Function-as-a-Service (FaaS) service models to optimize resource utilization, scalability, and operational efficiency. This paper presents a comprehensive study and practical implementation of a metrics-driven approach for migrating microservices from a traditional IaaS service model to a hybrid IaaS + FaaS model, using two microservice applications as case studies. The research develops an automated framework to analyze service-level performance metrics to identify microservices that are best suited for serverless execution. The findings of our research highlight the benefits and limitations of different cloud service models and provide a scalable and replicable automated methodology for optimized deployment of cloud-native applications.

Large Language Models (LLMs) are increasingly used for code generation, raising concerns that they may be misused to produce malicious code. Meanwhile, Grammar-Constrained Decoding (GCD) has been widely adopted to improve the reliability of LLM-generated code by enforcing syntactic validity. In this paper, we reveal a counterintuitive risk: this reliability-oriented technique can itself become an attack surface. We uncover a new jailbreak attack, termed CodeSpear, that exploits GCD to induce LLMs into generating malicious code. Our experiments show that simply applying a benign code grammar constraint can effectively jailbreak LLMs. To address this vulnerability, we propose CodeShield, a safety alignment approach that robustly preserves safe behavior even under attacker-controlled grammar constraints. CodeShield aligns the model in the code modality by teaching it to generate honeypot code under GCD. Such code is semantically harmless, so it does not implement the malicious request, and structurally diverse, so it is difficult to suppress through grammar tightening. At the same time, CodeShield still preserves natural-language refusals when natural language is available. Experiments on 10 popular LLMs across 4 benchmarks show that CodeSpear outperforms representative jailbreak baselines and increases the attack success rate by more than 30 percentage points on average. CodeShield also restores safety under CodeSpear while preserving benign utility. Our findings reveal a fundamental risk of GCD and call for greater attention to its potential security implications.

Forecasting real-world events requires language-model agents to reason under uncertainty from incomplete, time-bounded information. Yet evaluating whether agents genuinely forecast requires more than final-answer accuracy: a model may be correct by recalling memorized training facts, citing fabricated evidence, or producing an unsupported causal story. We present WorldReasoner, an evaluation framework for temporally valid event forecasting. Each task gives an agent a resolved forecasting question, a simulated forecast date, and access only to evidence available before that date; after resolution, the framework scores the submitted probability, cited evidence, and optional causal event graph. WorldReasoner reports three complementary axes: outcome quality against resolved answers, evidence quality over cited sources, and reasoning quality against post-resolution hindsight graphs. The benchmark is built by an agentic construction pipeline that generates forecasting questions, collects time-stamped evidence, and builds hindsight reference graphs at scale, yielding 345 resolved tasks derived from 14,141 articles with graphs covering 8,087 extracted events. Across six controlled agent settings, temporally valid retrieval is the strongest driver of outcome accuracy; causal graph construction improves key-event recovery; and correct graph-enabled forecasts are more strongly grounded in key events and relevant sources, yet agents still struggle to convert grounded evidence into calibrated probabilities.

Scalable distributed systems form the backbone of modern computing infrastructure. However, as scale grows, system complexity may lead to scalability faults. Scalability faults are challenging to uncover and diagnose, as they are often latent and only manifest at large-scale deployment. In this paper, we present the first comprehensive study on scalability faults and propose an approach for their detection. First, we systematically investigate 444 scalability issue reports from 10 large-scale distributed systems to understand the common anti-patterns and root causes of scalability faults. We found that the majority of these faults are caused by the synergy between dimensional code fragments and anti-patterns associated with them. Second, based on our findings, we design and implement ScaleLens, a novel approach to detect scalability faults. ScaleLens combines dynamic and static analyses to pinpoint dimensional code fragments and match them with anti-patterns. Our evaluation shows that ScaleLens detects 4.2x more dimensional code fragments associated with known scalability faults compared to the baseline. On the latest stable versions of Cassandra, HDFS, and Ignite, ScaleLens detects 334 dimensional code fragments with confirmed problematic behavior.

Machine-learning approaches to quantum state tomography can achieve high reconstruction fidelity, but the physical structure used by the trained model often remains implicit. Here we ask whether a sparsified Kolmogorov-Arnold Network (KAN) can be used not only as a regressor, but also as an inspectable reconstruction rule whose internal organization can be checked against known Pauli structure. We study a controlled three-qubit GHZ-family benchmark in which all 63 non-identity Pauli expectation values are used to reconstruct three GHZ-subspace variables: the population imbalance $z$, the real off-diagonal component $c$, and the imaginary off-diagonal component $s$. Under finite-shot sampling and depolarizing noise, external ablation identifies the extended 12-channel GHZ-relevant Pauli set from the 63 measurements, with exact top-12 recovery across the tested shot counts and depolarizing-noise strengths. These support patterns remain stable across multi-seed random-initialization and noise-level analyses, and collapse under random-label controls. The dominant pruned input-hidden-output pathways organize Z-type population observables and X/Y off-diagonal observables in a pattern consistent with the analytic GHZ Pauli grouping, and sparse formula recovery recovers the canonical signed Pauli relations. The contribution of the KAN is therefore pathway-level structural interpretability within a neural reconstruction model, rather than superior sparse regression. Together with negative controls, these probes provide a consistency chain for auditing learned reconstruction rules against known physical structure.

Production LLM systems accumulate reusable operational experience, but the practical deployment issue is not merely whether such experience can help. It is how different serving strategies trade off quality against online cost under realistic constraints. Injecting external experience can improve task quality, yet it also increases prompt burden, latency, and serving pressure. We study \textit{external experience serving} as a deployment-oriented quality-cost trade-off problem. We evaluate this question in a real production moderation setting, with tool-use and GPQA as supporting contrast tasks that expose different output-cost regimes. We compare no-experience baselines, random experience controls, global prompt injection, and retrieval-based selective injection, and analyze both task quality and serving cost. The results show that, once experience becomes case-dependent, selective retrieval provides a stronger operating point than unconditional global injection. They further show that retrieval quality matters more than simply increasing Top-$K$, and that the same serving policy can exhibit substantially different cost-benefit profiles across short-output and decode-heavy regimes. These findings suggest that external experience is best treated as a selective, cost-aware serving decision rather than as a universal add-on. Overall, in the settings studied here, external experience pays off only when both the serving interface and the task-specific cost structure make its quality gains worth the online cost.

Text-conditioned 3D generation has progressed rapidly for images and isolated objects, but producing a hand-object mesh remains challenging: the output must preserve language semantics, cross-view consistency, object geometry, articulated hand shape, and physically plausible contact. We present TextHOI-3D, a staged framework that uses generated multi-view observations as an explicit interface between text-conditioned visual generation and geometry-aware hand-object recovery. TextHOI-3D learns a compact VQ token space for fixed-camera hand-object observations, predicts multi-view visual tokens from text with a CLIP-conditioned visual autoregressive model, and recovers a unified hand-object mesh through prior initialization, multi-view joint optimization, and anti-penetration refinement. The design separates semantic generation from geometric recovery while keeping both stages connected by a discrete multi-view representation. On HO3D-derived evaluations, the multi-view setting reduces object CD from 17.26 mm to 4.92 mm and penetration volume from 5.3721 cm^3 to 0.2193 cm^3 compared with a single-view counterpart, while improving hand errors and surface F-scores. These results support multi-view visual tokens as an effective intermediate representation for text-driven 3D hand-object mesh creation.

Trustworthy AI requires reliable data-processing pipelines, not only robust downstream predictive models. As an upstream component, data summarization determines which information is retained and passed to subsequent learning or decision modules. Therefore, adversarial perturbations to the summarization process can compromise trustworthy AI in an upstream manner: they may alter the selected summary, reduce its representativeness, and further degrade the utility of subsequent learning tasks. In this paper, we study adversarial attacks on continuous data summarization under similarity-level perturbations through DR-submodular optimization. We show that a class of multi-resolution image summarization objectives can be formulated as multilinear extensions of non-negative submodular set functions and satisfy DR-submodularity with $m$-weak monotonicity. We then formulate multi-target attack generation as a min-max problem, where one admissible perturbation of the similarity structure is optimized to degrade multiple target summarization models. To mitigate such perturbations, we formulate robust defense against mixed attack types as a regularized max-min problem. For both problems, we develop approximation algorithms with theoretical guarantees. Experiments on real-data and controlled clustered benchmarks show that the proposed attack is effective in representative low-to-moderate budget regimes and can induce downstream task-performance loss. The proposed defense improves the robustness--mitigation trade-off in structured settings, while also revealing the parameter sensitivity of robust protection on real data.

In this paper, we develop a continuous-time model-free reinforcement learning algorithm to learn deterministic equilibrium policies in general time-inconsistent control problems. Utilizing the extended Hamilton-Jacobi-Bellman system, we recast the original time-inconsistent problem into an equivalent two-stage problem. In the first stage, for given auxiliary functions, we employ the deterministic policy gradient approach to learn an optimal policy in an auxiliary time-consistent control problem. In the second stage, given the updated policy, we exploit the inner fixed point iterations and some martingale characterizations to learn the auxiliary functions. As a theoretical contribution, we provide some mild model assumptions and establish the convergence of inner fixed point iterations. By repeating this actor-critic style of iterations across two stages, our algorithm aims to learn the equilibrium under different sources of time-inconsistency in a unified manner. The superior effectiveness of the proposed algorithm are illustrated in two classical financial applications with time-inconsistency: mean-variance portfolio management and optimal tracking portfolio under non-exponential discounting.

Studies on rodents such as mice have shown the capabilities to adapt their behavior when dealing with changing parameters (``drift'') of the environment even if no information about change is provided (uncertainty) -- a behavior that can be modeled by forgetting mechanisms. Non-stationary Reinforcement Learning (NSRL) deals with adapting state-of-the-art RL methods to deal with changing environments: these however usually require (partially) perfect information about the drift such as ``task IDs'' or ``context''. To mitigate the effects of drift, this work develops \emph{Space-sampled Value Decay} as an explicit forgetting mechanism for value-based deep RL architectures as a simple yet effective approach. In particular we demonstrate and discuss positive effects but also limitations in achieved returns for modifications of Deep Q-networks (DQN) and Soft Actor-Critic (SAC) when evaluated on non-stationary environments.

Neurodegenerative diseases such as Alzheimer's disease (AD) require accurate and scalable tools for assessing disease severity, yet current clinical staging remains time-intensive and prone to variability. We propose an attention-enhanced multimodal machine learning framework with ordinal regression for automated and interpretable AD severity staging. The framework integrates T1-weighted MRI with demographic and genetic variables and compares unimodal and multimodal architectures using ordinal and non-ordinal prediction heads. Models were trained and validated using cohort-stratified splits derived from the ADNI, AIBL, and NIFD datasets. A strictly held-out test set was constructed using subjects excluded from all training, validation, preprocessing, and hyperparameter tuning procedures, with subject-level splitting employed throughout to prevent data leakage. Among unimodal approaches, the T1-weighted MRI model achieved slightly higher adjacent-stage accuracy (0.963) and agreement with clinical staging (QWK 0.444) than the tabular model (QWK 0.433). Integrating imaging, demographic, and genetic information improved overall performance. The multimodal non-ordinal baseline achieved the lowest prediction error (MAE 0.340), whereas the ordinal multimodal model achieved the highest adjacent-stage accuracy (0.970) and strongest agreement with clinical staging (QWK 0.549). These findings indicate that ordinal formulations better capture the ordered structure of the CDR scale and yield predictions more consistent with clinical staging. Explainability analyses using Grad CAM++ and SHAP demonstrated anatomically and clinically plausible model behavior, supporting transparent decision-making. Overall, attention-based multimodal learning with ordinal regression represents a robust, interpretable, and scalable approach for automated AD severity staging and AI-assisted clinical decision support.

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.

Large Language Models (LLMs) offer new potential for translation tasks but often experience performance degradation when handling low-resource languages. To address this limitation, we propose an approach for fine-tuning LLMs on a low-resource language, Kupang Malay. Our approach involves designing a set of instructions by leveraging explicit lexical and semantic features from a bilingual dictionary, and introducing Continual Instruction Tuning (CIT), a training paradigm that enables iterative instruction-based training. Experimental results demonstrate that our model, named Lius, yields notable improvements over standard instruction-tuned models by outperforming 4-6 points, and surpassing both Neural Machine Translation (NMT) and Multilingual LLM models by 10-13 points on several evaluation metrics. These findings highlight the potential of our approach to mitigate the reliance on large-scale parallel data in low-resource language translation.

We establish conditions for embedding a corpus of $N$ documents as $d$-dimensional vectors such that every $k$-subset $S \subseteq [N]$ is realizable as a result of top-$k$ retrieval by some query vector. Recent work shows that $d = O(k)$ suffices for such embeddings to exist in $\mathbb{R}^d$, independently of $N$. We theoretically prove that this corpus-independent bound is specific to infinite precision. With $B$ bits per coordinate, perfect top-$k$ retrieval requires $Bd = Ω(k \ln N)$; thus, at any fixed precision, the dimension must grow at least logarithmically with $N$. Specializing to a $\ell_2$-normalized $B$-bit uniform scalar quantization model, we also identify a threshold on the precision $B^{*} = O(\ln \ln N)$ below which no dimension suffices, together with two further regimes that bound the feasible $(B, d)$ pairs. Our result implies that in practical vector databases and dense retrieval systems where quantization is standard, the embedding dimension and possibly the precision must grow with the corpus size.

We establish the correct parameter governing the convergence time of the 3-Majority and 2-Choices dynamics on the complete graph in the synchronous model. Recent work [Shimizu and Shiraga, PODC'25] provides matching upper and lower bounds on the number of rounds to consensus, but only in a weak sense: the bounds are shown to coincide for some initial opinion configuration. In contrast, we obtain tight bounds in a strong sense, with upper and lower bounds matching up to logarithmic factors for every initial configuration. Let $α$ (0) be the initial opinion-frequency vector, and denote by ___$α$ (0) ___ $\infty$ its maximum entry. We show that 3-Majority reaches consensus in $Θ$(min{___$α$ (0) ___ -1 $\infty$ , $\sqrt$ n}) rounds w.h.p., while 2-Choices reaches consensus in $Θ$(___$α$ (0) ___ -1 $\infty$ ) rounds w.h.p. Our results demonstrate that the convergence time of both dynamics is governed not by global parameters such as the number of opinions k or the squared ${\ell}$ 2 norm of the initial opinion distribution, but rather by the ''local'' parameter ___$α$ (0) ___ $\infty$ , the maximum initial opinion density.

Optimistic Gradient Descent Ascent (OGDA) and Optimistic Multiplicative-Weights Update (OMWU) are two very popular algorithms to solve convex/concave saddle-point problems, where OMWU is the non-Euclidean, entropic version of OGDA. It is known since the '80s that the last iterate of OGDA asymptotically converges to a saddle point in smooth problems. On the other hand, it is unknown if OMWU has the same property. In this paper, I show that OMWU converges asymptotically for smooth convex-concave saddle-point problems, with a small enough constant learning rate. The result does not require uniqueness, strict complementarity, an error bound, or initialization near a solution. The main new ingredient is a boundary argument showing that every cluster point satisfies the inactive-coordinate KKT inequalities. The boundary argument was discovered with assistance from ChatGPT and is documented in the appendix.

Spatial reasoning remains a challenge for Multimodal Large Language Models (MLLMs), as it requires reliable multi-hop inference over both intermediate states and state transitions. Current studies often leave intermediate states unverified and treat state transitions as implicit processes, which limits reliability in multi-hop spatial reasoning. To address this, we propose State-aware Visualization-of-Thought (SVoT), a reinforcement learning framework that generates interleaved, verifiable intermediate states and visualizations. SVoT integrates transition reasoning chains into the generation processes, enabling the model to verify action preconditions and effects through interleaved textual and visual reasoning. We train SVoT via Group Relative Policy Optimization (GRPO), instantiating verification through reward design and evaluating the efficacy of different fine-grained rewards. As existing benchmarks reduce state transitions to single-variable updates, substantially simplifying the problems, we establish five domains by extending classical environments and introducing two novel domains, Pacman and Gather, that require multi-object interactions and numerical reasoning. These domains support systematic evaluation of multi-hop spatial reasoning with quantitative verification of generated intermediate states and transition reasoning. SVoT with transition-aware supervision achieves state-of-the-art performance across the introduced domains, yielding up to a 65% absolute accuracy gain on out-of-distribution test sets.

The European AI Act is the first comprehensive regulation of artificial intelligence (AI), setting out extensive obligations, particularly for so-called high-risk and general-purpose AI systems. A key distinguishing feature of AI systems under the AI Act is the capability to infer. Since the AI Act does not clearly define what inference is, there is a gray area for certain data-driven systems. A specific example is credit scoring systems, which are listed by Annex III of the AI Act. At the same time, however, these are often implemented using statistical models for which it is unclear whether they have the capability to infer and thus fall under the AI definition of the AI Act at all. Motivated by statistical learning theory, this work develops a framework for grading different levels of the capability to infer. Based on the AI Act and the Commission Guidelines on the definition of an artificial intelligence system, we analyze which levels constitute sufficient capability to infer within the meaning of the AI Act and where further regulatory clarity is needed. We illustrate the framework by creating two realistic credit scoring workflows and show whether and where inference occurs in them. Our analysis illustrates that not only individual models but the entire data processing workflow must be considered. It also shows that the involvement of human experts during development can have significant influence on the capability to infer. Code can be found at https://github.com/fraunhofer-iais/inference-framework-creditscorecards.

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.

Distilling a large speech foundation model (SFM) into an efficient student model has been successfully applied to low-resource environments. Although distillation reduces inference latency, it requires an additional student model training. However, the training efficiency of SFM distillation remains underexplored. In this work, we explore training acceleration of SFM distillation to speed up model deployment. We examine the potential of stacking, in which the model depth is progressively increased through training until the target model depth is reached. While existing stacking methods improve training speed, they suffer from performance degradation. To handle this limitation, we propose interleaved stacking, a novel stacking method that consistently preserves layer position throughout the stacking process. This property is particularly critical in SFMs, in which each layer encodes distinct layer-specific knowledge. We validate the effectiveness of the proposed method on SUPERB.

Large language models (LLMs) have achieved remarkable progress in language understanding, reasoning, and generation, sparking growing interest in their creative potential. Realizing this potential requires systematic and scalable methods for evaluating creativity across diverse tasks. However, most existing creativity metrics are tightly coupled to specific tasks, embedding domain assumptions into the evaluation process, and limiting scalability and generality. To address this gap, we introduce an automated, domain-agnostic framework for quantifying LLM creativity across open-ended tasks. Our approach separates the measurement apparatus from the creative task itself, enabling scalable, task-agnostic assessment. Divergent creativity is measured using semantic entropy, a reference-free and robust metric for novelty and diversity, validated against human annotations, LLM-based novelty judgments and baseline diversity measures. Convergent creativity is assessed via a novel retrieval-based multi-agent judge framework that delivers context-sensitive evaluation of task fulfilment with over 60% improved efficiency. We validate our framework in three qualitatively distinct domains: problem-solving (MacGyver), research ideation (HypoGen), and creative writing (BookMIA), using a broad suite of LLMs. Empirical results show that our framework reliably captures key facets of creativity, including novelty, diversity, and task fulfilment, and reveal how model properties, such as size, temperature, recency, and reasoning, impact creative performance. Our work establishes a reproducible and generalizable standard for automated LLM creativity evaluation, paving the way for scalable benchmarking and accelerating progress in creative AI.

Dynamic data pruning techniques aim to reduce computational cost while minimizing information loss by periodically selecting representative subsets of input data during model training. However, existing methods often struggle to maintain strong worst-group accuracy, particularly at high pruning rates, across balanced and imbalanced datasets. To address this challenge, we propose RCAP, a Robust, Class-Aware, Probabilistic dynamic dataset pruning algorithm for classification tasks. RCAP applies a closed-form solution to estimate the fraction of samples to be included in the training subset for each individual class. This fraction is adaptively adjusted in every epoch using class-wise aggregated loss. Thereafter, it employs an adaptive sampling strategy that prioritizes samples having high loss for populating the class-wise subsets. We evaluate RCAP on six diverse datasets ranging from class-balanced to highly imbalanced using five distinct models across three training paradigms: training from scratch, transfer learning, and fine-tuning. Our approach consistently outperforms state-of-the-art dataset pruning methods, achieving superior worst-group accuracy at all pruning rates. Remarkably, with only $10\%$ data, RCAP delivers $>1\%$ improvement in performance on class-imbalanced datasets compared to full data training while providing an average $8.69\times$ speedup. The code can be accessed at https://github.com/atif-hassan/RCAP-dynamic-dataset-pruning

With the widespread adoption of Large Language Models (LLMs) in software engineering (SE) tasks such as code understanding, debugging, and vulnerability detection, their powerful semantic reasoning ability has also introduced new security and privacy risks. LLMs can analyze, reconstruct, or even reverse-engineer source code logic, potentially leading to the leakage of intellectual property. To address this issue, we propose Acoda, a genetic algorithm-based adversarial code obfuscation framework that defends against LLM-based code analysis. Acoda leverages two key mechanisms of LLMs, namely safety alignment and token-based information processing, to design 8 semantics-preserving obfuscation methods. It iteratively optimizes obfuscation strategies through a genetic algorithm to generate adversarial samples that maximize defensive effectiveness. In addition, we propose a quantitative evaluation framework based on LLM responses, which combines an auxiliary LLM and four evaluation metrics to assess how target LLMs analyze obfuscated code comprehensively. Experimental results show that Acoda can effectively induce LLMs to refuse or misinterpret code analysis. On 7 state-of-the-art LLMs, including GPT-4o, DeepSeek, Qwen, Llama, and Gemma, Acoda achieves an attack success rate (ASR) of up to 70%, with strong cross-model transferability and minimal runtime overhead, while ensuring that the semantics of the original code remain unchanged. Overall, this study provides a new perspective for code protection and LLM security defense in the era of LLMs.

Multi-turn image editing is essential for iterative design, yet current models often struggle with identity drift and error accumulation over successive steps. While existing research leverages video priors for consistency, their reliance on bidirectional attention is fundamentally misaligned with the causal, sequential nature of interactive editing. In this paper, we propose AnchorEdit, the first autoregressive (AR) diffusion-based framework designed specifically for high-resolution, long-term multi-turn editing. AnchorEdit bridges the gap between video priors and causal inference through a three-stage training curriculum: identity-preserving sing-turn pretraining, causal AR forcing fine-tuning with a novel self-rollout strategy to mitigate exposure bias, and consistency distillation for efficient 4-step generation. During inference, we introduce a memory mechanism to anchor the initial subject identity and ensure stable extrapolation across extended editing trajectories. To evaluate performance, we provide a new high-resolution multi-turn editing benchmark designed to stress-test long-horizon stability. Extensive experiments demonstrate that AnchorEdit achieves state-of-the-art results, maintaining exceptional subject fidelity and instruction following even over 10+ interaction rounds.

Visual causal reasoning is essential for understanding and intervening in the physical world, requiring identification of causal variables from visual inputs and reasoning over intervention effects. Despite recent progress, large vision--language models (VLMs) remain brittle at such tasks, especially for interventional and counterfactual queries over multi-image inputs. Most existing explorations inject causal knowledge via textual prompts, leaving causal mechanisms external to model execution and limiting reliable control during inference. To address this problem, we propose BridgeVLM, which internalizes visual causal reasoning by inducing a causal graph from multi-image inputs and converting it into structured Causal Tokens executed by RAMP layers injected into the LLM decoder for causal message passing. We further introduce a unified training interface M3S for fine-grained causal supervision from different granularities (local/global level). BridgeVLM achieves 54.4% accuracy on intervention tasks on CausalVLBench (vs. 33.2% with prompt-level supervision), improves results on Causal3D from 43.6% to 49.0%, and substantially improves causal structure learning on CausalVLBench ($F_1$: 33.4% $\rightarrow$ 75.1%).

Large language models are now widely used for everyday learning, but the underlying interactions are typically unstructured chats rather than following a curriculum. Unlike formal online learning systems, these interactions carry no prior record of the student, so any estimate of what the student already knows must be inferred from the dialogue itself. We show that this gap is not closed by scaling models alone. Frontier and education-tuned LLMs perform poorly when asked to tutor a student over an extended session, because doing so requires three things at once. The tutor must sequence a curriculum, conduct Socratic dialogue, and infer the student's knowledge state from that dialogue. We propose separating these responsibilities. Given a student query, our system constructs a prerequisite knowledge graph in which subtopics are nodes and dependencies are edges, and frames tutoring as deciding which node to teach next and how many dialogue turns to spend on it before moving on. A lightweight PPO policy handles this sequencing decision, while an LLM conducts the Socratic exchange at the chosen node and returns a signal of student progress. Across held-out STEM and non-STEM topics, our PPO-paired tutor outperforms heuristic baselines, frontier general-purpose models, and a model specialised for Socratic dialogue: on both the rate at which students reach full curriculum mastery and the number of turns required. Explicit curriculum structure delivers gains that scaling the underlying model does not.

Vision-language-action (VLA) models provide strong visual, language, and action priors for robot manipulation, but visual observations alone often miss the local contact state required for contact-rich tasks. We present TacCoRL, a scalable framework that injects Tactile feedback into VLA policies and improves them through sim-real Co-training and simulation-based reinforcement learning (RL), without requiring large-scale tactile pretraining or extensive real-world contact exploration. The key idea is not only adding touch as an input, but learning how contact readings should modulate action responses in near-failure states that are rare in demonstrations and risky to collect on hardware. We use a real-aligned simulator as a closed-loop training environment for contact interaction. Mixed simulated and real trajectories first warm-start tactile-conditioned actions in the pretrained policy. Reinforcement learning with verifiable task rewards then optimizes the policy using simulated contact rollouts. It reinforces tactile-conditioned actions that lead to task completion, while a supervised objective on real trajectories keeps the refined policy anchored to deployment visual, tactile, and action distributions. The resulting policy transfers directly to the real robot without privileged simulation state or online real-world RL. Across four bimanual contact-rich tasks, the final visuo-tactile policy achieves an average success rate of 72.5%, compared to baseline of 50.0%. Result videos and more details are available at https://tac-corl.github.io/

We study whether grounded reasoning supervision from abundant 2D medical images can improve 3D medical VQA when both input types are aligned through a common reasoning interface. We introduce UniReason-Med, a single-checkpoint framework that processes either a 2D image or a slice-serialized 3D volume at inference time, generating interleaved textual reasoning and localized visual evidence through shared box syntax, region-token injection, and a common grounded reasoning policy. To train this interface, we construct UniMed-CoT, a 220K instruction-tuning dataset with interleaved textual reasoning and grounded visual evidence, including 170K 2D and 50K 3D samples. Through supervised fine-tuning followed by outcome-level reinforcement learning, UniReason-Med learns to generate grounded reasoning traces without IoU/Dice-based localization rewards during RL. Data-mixture and component ablations show that joint 2D+3D grounded supervision substantially improves 3D reasoning over 3D-only training, while grounding and region-token injection consistently benefit both 2D and 3D tasks. These results suggest that a shared grounded reasoning interface can transfer reasoning structure from 2D images to slice-serialized volumetric medical understanding. The code and data are publicly available at https://github.com/IQuestLab/unireason-med.

We introduce a multi-view in-cabin monitoring dataset for public transportation with synchronized RGB and depth images from four inward-facing cameras and a rotating LiDAR covering the vehicle interior of a digitalized and partly automated German city bus. The dataset contains 9.136 synchronized samples with annotations and is accompanied by a calibration and pseudo-labeling pipeline that generates 3D human pose estimates and oriented 3D bounding boxes for occupants. We further provide a nuScenes-format conversion and benchmark representative multi-view 3D detection models (e.g., Lift-Splat-Shoot and BEVFusion), supporting comparative evaluation and small-scale training of multi-view in-cabin perception models. The dataset and tools are available at https://github.com/EvgenyGorelik/multiview_incabin_dataset.

We study online estimation for high-dimensional generalized linear models with streaming data. First, for the non-distributed setting, we propose a gradient-enhanced surrogate loss that approximates the cumulative loss using only historical summaries, which modifies and improves upon the existing renewable estimation approach for the same model in the high-dimensional setting, and removes the batch-number constraint in previous studies. We then extend the method to distributed streaming data under the master-client architecture, where batches are partitioned across sites and only summaries (gradient vectors) are exchanged. Instead of directing applying the popular method of Jordan et al. (2019) to the surrogate quadratic loss, our adjusted approach does not require the clients to compute the full surrogate loss. We derive non-asymptotic error bounds under the high-dimensional scaling, without the stringent constraint on the number of batches in the previous studies. Simulation results under linear and logistic models, together with a real-data application, show improved accuracy over existing renewable estimators.

Close-in exoplanets exhibit a wide range of orbital architectures and physical properties shaped by both formation conditions and migration processes. Although population-synthesis models predict distinct planetary populations, establishing a quantitative connection between observed exoplanets and synthetic populations remains challenging. We investigate the intrinsic organisation of close-in exoplanets using physically motivated dynamical parameters and connect the resulting populations to pebble-accretion formation pathways. A two-stage Gaussian mixture model (GMM) is applied to an observed sample of close-in exoplanets, performing unsupervised probabilistic clustering in a feature space dominated by dynamical descriptors of planet-star interactions. The resulting clusters are mapped onto a pebble-accretion synthetic population within a statistically motivated three-dimensional parameter space. Formation-related quantities, including gas availability, gas fraction, and ice-rock mass ratio, are then used to interpret the mapped populations. We identify statistically supported sub-populations without imposing predefined classification boundaries, including very-massive gas giants, hot giants, warm-Jupiter-dominated systems, and lower-mass giants. The mapped synthetic populations reveal systematic differences in formation timing, gas accretion, and solid growth histories. In particular, very-massive gas giants are preferentially associated with earlier formation epochs than hot-giant and warm-Jupiter-dominated populations. These results demonstrate that physically motivated machine-learning approaches can provide a statistically robust framework for linking observed exoplanet populations to theoretical planet formation pathways.

State root computation dominates (78%) blockchain block processing time. Ethereum's canonical authenticated data structure, i.e., Merkle Patricia Trie (MPT), suffers from severe tree-height growth and is vulnerable to \textit{Nurgle attacks} (SP'24), where adversaries inflate path depth via hash collisions and degrade system performance at negligible cost. Existing defenses increase node fanout (span) to bound tree height, but higher span inflates proof size exponentially. Prior work mitigates this trade-off using vector commitments, at the cost of trusted setup or expensive verification. We present \textsc{Mhot}, a height-optimal authenticated data structure for blockchain state commitment that preserves standard hash-based verification without trusted setup. Unlike MPT's fixed-prefix indexing, which couples span and fanout exponentially, \textsc{Mhot} indexes by discriminative bits that actually distinguish keys, achieving adaptive span with linear fanout coupling and provably minimal height. To prevent high fanout from inflating proofs, we introduce hierarchical proofs, a two-layer Merkle construction that reduces per-node proof overhead from O(k) to O(log k). On Ethereum mainnet workloads, \textsc{Mhot} achieves up to 9X higher write throughput, 4X lower write amplification, and 2X smaller proofs than MPT. Under Nurgle attacks, even when the adversary consumes an entire block's gas budget, \textsc{Mhot} maintains a 0% attack success rate (v.s., 99.97% for MPT). Our results, somewhat surprisingly, show that height optimality (not new crypto primitives!) is the key abstraction for scalable and attack-resilient blockchain state commitment.

Theory of Mind (ToM) reasoning requires inferring agents' beliefs from partial and asymmetric observations, which remains an open challenge for LLMs. Existing prompting-based approaches improve ToM reasoning through observable-event filtering or temporal belief chains, without explicitly modeling nested beliefs. We introduce RecToM, an inference-time framework for ToM reasoning that models nested beliefs via recursive perspective construction. RecToM constructs each character perspective from the preceding character perspective along the character chain specified by the question, reducing higher-order belief questions to actual-world questions within the final constructed perspective. We further provide a KD45 analysis showing that RecToM's perspective construction induces a well-formed belief modality beyond simple event filtering. Experiments on ToM benchmarks, including Hi-ToM, Big-ToM, and FanToM, across multiple LLM backbones show that RecToM consistently outperforms recent advanced approaches, achieving state-of-the-art performance. Notably, RecToM reaches 100\% accuracy on Hi-ToM with GPT-5.4 and Qwen3.5, a benchmark requiring higher-order ToM reasoning.

Finding interpretable directions in language-model representations is critical for understanding and controlling model behavior. Sparse autoencoders (SAEs) have become the standard tool for this purpose, but using them as the default first lens often requires training, storing, and evaluating large overcomplete dictionaries. This bottleneck limits rapid exploration and raises a fundamental question: how much interpretable structure is already visible from activation geometry before training another neural dictionary? Our intuition is simple: many interpretable directions are selective on tokens, and these directions should look less Gaussian than random directions. We therefore revisit independent component analysis (ICA), a classical method for finding non-Gaussian directions, as a compact lens for language-model interpretability. We find that ICA has been underestimated for LLM interpretability, because prior uses often relied on off-the-shelf ICA implementations that are brittle on LLM activations and lacked systematic tools for inspecting and evaluating the recovered directions. To bridge these gaps, we introduce ICALens, the first practical workflow for stable, efficient, and auditable ICA analysis of LLM representations. It combines an optimized GPU-parallel FastICA pipeline with LLM-specific stability recipes and better fitting diagnostics, enabling efficient and reliable layer-wise analysis. Across GPT-2 Small, Gemma 2 2B, and Qwen 3.5 2B Base, ICALens efficiently recovers compact, human-interpretable directions without per-layer gradient-based dictionary training. On SAEBench, ICA is competitive with public SAEs in sparse probing and outperforms them in targeted probe perturbation under small-to-medium budgets. These results suggest that ICA should not be viewed as a weak baseline, but as an efficient and complementary first lens for exploring language-model representations.

Spatial reasoning remains a persistent challenge for multimodal large language models (MLLMs). Existing approaches largely rely on large-scale, statically curated datasets, where all training samples are treated uniformly regardless of the model's evolving capabilities. This static paradigm is inherently data-inefficient: training capacity is often spent on samples that are either trivial or overly difficult for the model at its current stage. To address this limitation, we propose Ouroboros-Spatial, a self-evolving training framework in which the model plays dual roles as a proposer and a solver. In each iteration, a frozen proposer generates spatial question-answer (QA) pairs from 3D scene metadata and raw video frames, together with executable code for deriving reliable ground truth. A learnable solver is then fine-tuned on the accepted samples, and its per-sample prediction confidence is used as a difficulty signal. This signal is fed back to the proposer in the next iteration, guiding it to generate questions better matched to the solver's current capabilities. Through this closed-loop design, the training distribution co-evolves with model ability, reducing redundant trivial examples while filtering out ambiguous or uninformative samples with limited learning value. Across six spatial reasoning benchmarks, Ouroboros-Spatial substantially improves Qwen3-VL-4B and Qwen3-VL-8B while using an order of magnitude fewer training examples than recent large-scale curated datasets. On VSI-Bench, it yields absolute gains of 9.9 and 6.8 points for the 4B and 8B models, respectively, enabling both to outperform a wide range of strong open-source and proprietary baselines.

Multi-chiplet GPUs scale compute throughput and high-bandwidth memory (HBM) capacity, but their non-uniform memory system makes locality between chiplets and their data critical to the GPU's performance and energy efficiency. Locality-aware scheduling and data placement identify which data should reside near each chiplet. However, in general matrix multiplication (GEMM), locality-aware data placement often becomes incompatible with a fixed page-granularity data interleaving, since the optimal granularity for mapping data across chiplets varies widely across workloads. We propose Chiplet-Contiguous Layout, a global memory layout that stores chiplet-local data contiguously. Chiplet-Contiguous Layout enables locality-aware placement compatible with page-granularity placement across diverse large language model (LLM) GEMM shapes, without changes to the operating system or hardware. On representative LLM inference and training GEMMs from Qwen 3 30B and Llama 3.1 70B, Chiplet-Contiguous Layout on average reduces remote HBM traffic by 24.7x on Qwen and 19.2x on Llama over 4KB interleaving, and by 4.1x and 2.1x over coarse locality-aware placement.

Multi-chiplet GPUs split memory into local and remote HBM regions across a silicon interposer, and reducing the remote HBM traffic is crucial for the performance and energy efficiency of multi-chiplet GPUs. For general matrix multiplication (GEMM), the dominant operator in large language models (LLMs), the resulting inter-chiplet traffic depends strongly on kernel choices such as operand layout, CTA traversal order, and data placement, and the optimal strategy to minimize remote accesses is nontrivial. We present a fast, functional, tile-level locality simulator that models CTA scheduling, per-chiplet L2 caches, and local/remote HBM accesses to evaluate a full-size LLM GEMM configuration. Across representative LLM GEMMs, the simulator shows that remote traffic varies by up to 90x across the design space for the same GEMM dimensions. Moreover, using the simulator as feedback, an agentic AI discovers that a 2D block-swizzle CTA traversal reduces remote traffic over the best 1D traversal by up to 5.1x under round-robin placement, identifying CTA traversal order as a first-order, GEMM-dependent design knob for inter-chiplet traffic.

User-side memory in LLMs is typically scored as a single "personalization" capability: given a user's history, is the output more user-aware? We show this aggregate metric hides opposite-direction failures. Memory factorises into at least three orthogonal axes -- behavioral consistency (style, voice), factual presence (recall facts in history), and factual absence (abstain when a fact is absent) -- and no single substrate wins all three. Comparing per-user gamma-LoRA (a small LoRA adapter trained on each user's history; gamma denotes per-user, not per-task) against BGE-large dense top-K retrieval on a controlled 50-user synthetic corpus and a real-data probe (LaMP-3), we find gamma-LoRA decisively wins behavioral style while RAG decisively wins factual absence -- and the same query-projection cells in attention layers 21-35 causally load-bear both effects in opposite directions (zeroing those LoRA weights raises absence-probe TPR by +33 pp and drops presence-probe TPR by 20 pp). On the more heavily RLHF-tuned Llama-3.1-8B-Instruct the asymmetry strengthens, not heals: parametric memory's behavioral advantage collapses while its absence-calibration deficit against retrieval widens -- an alignment tax on parametric user-memory. On real-data LaMP-3, gamma-LoRA underperforms a majority baseline; a 9-condition mitigation sweep diagnoses this as instruction-following collapse, not substrate failure (a 9x2 cross-product shows the eval-time {1..5} logit mask drives main_acc to >=0.995 on every recipe), and the best training-time fix replicates bit-identically on Llama. Finally, substrate-selection routing is question-classification, not calibration: a 110M DistilBERT on the question text alone beats every logit-based router. We contribute the diagnostic framework, the diagnosed real-data negative, the alignment-tax replication, and the routing-as-classification finding.

Online learning with delayed feedback typically assumes that the learner can track all pending rounds until their feedback arrives. In practice, tracking resources are finite, and feedback from untracked rounds is permanently lost. In this paper, we study delayed online convex optimization (OCO) under a hard capacity constraint, where at most $C$ pending rounds can be tracked at any time. To model delay information, we introduce a semi-clairvoyant model that refines the clairvoyant assumption from prior work: rather than requiring delays to be known at prediction time, the learner observes delay expirations online, consistent with the classical unconstrained delayed setting. Our approach proceeds via a reduction to a novel ``delayed and weighted'' OCO problem, using a scheduler that randomizes tracking decisions and importance-weights the resulting observations. For this base problem, we propose and analyze Delayed-Weighted FTRL and its bandit analogue, establishing regret bounds that explicitly characterize the interaction between time-varying weights and delayed feedback. Combining these base learners with our schedulers yields the first regret guarantees for capacity-constrained OCO under convex and strongly convex losses, for both first-order and bandit feedback. For first-order feedback, capacity $C = Ω(\log T)$ suffices to recover standard delayed OCO rates up to logarithmic factors. For bandit feedback, the regret rates are modulated by powers of $(1 + σ_{\text{max}}/C)$, where $σ_{\text{max}}$ is the maximum number of pending observations at any time. This allows the regret bound to degrade gracefully when $C < σ_{\text{max}}$, while remaining sublinear.

On-policy self-distillation (OPSD) provides dense, token-level supervision for reasoning models by aligning a model's own distribution with the distribution it produces under privileged context, typically a verified solution. However, we show that the learning signal drawn from this distributional gap concentrates on style tokens rather than task-bearing ones, as the hinted model tends to produce more direct, shorter outputs. We term this pathology \emph{privilege-induced style drift}, which destabilizes training or causes response length to shrink. To address this, we propose \textbf{RLCSD} (Reinforcement Learning with Contrastive on-policy Self-Distillation), which mitigates this drift by contrasting the teacher-student gap under a correct hint against that under a wrong hint, suppressing the style shift that conditioning on a hint tends to induce regardless of correctness, and yielding a signal that is more concentrated on task-bearing tokens. Experiments on Qwen3 (1.7B/4B/8B) and Olmo-3-7B-Think across mathematical and logical reasoning show that RLCSD consistently outperforms GRPO and prior OPSD methods. We further show that the contrastive principle is general: it plugs into existing OPSD methods to improve them, and its underlying insight extends to the broader cross-model on-policy distillation setting.

To make clinically grounded decisions, medical AI agents are expected to go beyond simple recognition and be capable of tool retrieval, evidence acquisition, and integration. Existing benchmarks largely evaluate isolated perception or single-turn question answering, and therefore provide limited visibility into failures of planning, tool recruitment, and rollout reliability. We introduce MedCTA, a benchmark for evaluating medical tool agents on clinician-validated, step-implicit tasks grounded in realistic multimodal clinical inputs, including radiology images, pathology slides, and reports. MedCTA comprises 107 real-world clinical tasks with clinician-verified executable trajectories over 5 deployed tools, and supports process-aware evaluation of tool selection, argument validity, execution stability, trajectory fidelity, and outcome quality. We benchmark 18 open- and closed-source multimodal models and find that even frontier systems remain brittle in multi-step clinical tool use: autonomous rollouts are dominated by protocol failures, premature stopping, and incorrect tool recruitment, while gold-standard tool routing yields large but still incomplete gains. These results show that strong backbone perception does not translate into reliable agentic behavior in clinical settings. MedCTA provides a rigorous testbed for auditing, diagnosing, and advancing trustworthy medical AI agents. The dataset and evaluation suite are available at https://ivul-kaust.github.io/MedCTA/

The performance of machine learning and deep learning models largely depends on the quality of the training data. However, the quality of the real-world datasets is often compromised by noisy labels, which can substantially degrade model accuracy and reliability. To address this challenge, we propose Relabeler, an end-to-end data-centric framework for detecting and correcting corrupted labels. For corrupted label detection, Relabeler jointly leverages both local and global relationships among data instances to identify potentially noisy samples. After detecting suspicious instances, Relabeler further performs label correction by estimating the most probable clean label for each instance based on both its input features and observed noisy label. Extensive experiments across multiple datasets, noise types, and noise rates demonstrate that Relabeler consistently outperforms state-of-the-art baselines, achieving up to 58% improvement in label correction precision and 6% improvement in downstream task performance.

Model watermarking safeguards AI model intellectual property by embedding distinctive knowledge that induces unique behavioral signatures. The primary technical challenge lies in ensuring watermark robustness against various post-processing attacks on the watermarked model. Model extraction attacks emerge as the most severe threat, where adversaries exploit prediction outputs to train surrogate models that illegally replicate the original model's functionality. In this work, we propose a rehearsal-based watermark embedding framework to enhance the robustness of model watermarks against model extraction attacks. By simulating the extraction process, our method leverages the loss of a \textit{simulated stolen model} on a trigger set as a training signal to fine-tune the watermark knowledge within the target model. This fine-tuning step encourages the watermark to be embedded in a way that boosts transferability, thereby increasing its chances of persisting and remaining detectable in stolen models. Comprehensive experiments conducted under diverse settings demonstrate that the proposed method significantly improves the robustness of model watermarks against both model extraction and subsequent watermark removal attacks.

High-quality labeled data is essential for training reliable ML/DL models. However, real-world datasets often contain a considerable proportion of corrupted labels, which can severely degrade model performance. To address this problem, we propose CANOLA, a novel framework for correcting corrupted labels through noise-aware learning and iterative label refinement. CANOLA explicitly estimates the underlying noise distribution of the dataset and incorporates this information into the training of a noise-aware Deep Neural Network. By incorporating noise characteristics during learning, CANOLA enables the model to down-weight unreliable supervision signals and focus on trustworthy patterns, thereby improving robustness and generalization. Label correction is performed via cautious, iterative soft label refinement, in which model predictions are blended with observed labels to prevent premature or erroneous updates. This progressive refinement allows the dataset to be repaired in a stable and controlled manner. We evaluate CANOLA on six widely used datasets under realistic noisy labeling scenarios. Experimental results show that CANOLA consistently outperforms SOTA label correction methods, achieving relative improvements ranging from 19% to 52% in error reduction. Moreover, models trained on datasets corrected by CANOLA obtain substantial downstream performance gains. Even simple classifiers trained on CANOLA's corrected data can outperform complex model-centric approaches by margins of up to 67%.

Deliberative polling promises to improve collective decision-making by exposing shareholders to a broad range of arguments before they vote. Yet ensuring that every voter encounters a representative sample of the reason space, the coverage problem, remains an open challenge, particularly at scale and in adversarial or strategically motivated electorates. This paper introduces a way of evaluating solutions using the LLM-based Agentic Bipolar Argumentation Simulator, grounded in a framework which formalises a poll as a six-tuple <Jend, Jopp, Ratt, Renh, VA, VR> of endorsing and opposing justifications, attack and enhance relations, and shareholder- and relation-weights. ABAS simulates N autonomous shareholder agents, each assigned a latent opinion according to desired distributions in [-1, 1], who sequentially vote, choose or author justifications, and optionally submit argumentation-graph links. The simulator implements recommendations that rank existing justifications by their observable endorsement mass. It evaluates the mechanism's success by coverage, namely the fraction of the corpus reason-tag set represented in the K recommendations presented to each shareholder, as a solution to the NP-hard Subsuming Justification Problem. Reported experiments characterise how creativity rate (pown), recommendation size (K), argumentation density (plinks), and population size (N) affect coverage and corpus diversity. In an authenticated electorate where Sybil attacks are impossible and only the relation graph is gameable, we stress-test the scoring with coordinated strategic voting attacks: a tag-flood attack collapses coverage, while author-count relation weighting through a reversed-PageRank rule resists the flood markedly better than uniform weights.

Latent diffusion and flow matching have emerged as leading approaches for synthetic turbulence generation, yet they systematically under-represent dissipation-range amplitudes. We introduce a latent flow matching framework with a spectrally regularized compression stage that directly targets this failure mode. On a 256^2 DNS dataset at Re_f \approx 2250, replacing an MSE-trained VAE with a zone-weighted log-spectral objective raises deep-dissipation retained spectral power from 25% to 94% in reconstruction and from 20% to 79% in unconditional generation. The improved latent representation also yields a substantially better sampling cost-fidelity tradeoff: the MSE-trained latent space imposes a fundamental quality ceiling near DD bias -0.70 that no integrator or step-count can overcome, while the spectrally regularized latent space reaches DD bias -0.117 at just 20 function evaluations. Mechanistically, encoder-decoder swap experiments show that the improvement is driven primarily by encoder-induced latent reorganization rather than decoder capacity, while a support-amplitude decomposition reveals that MSE-trained models behave as conservative suppression models, minimizing pointwise error by attenuating intermittent high-wavenumber structure. Both pipelines recover the second-order structure function and the correct sign of S_3, indicating the correct cascade direction without explicit supervision. A small residual gap in the magnitude of S_3 suggests that phase-coherent triadic organization remains a complementary axis to amplitude fidelity for future generative turbulence models.

Every public LLM cost calculator we surveyed treats GPU utilization as a fixed input -- entered by the user, baked in as a preset, or silently assumed at 100% -- never measured against the operator's actual load. We show that this assumption is the dominant source of error: on identical H100 hardware, effective cost spans \$0.21 to \$15.25 per million output tokens, an underutilization penalty of 2.5-24x across low-to-moderate enterprise loads (1-10 rps) and up to 36.3x near idle -- driven by one operator-controlled variable, offered request rate lambda, which sets in-flight concurrency via Little's Law and which no open-source calculator exposes. Because calculators take utilization as a user-supplied input, any utilization-naive estimate understates true cost by exactly 1/U, systematically mispricing self-hosting -- most severely over-selling it for low-traffic workloads. We propose a measurement methodology that parameterizes the relationship as C_eff = f(H, M, Q, lambda, L), validate it with 42 benchmarks across dense, ultra-sparse MoE, and sparse MoE models, and release vllm-cost-meter, an open-source cost meter that attaches to a live vLLM server and reports real \$/M-tokens against the operator's own traffic. We further show that FP8 quantization benefits the MoE architectures we tested roughly 2.2-2.4x more than the dense model (+69 to +74% vs. +31% peak throughput; n=3, broader validation needed), and our data are consistent with active parameter count, not total model size, being a primary predictor of saturation economics. To rule out single-hardware confounding we repeat the core sweep on A100 80GB PCIe (56 runs): the load-driven spread reproduces at 7.0-11.4x, the active-parameters ordering survives at FP8, and the dense-FP8 advantage inverts on silicon without native FP8 tensor cores -- a hardware-conditional caveat the framework already accommodates.

Long-horizon LLM agents are not trusted to run unattended: with no human watching, they confidently report success they never verified. We treat honesty -- bounding what an agent may claim at termination -- as a first-class metric for unattended autonomy, distinct from capability. We present Autopilot, an execution model that makes silent fabricated success structurally impossible rather than merely rarer. Autopilot externalizes all working state into a durable, gated finite-state machine that a scheduler advances one stateless tick at a time; a hard floor forbids any terminal "done" claim whose falsifiable gate did not actually execute and pass. We prove a No-False-Success theorem -- under gate soundness, floor enforcement, and plan coverage, termination implies the goal holds -- whose only trust points are empirically measurable, and show the worst case degrades to an honest stall, never a fabricated success. Because each tick rehydrates only the state machine, per-step context cost is constant in the horizon. Across a 3,150-cell paired corpus (70 tasks $\times$ 3 systems $\times$ 3 models $\times$ 5 seeds, including 50 SWE-bench Lite tasks across 11 OSS repos), Autopilot fabricates on 0.95% of cells [95% CI 0.38--1.62] while Reflexion and StateFlow baselines fabricate on 8.10% [6.48--9.81] and 25.05% [22.48--27.62] respectively. The headline contrast lives in the hard regime: on SWE-bench Lite, the firewall reduces fabrication from 33.7% (StateFlow) to 0.67%, a paired difference of $-33.07$ pp [95% CI $-36.53, -29.73$]. The mechanism is the gate, not the model: all ten Autopilot fabrications come from the strongest model, while two weaker mid-tier models never fabricate across 700 paired cells. The firewall trades coverage for honesty by design -- an honest stall is recoverable; a confident wrong output shipped downstream is not.

Unmanned Aerial Vehicle (UAV) threats have emerged as a defining security challenge of the 21st century. This paper presents DroneShield-AI, a unified open framework integrating six processing layers: RF signal classification, acoustic motor-signature detection, YOLOv8-based visual detection, evidence-weighted sensor fusion, a Behavioral Intent Classification Engine (BICE), and a Graph Neural Network Swarm Intelligence Module (GNN-SIM). BICE introduces the first systematic six-class threat taxonomy for drone flight patterns, enabling predictive operator alerts with a 30-second advance-warning horizon. GNN-SIM is the first open framework for adversarial multi-drone formation analysis using Graph Attention Networks. Evaluated on three publicly available real-world datasets, the fused pipeline achieves 96.1% detection accuracy, 3.2% false alarm rate, AUC-ROC: 0.981, and 142ms end-to-end latency on commodity CPU-class hardware at approximately $500-$780 USD total system cost. All code, model weights, and simulation datasets are publicly released at submission.

End-to-end task-success is the dominant way to evaluate LLM agents, but one aggregate number tells you that an agent regressed, not where. We present layer-isolated evaluation: a deployed ordering agent is decomposed into a fixed taxonomy of layers (ontology, intent, routing, decomposition, escalation, safety, memory, and cross-cutting envelope/defense), each exercised by its own assertion slice in a deterministic, no-LLM "pure" mode. The pure suite (238 cases across 23 slices; 225 run in 2.39 s, ~10 ms/case) runs in CI on every change against a locked per-slice baseline. We validate by controlled regression injection, degrading one layer at a time across seven non-safety layers. The effect we did not design in is masking: the aggregate pass-rate barely moves (-1.7 to -5.9 pp for six local regressions), while the matching slice craters (-25 to -91 pp). A layer's slice reacting to its own fault is partly by construction; the measured results are (i) the aggregate masking and (ii) that damage stays off the other slices: the injected layer's slice is the single worst-hit in 5 of 7 cases and top-3 in 7 of 7 (mean rank 1.29 of 19). Localization replicates on a second, structurally different tenant (Starbucks SG): all seven matching slices crater, so it is not a single-catalog artifact. We position it as a concrete, deterministic instantiation of the component-level evaluation EDDOps prescribes but leaves unimplemented, with CheckList as ancestor and as the deterministic mirror image of whole-workflow stochastic mutation testing. Our contributions: (a) a fully decomposed, sub-second, no-LLM per-layer harness for a production agent, (b) a coverage-honesty test-adequacy criterion that refuses to score an unexercised layer, and (c) the regression-injection demonstration that per-slice baseline-locked gates localize regressions an aggregate metric masks.

Spatial reasoning from egocentric videos is inherently challenging because the observable evidence is constrained by the camera trajectory. Existing methods rely on single-turn inference, forcing models to resolve geometric ambiguity through semantic priors rather than verifiable evidence. We argue that spatial reasoning should be revisitable: conclusions formed under limited evidence should remain open to revision when complementary viewpoints become available. Building on this insight, we propose Reason, then Re-reason (ReRe), a training-free, inference-time framework with two phases: in the Reason Phase, an MLLM forms a spatial hypothesis from the original video; in the Re-reason Phase, it verifies or revises the hypothesis by observing a synthesized novel-view video. To enable effective cross-view revisiting, we design a Geometry-to-Video pipeline that renders strategically complementary novel views from predicted 3D geometry. These views feature an elevated, oblique perspective with scene-spanning coverage, while preserving the MLLM's native video interface without architectural modifications. Extensive evaluations on VSI-Bench and STI-Bench demonstrate that ReRe substantially boosts open-source MLLMs to rival proprietary state-of-the-art performance. Project page: https://zhenjiemao.github.io/ReRe/

Tabular-image multimodal learning aims to improve predictive modeling by jointly using structured tabular attributes and visual data. Although pretrained encoders provide strong modality-specific representations, full fine-tuning can be computationally expensive, while keeping encoders frozen may limit task-specific adaptation. We propose the Tabular-Image Adapter (TI-Adapter), a modality-specific adapter-based fine-tuning framework for efficient multimodal adaptation. TI-Adapter freezes the pretrained tabular encoder and learns an adapter after the extracted tabular embedding, while adapting the image branch with embedding-level and bottleneck-level adapters instead of full fine-tuning. Experiments on 20 tabular-image datasets show that TI-Adapter achieves competitive or better predictive performance than full fine-tuning while using substantially fewer trainable parameters. Ablation studies further demonstrate the importance of adapter placement for balancing performance and practical efficiency.

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.

Large language model (LLM) agents struggle with long-horizon tasks due to their inherent statelessness, requiring all task-relevant information to be encoded in growing input contexts. The resulting degraded reasoning quality, increased inference cost, and higher latency necessitate efficient working memory mechanisms. However, existing approaches either rely on lossy compression or similarity-based retrieval, which often fail to capture temporal structure and causal dependencies required for multi-step agentic tasks. In this work, we present HORMA, a Hierarchical Organize-and-Retrieve Memory Agent that organizes experience into a file-system-like hierarchical structure, where summarized entities are linked to the corresponding raw trajectories, enabling efficient access without losing detailed information. HORMA decomposes working memory into two stages: structured memory construction and navigation-based retrieval. The construction module iteratively refines how experiences are structured by distinguishing between failures caused by missing information and those caused by misleading or overloaded context. The navigation module retrieves task-relevant context by traversing the hierarchy using a lightweight agent trained with reinforcement learning to select minimal yet sufficient context, thereby reducing latency along the critical execution path. Across ALFWorld, LoCoMo, and LongMemEval, HORMA improves task performance under constrained context budgets while requiring at most 22.17% of the baseline token usage in long conversation tasks. Compared to existing methods, it consistently achieves better efficiency-performance trade-offs and generalizes effectively to unseen tasks.

Problem, Research Strategy, and Findings: The rise of large language models (LLMs) raises a key question for urban planning: which forms of professional planning knowledge can AI replicate, and which still require human judgment? Although AI tools are increasingly used in planning practice, there is still no systematic framework for testing whether they can reason with the contextual sensitivity, value awareness, and institutional literacy central to planning expertise. This paper introduces Urban Planning Bench (UPBench), a domain-specific evaluation framework that assesses LLM reasoning through a 4x5 matrix of four knowledge pillars and five cognitive levels adapted from Bloom's revised taxonomy. Evaluating 25 LLMs with automated scoring and expert review, we find a non-monotonic cognitive curve: models perform better on higher-order analytical tasks than on factual recall and integrative judgment. This suggests that planning knowledge often treated as lower-order is deeply shaped by institutional, jurisdictional, and temporal context, making it hard for LLMs to generalize. We summarize these limits as four epistemic diagnostics: regulatory hallucination, conceptual conflation, wickedness paralysis, and phronetic deficit. Takeaway for Practice: The findings support differential delegation in planning. LLMs can assist with cross-disciplinary synthesis, literature review, scenario generation, and preliminary policy analysis. However, they remain unreliable for jurisdiction-specific regulation, normative conflict resolution, and context-sensitive procedure. Agencies should require verification for AI-assisted regulatory analysis, while planning education should emphasize institutional literacy, normative judgment, and contextual sensitivity.

Traditional wildfire models rely on rigid, low-dimensional parameters and static fuel maps, frequently underpredicting fire spread. To address this weakness, we introduce a hybrid deep-learning parameterized Probabilistic Cellular Automata (CA) framework implemented in JAX. Our approach employs a Multi-Scale Convolutional Neural Network to dynamically generate spatially varying parameters that govern fire-spread probability, wind alignment, and slope influence. This hybrid design captures complex, nonlinear environmental interactions while preserving the physical interpretability of the underlying three-state CA. The JAX implementation enables hardware acceleration and gradient-based parameter calibration. Evaluated on six large-scale wildfires in the western United States, the model maintains IoU > 0.6 over 72-hour forecast horizons after a 10-day data assimilation window during which the model is fitted incrementally to observed perimeters; the resulting forecast is a conditional projection of fire growth under the suppression regime already ncoded in those observations.

Diagnosing pulmonary diseases requires integrating heterogeneous evidence amid phenotypic variability and cross-disease overlap. Although large language models (LLMs) have shown progress on pulmonary knowledge question answering (QA) and information-processing tasks, reliable pulmonary diagnosis requires patient-specific, relation-aware reasoning over electronic medical record (EMR) evidence rather than isolated knowledge recall. We define this gap between pulmonary knowledge and case-level diagnostic reasoning as the Pulmonary Knowledge-to-Diagnosis Gap. To address it, we introduce LungKG, the first structured pulmonary knowledge graph for diagnostic knowledge organization and record-grounded reasoning. LungKG contains 59,038 nodes and 164,308 edges across 15 entity types and 112 relation types, serving as both a reusable pulmonary knowledge resource and the foundation for LungKG-guided model adaptation. Built on LungKG, we propose Lung-R1, a LungKG-guided pulmonary LLM trained through KG-constrained reasoning-chain construction and KG-guided reinforcement learning. In a 20-system evaluation, Lung-R1-14B achieves state-of-the-art performance across Choice, Pulmonary-QA, and EMR Diagnosis, reaching an EMR Diagnosis score of 4.3583 and surpassing the strongest non-Lung-R1 baseline by 0.1476 points. These results demonstrate the value of LungKG-guided training for EMR-based pulmonary diagnosis.

We present SpAArSIST, a deployment-oriented refinement of the widely used AASIST graph pooling backend for self-supervised learning (SSL) based anti-spoofing. Motivated by redundant operations in public implementations, we replace learned pooling and stack-node attention with explicit, lightweight choices: separate train and inference graph pooling ratios $(k_{\mathrm{tr}},k_{\mathrm{inf}})$, magnitude-based node scoring, and mean aggregation of graph nodes. The best overall configuration (rank 1) cuts backend compute by 20.7% (195.045M $\rightarrow$ 154.706M MACs) and model size by 4.1% (611.8k $\rightarrow$ 586.4k params), while improving out-of-domain robustness on In-the-Wild to 2.82% EER and 0.078 minDCF (from 4.64% and 0.133) and remaining competitive on ASVspoof5. We further provide a composite selection score that summarizes accuracy, calibration, and compute to support balanced deployment-oriented model choice.

Standard dot-product self-attention computes, in a single layer, only pairwise (order-2) interactions between tokens; representing a generic order-$k$ interaction is known to require either super-quadratic resources in one layer or composition across depth. We introduce \textbf{Quantum Higher-Order Attention (QHA)}, a shallow, hardware-realizable quantum attention head that, via data re-uploading and an all-to-all non-Clifford entangler, synthesizes order-$k$ token interactions inside the circuit and exposes them through a local single-qubit read-out. We prove (i) an expressivity separation: any single standard self-attention layer with embedding dimension $m$, $H$ heads and $p$-bit precision satisfying $mHp=o(N/\log\log N)$ cannot represent the order-$k$ correlation family that one QHA head represents with circuit depth $O(\log k)$ ($O(k)$ two-qubit gates); and (ii) a trainability guarantee for its local-design instantiation: with a local read-out and $O(\log n)$ depth the gradient variance is $Ω(1/\mathrm{poly}(n))$ (no barren plateau), which we confirm empirically -- while being explicit that the more expressive all-to-all instantiation we benchmark is trained empirically and shows exponentially decaying gradients. Empirically, at a $6.5\times$ smaller parameter budget, QHA generalizes hidden-subset parity of every order $k\le6$ from disjoint inputs, whereas the larger classical attention head collapses past order~2; consistent with theory, the size of the advantage tracks the target's Fourier degree - largest for parity and shrinking when low-order structure is present. As an application, QHA serves as a compact high-order interaction detector across three domains - genetic epistasis, learning-parity-with-noise, and graph triangle detection - reaching the noise ceiling at the smallest parameter budget where field-standard linear methods fail.

This paper explores the value of agentic AI tools for cybersecurity purposes. We evaluate the efficacy of a general-purpose GenAI Large Language Model- (GenAI-) based agent when powered by three different Ollama-hosted general-purpose open source models. We assess each agent's performance using precision, recall, false positive count, and a calculated composite score based upon the interplay of the captured metrics, against the baseline performance of an existing, vetted Static Application Security Testing (SAST) tool, Bandit. Our findings refute the notion that a modern open-source GenAI LLM-based agent is currently suitable for the specialized task of SAST scanning under realistic conditions.

Agent skills let LLM agents reuse instructions, resources, tools, and workflows, but they also create a new place for malicious behavior to hide. A skill may look benign in its documentation or code while becoming harmful only when it is invoked with particular user requests, local assets, persistent state, or multi-step tool interactions. This makes purely static vetting brittle. We present Runtime Skill Audit (RSA), a dynamic analysis method that audits skills by asking what the skill-mediated agent actually does under targeted runtime conditions. Instead of testing every skill with the same generic tasks, RSA profiles risk-relevant interfaces, prepares the execution context needed to exercise them, and assigns security labels from the resulting trace evidence. We instantiate RSA on OpenClaw and evaluate it on 100 skills against representative static baselines. RSA achieves 90.0\% accuracy with an 88.0\% true positive rate and an 8.0\% false positive rate, improving accuracy by 13.0 percentage points over the best static baseline. Under self-evolving attacks, static detectors collapse after one or two rounds, while RSA continues to detect 19--20 out of 20 malicious skills across rounds.

Subject-preserving video generation is not solved by frontal-face similarity alone: a generated person must remain recognizable across motion, large viewpoint changes, expression shifts, occlusion, scale variation, and conflicts among text, first-frame, and identity references. We argue that the central bottleneck is the point-reference paradigm, which collapses identity into a single static observation entangled with pose, accessories, lighting, background, and camera statistics. We introduce Argus, a Wan-based framework centered on Stacked Multi-View Identity Mosaic Injection (SMII). SMII converts MLLM-selected image/video identity evidence into a 3*3 stacked mosaic, synchronizes the mosaic with the current diffusion time, and injects it as negative-time read-only memory in Wan's native token space. This turns identity from an external clean adapter or a single reference image into a compact dynamic distribution. Around SMII, an MLLM Identity Director selects informative identity moments and resolves condition conflicts, while no-cross-pair counterfactual training, Temporal Identity Annealing, and Adaptive Self-Likeness Guidance improve robustness without paired subject-video supervision. We further release HardID-Celeb, a public-figure identity-stress benchmark, and introduce YawScore and OccScore to probe large-yaw and first-frame-occlusion robustness. Argus achieves state-of-the-art results on OpenS2V-Eval Human-Domain, reaching 64.38 Total Score, 71.86 FaceSim, 51.62 NexusScore, and 79.14 NaturalScore. On HardID-Celeb, Argus obtains 76.80 FaceSim and improves YawScore and OccScore by 12.60 and 15.10 points over the strongest baselines, demonstrating that dynamic identity memory and large-scale counterfactual self-supervision are highly effective for subject-preserving video generation.

Accurate salary prediction is critical for bridging the information gap between employers and job seekers in modern labor markets. Existing approaches predominantly yield a single point estimate and treat job attributes such as location, occupation, and industry as independent categorical features, ignoring both the inherent uncertainty and multi-modality of real-world compensation data and the rich hierarchical and semantic-similarity relationships that govern pay norms. In this paper we propose GAT-MDN, a unified framework that addresses both limitations simultaneously. For each of the three attribute domains we construct a domain-specific graph whose edges encode (i) hierarchical parent-child containment and (ii) weighted similarity links derived from a pre-trained Sentence-Transformer. Parallel Graph Attention Networks (GATs) with edge-feature-aware attention learn rich, context-sensitive node representations from these multi-relational graphs. A priority-based hierarchical selection module then assembles a composite feature vector that gracefully handles missing or coarse attributes, and a Mixture Density Network (MDN) head maps this vector to the parameters of a Gaussian Mixture Model (GMM), yielding a full conditional salary distribution. Extensive experiments on a real-world Dutch job-posting dataset of over 1 million records demonstrate that GAT-MDN significantly outperforms a non-graph MLP-MDN baseline in both Negative Log-Likelihood (NLL) and Mean Squared Error (MSE).

Deep search requires agents to answer complex questions through multi-step web search, browsing, evidence comparison, and synthesis. A central challenge is deciding how to search when several directions look plausible but only some will later lead to reliable evidence. If an agent greedily follows the current best-looking direction, it may keep extending a weak continuation. If it explores without discipline, it may waste budget on disconnected trials. We propose TreeSeeker, an inference-time framework for controlled trial-and-error in deep search. TreeSeeker organizes search as branch-and-return search over tree-structured states, where each branch is a tentative direction for a sub-goal. At each round, TreeSearch reads all sub-goal trees, identifies active goals, and uses textual UCB signals of value, uncertainty, and risk to select among exploiting a promising branch, exploring an uncertain alternative, or pruning an unproductive continuation and returning to an earlier branch point. TreeMem supports this control loop by keeping evidence, uncertainty, conflicts, progress, and failure cues attached to the branches that produced them, so trial outcomes can guide later decisions. Experiments on XBench-DeepSearch, BrowseComp, and BrowseComp-ZH show that TreeSeeker consistently outperforms strong open-source baselines, suggesting that explicit branch-and-return control complements stronger reasoning and tool execution.

Clothes-changing person re-identification (CC-ReID) aims to recognize individuals despite drastic appearance changes caused by clothing variation. While existing methods rely on adversarial learning to disentangle clothing features, we propose Ortho-ReID, which explicitly models a low-rank clothing subspace from VLM text descriptions and extracts clothing-invariant representations via direct geometric constraints. A critical component is our transformer-based Basis Maker, which refines a shared, low-dimensional clothing prior into an instance-adaptive low-rank subspace through cross-attention with image patches, enabling robust clothing feature extraction even under varying visibility conditions. This instance-adaptive subspace is supervised via alignment with clothing text embeddings, while identity features are extracted via a learnable projection head and geometrically constrained to be strictly orthogonal to it. Extensive experiments demonstrate state-of-the-art performance on PRCC (+5.9% top-1), Celeb-reID-light (+3.5%), and LaST (+5.3%), with competitive results on LTCC.

Data attribution is a promising field in interpretability that aims to explain model behavior through the influence of its training data, with applications including debugging undesirable model behavior and training dataset curation. However, significant engineering effort is required to perform it at scale, and many cutting edge techniques lack open-source tooling and support. Bergson is an open source library that aims to enable faster progress in the field by providing a host of techniques that scale to very large language models and pre-training datasets. The library natively supports on-disk gradient stores and multi-node distributed training, and provides quality of life tools for researchers. Finally, we introduce the first open-source implementations of three leading data attribution methods: MAGIC, SOURCE, and TrackStar. The library is available at https://github.com/EleutherAI/bergson .

Generative AI emulators are increasingly used in scientific domains where we already have strong theory, benchmarks, and physical intuition. This raises a central evaluation and interpretability question: when a foundation-style model can reproduce known continuum dynamics, what internal mechanism supports that behavior, is the internal behaviour consistent with known physics, and how does it relate to where the emulator succeeds or fails? We investigate a cross-domain foundation model for continuum dynamics, Walrus by Polymathic, using mechanistic interpretability guided by physical principles. We apply a sparse autoencoder (SAE) to probe a selected layer, and address the practical challenge of triaging a large feature set (over 20,000) using enstrophy as a physically grounded metric. As a deliberately simple testbed, we focus on shear flow and compare feature recruitment across multiple shear-flow setups, i.e. parameter values in the numerical simulation. Across setups we find evidence of piecewise consistency, with subsets of features recurring in similar roles, but this structure is intermittent and does not map cleanly onto standard physical decompositions. In parallel, direct comparisons between numerical simulation and the emulator reveal systematic output-level discrepancies, including regimes where energy/structures become too diffuse or too localized. We connect parts of these discrepancies to changes in specific SAE feature usage. Our work highlights open questions for scientific foundation models: how to robustly prioritize mechanistically meaningful features, how to separate stable structure from analysis artifacts (including single-layer and SAE limitations), and how to use established benchmarks to decide when "different" internal representations are genuinely informative rather than merely effective.

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.

This paper investigates reinforcement learning (RL) methods for improving tool-calling capabilities in multimodal small language model (SLM) agents. While existing works have explored various reward designs to improve agentic tool-calling ability, these approaches face inherent limitations for SLM training, especially under multimodal scenarios. First, many existing methods evaluate tool use correctness through exact matching against certain ground-truth or predefined formats. However, this assumption is often unsuitable for multimodal tasks, where multiple tool use paths may be valid and annotated tool trajectories are typically unavailable. Second, such sparse and brittle binary rewards provide little guidance on how to improve the underlying decision process, making them particularly difficult for multimodal SLM to learn from. To address these issues, we propose Input Attribution-Aware Policy Optimization (IAPO), an RL algorithm for improving tool use in multimodal SLM by aligning the model's attribution across input components with that of a stronger teacher. Experiments on Qwen2.5-VL-3B show that the proposed method improves visual question answering accuracy by an average of 3% across six test sets compared with existing visual tool use work, by helping the model attend to the most relevant input evidence.

Synthetic random heteropolymers (RHPs), consisting of a predefined set of monomers, offer an approach toward the design of protein-like materials. These RHPs, if designed appropriately, can mimic protein behavior and function. As such, there is a need for computational tools to efficiently guide RHP design. We bridge this gap by developing DeepRHP, a modified variational autoencoder (VAE) model under a semi-supervised framework. By equipping a classical VAE with an additional feature-based VAE, DeepRHP forces the latent space to capture structures of critical chemical features as well as individual RHP sequence patterns. In this sense, our method is versatile by allowing any relevant features to be incorporated in a hybrid manner. We demonstrate the effectiveness of DeepRHP by suggesting potential monomer compositions that stabilize membrane proteins (e.g. Aquaporin Z) in non-native environments and cross-validating our prediction with published results. The concordance between our model and true RHP function suggests strong potential in utilizing hybrid autoencoder architectures to guide RHP design for proteins and other biological compounds.

Recent advances in scientific machine learning provide a means of near-real-time solution to partial differential equations (PDEs), but lack the theoretical underpinnings of conventional simulators that support contemporary verification and validation. In this work, we construct data-driven reduced-order models that serve as structure-preserving, real-time surrogates. Remarkably, the exterior calculus that imposes physical conservation structure also exposes topological structure that we use to build a Gaussian process (GP) representation of uncertainty in state-flux relationships, ultimately yielding a Dirichlet-to-Neumann map for quantities of interest with closed-form expressions for posterior uncertainty. We specifically propose structure-preserving $H(\mathrm{div})$--$L^2$ subspaces of conventional Raviart--Thomas and $dgP_0$ elements prescribed by a lightweight transformer. Reduced-order dynamics consistent with this subspace are learned by posing a conservation law in which a GP describes the fluxes between volumes. This work hinges on a novel interface between mixed FEM spaces and GP regression; when training is posed as the optimal recovery problem (ORP), the resulting GP regression can be written as an optimization problem with equality constraints that impose a conservation structure, amenable to a fast Schur-complement training strategy. The trained model can then be solved in real time with closed-form estimators for boundary fluxes driven by prescribed Dirichlet data. The paper includes RKHS posterior error bounds for linear functionals to support uncertainty quantification, as well as numerical experiments demonstrating the accuracy of the posterior distribution as a surrogate for error estimation.

Backdoor attacks pose a serious threat to the safety and reliability of Large Language Models (LLMs), as they cause models to behave normally on clean inputs while producing attacker-specified responses when hidden triggers are present. Removing such unknown backdoors is particularly challenging when the defender does not know the backdoor attack types or the internal mechanisms formed through backdoor training. In this work, we propose a simple but effective backdoor removal method based on shared internal mechanisms across different backdoors. First, we show that different backdoors with the same task (attack objective) induce similar trigger-activated changes in the internal activations. Motivated by this observation, our method intentionally embeds a backdoor with a known trigger (\emph{dummy backdoor}) and then removes it through further fine-tuning on dummy-triggered inputs paired with clean responses. Since the dummy backdoor and the unknown backdoor can rely on shared internal mechanisms, removing the dummy backdoor also reduces the effect of the unknown backdoor. We evaluate our method on three backdoor attack types across multiple model families. Experimental results show that our method substantially reduces the attack success rate of the unknown backdoor while preserving model utility, outperforming representative existing defense methods in both backdoor removal effectiveness and utility preservation. These findings suggest that a defender-controllable backdoor can serve as a helpful proxy for mitigating unknown backdoors in generative LLMs.

Compositional data -- vectors encoding relative proportions -- arise across scientific domains, including ecology, geochemistry, and genomics. The features in these data often come with known hierarchical structure (e.g., taxonomies, phylogenies, ontologies), yet existing methods either ignore this structure, discard the intrinsic Aitchison geometry, are designed for binary trees, or yield incomplete coordinate systems. We describe PolyILR, a canonical orthonormal decomposition of the Aitchison tangent space aligned with any tree topology. Our construction defines a weighted local geometry at each internal node capturing full branching structure, then lifts these to a global orthonormal basis where every coordinate corresponds to a specific tree location. On microbiome and single-cell benchmarks, PolyILR yields stable, interpretable features and enables inference at multiscale tree resolution. We also establish a novel theoretical connection to softmax classifiers, suggesting possible applications to probabilistic modeling.

Large language models often remain sensitive to answer format: a question solved correctly in one form may fail in another semantically equivalent form. To study this gap, we define cross-format robustness as the extent to which a model answers the same underlying question consistently across formats. We then compare full-format training with FormatMix, which expands only a subset of training items into multiple equivalent formats using either random or targeted selection. Across GLM4 and Llama-3.1, multi-format supervision consistently improves both task performance and cross-format robustness, whereas Multiple-choice question (MCQ)-only supervision alone brings little benefit and can even reduce robustness. We further find that expanding only about 30% of the training set into multiple formats often recovers most of the gain from full-format training, and this effect appears across the model families and sizes we study. These results suggest that format diversity, rather than additional supervision alone, is the key driver of robustness. That lightweight multi-format augmentation is a practical way to make LLMs less sensitive to answer format without changing the base model.

How far can we reduce the number of physical keys if we endow an ambiguous keyboard with modern language models? Fewer keys increase hardware design freedom in constrained settings such as assistive devices and mobile form factors. This paper systematically evaluates text entry systems using 2-5 physical keys combined with language-model-based disambiguation. On a 300-sentence English corpus (100 sentences each for Business / Conversational / Technical), we compare key counts (2-5), letter-to-key mappings (layout-based / frequency-based / intentionally worst-case), and decoders (Trie-only, GPT-2 beam search, GPT-4o selection). We find that 3 keys + GPT-4o achieves character error rate (CER) 9.46% and word error rate (WER) 12.20%, reducing CER by 59% relative to 2 keys (CER 23.3%). At 3 keys, the key-stream entropy is 1.54 bits/char; while increasing to 5 keys improves accuracy (CER 5.4%), the marginal gains diminish. Mapping choice has a small impact under standard designs (ΔCER < 0.5 pp), and even an intentionally worst mapping degrades CER by only +0.5 pp, whereas Technical sentences yield roughly twice the error rate of Business. These results suggest that, in our evaluated offline setting under a strong LM prior, 3 keys are a practical minimum for general English.

Few-shot tabular learning provides a cost-effective approach for real-world applications where annotation is costly and collecting sufficient samples for new tasks is difficult. Existing Traditional and LLM-based methods have demonstrated effectiveness in few-shot scenarios. However, traditional methods need additional training on unlabeled or generated data, which incur significant computational overhead. In addition, LLM-based methods that directly feed raw tabular data into LLMs raise privacy and compliance concerns. More importantly, both paradigms largely overlook the semantic relationships between features, which provide structural and semantic prior for constructing a semantic graph. Semantic graph is essential for modeling meaningful feature interactions in few-shot scenarios. In this paper, we propose TAROT, a GNN-based framework that encodes the structural and semantic prior by constructing and refining a task-adaptive semantic graph from this prior, thereby improving predictive performance in few-shot tabular learning. TAROT first encodes heterogeneous tabular data into unified node semantic representations via a Unified Semantic Tabular Node Encoder (USTNE). Then, it prompts LLMs to infer the semantic relationship between features based on the task description and feature names to construct a semantic graph. To mitigate structural noise introduced by the hallucination of LLMs, TAROT introduces Task-adaptive Semantic Graph Refinement that prunes spurious or task-unrelated edges and adds missing task-related ones, aligning the graph structure with the downstream objective. Finally, a GNN performs message passing over the refined graph to capture task-related semantic dependencies for prediction. Extensive experiments on various few-shot tabular learning benchmarks demonstrate the superior performance of TAROT, establishing it as a state-of-the-art approach in this domain.

The popularization of automatic speech recognition (ASR) systems has increased exploration of the demographic biases related to race, age, gender, and accent, often formed from imbalanced training data. Most of these studies focused on standard grapheme-based ASR systems with comparatively little emphasis on phoneme-based systems, such as models that produce International Phonetic Alphabet (IPA) representations. As ASR systems shift toward multilingual support and low-resource language modeling, IPA-based layers serve as a critical, language-agnostic foundation. In this study, we evaluate the performance of two state-of-the-art open-source ASR systems, WhisperIPA and ZIPA, that generate IPA transcriptions across diverse accents and language sources. Our evaluation includes existing multilingual speech corpora and demographically annotated English-language corpora. We measure model performance by comparing model-generated IPA transcriptions against grapheme-to-phoneme (G2P) systems using both standard phoneme error rate (PER) and a proposed Soft PER metric that tolerates linguistically similar phoneme substitutions. Our analysis examines how performance varies across languages and demographic groups such as gender, accent, ethnicity, and age, revealing persistent disparities even after accounting for acceptable phonemic variation. These findings provide insight into potential sources of bias and inform the development of more inclusive and linguistically robust phoneme-based ASR systems. Our code and data will be made publicly available to the community.

Touch is a key modality for embodied agents to understand the physical world. Although recent work has incorporated tactile signals into language systems for tactile commonsense reasoning, scaling such systems to realistic open-world settings remains challenging due to two key bottlenecks: (1) current tactile reasoning datasets remain limited in format and scale, providing insufficient supervision for reasoning from tactile observations to physical commonsense and hindering the learning of transferable tactile commonsense; (2) Tactile signals are inherently redundant and action-specific, yet existing methods often overlook these properties, resulting in inefficient representations with limited semantic expressiveness. To address these limitations, we propose TouchThinker, a tactile-language framework that scales tactile commonsense reasoning to the open world from both data and representation perspectives. First, we construct TouchThinker-1M, a million-scale, multi-source tactile reasoning dataset covering \textbf{415} objects, \textbf{8} scenarios, and \textbf{7} sensor types, providing a solid data foundation for open-world generalization. We further introduce TouchThinker-Bench, an open-world benchmark with more realistic and diverse tasks. Then, we propose action-aware modeling mechanism to improve tactile representation efficiency and enable efficient reasoning. Experimental results demonstrate that TouchThinker achieves competitive performance against state-of-the-art models across multiple datasets. Our code and dataset will be made available at: https://github.com/lvkailin0118/TouchThinker.

For highly capable AI systems to operate safely in dynamic, open-ended environments, they must be able to identify, understand, and respond to moral reasons for action, and constrain their behaviour accordingly. A growing body of research aims to evaluate this capacity -- moral competence -- in today's most capable AI systems, recently reaching broadly pessimistic conclusions. One of the most ambitious such papers collects gold-standard human-authored rubrics for evaluating moral reasoning in 1,000 cases, and benchmarks frontier AI models against those rubrics, with underwhelming results. In this paper, we argue that the MoReBench dataset can be redeployed to give a much more optimistic picture of LLMs' moral reasoning (an essential part of moral competence). We show that if, instead of scoring LLMs' responses to these cases against these rubrics, we instead give the LLMs the same task given to humans -- to generate scoring rubrics for the moral analysis of particular cases -- the rubrics they generate are both better calibrated to the human rubrics than their open-ended responses, and, where they differ, plausibly reflect nothing more than the vast dimensionality of most moral problems, as well as highlighting some human departures from the "rubric for creating rubrics". Taking these points into consideration, the MoReBench dataset suggests that LLMs are significantly more capable at moral reasoning than was previously believed.

The rapid progress of reasoning and agentic large language models (LLMs) has increased the demand for long-context inference, but self-attention (SA) scales quadratically with context length. To address this, we study SWARR (Sliding-Window Attention with Reinforced Adaptation for Math Reasoning), a practical recipe for adapting SWA models to mathematical reasoning. SWARR has two stages: (1) efficient conversion from a pretrained SA model to SWA with supervised fine-tuning (SFT), which avoids pretraining a new base model, and (2) policy adaptation with reinforcement learning (RL). We find that SWA still underperforms SA after SFT, and we hypothesize that this gap is caused in part by a data-architecture mismatch: most SFT data are prepared for SA models and may contain long-range dependencies that are difficult for SWA to model. Because on-policy RL optimizes self-generated trajectories under the SWA constraint, it can adapt trajectories to better match SWA. Experiments on mathematical reasoning benchmarks show that this recipe substantially narrows the gap between SWA and SA, recovering much of the accuracy lost during SWA conversion while preserving the efficiency benefits of linear-complexity attention. Our central contribution is the empirical finding that RL changes the conclusion one would draw from conversion and SFT alone about SWA's viability for math reasoning.

Agentic infrastructure introduces a critical control-plane authorization problem: non-deterministic reasoning systems can propose high-stakes mutations to production resources, yet existing security mechanisms -- such as identity and access management (IAM), policy engines, consensus protocols, and audit logs -- either enforce static, context-unaware permissions or merely record actions post-execution. This paper introduces the Sovereign Assurance Boundary (SAB), a certificate-bound runtime admission layer for autonomous execution authority. SAB intercepts agent proposals at an assurance airlock, compiles them into typed execution contracts $C$, and binds these contracts to cryptographic evidence digests $H(E)$ and policy versions. The contracts are then routed through consequence-aware certification paths. Upon successful admission, the system emits a signed Sovereign Assurance Certificate ($Ω$) that is strictly scoped to a specific execution identity, revocation epoch, and validity window. Finally, a sovereign execution broker verifies $Ω$ and performs fresh pre-execution revocation and drift checks before invoking infrastructure APIs. We detail the airlock-broker architecture, formalize its admission and revocation invariants, and report preliminary feasibility measurements from a Go prototype evaluated over 2,500 admission attempts. Ultimately, this broker-enforced model prevents autonomous reasoning from directly mutating state, transforming delegated execution authority into a cryptographically verifiable, evidence-bound, revocable, and replayable runtime artifact.

This manuscript proposes an integral formulation of the newly defined quadratic embedding method for identifying nonlinear systems (QENDy). In the original algorithm, trajectory data points along with their time derivatives are used. Methods for calculating time derivatives make the algorithm sensitive to noise. Our integral formulation does not use the time derivatives. This results in a more robust method to learn the dynamics.

The most widely-adopted robot learning pipelines today learn skills from robot demonstrations or structured human data, which are expensive to collect and tied to specific embodiments. In contrast, unstructured human videos provide a scalable alternative. They contain diverse manipulation demonstrations across objects, scenes, and strategies, but are not directly connected to robot action. We propose LUCID, a two-stage framework that learns task intent from unstructured human videos drawn from internet-scale datasets and learns robot control in massively-parallel simulation. The intent model predicts short-horizon intent (what should happen next in the scene) from the current observation in closed loop. An embodiment-specific sensorimotor policy converts this intent into robot actions. The intent interface is shared across controllers, so the same intent model can be applied to different embodiments, from our primary dexterous hand to a parallel-jaw gripper. We evaluate LUCID on five real-world manipulation tasks: stirring, wiping, and binning supervised by only internet video, with zero-shot transfer to novel scenes and object instances; and push-T and cable routing supervised by 1 hr each of self-collected smartphone video. Project page: https://lucid-robot.github.io/.

Recent work has shown that on-policy distillation can internalize privileged context, such as system prompts or task hints, into a student model so that the context is no longer needed at inference time. Although this approach successfully improves the student's no-context performance, we identify an interesting and previously unstudied phenomenon: in many settings, reintroducing the original privileged context to the distilled student actually degrades its performance, even on instances it already solves correctly without context. We term this context-induced degradation and argue that robust internalization demands not only matching the teacher's context-conditioned behavior, but also remaining stable when the context is reintroduced, a property we call context removability. Motivated by this observation, we propose a lightweight consistency regularizer that first anchors the student's no-context output via stop-gradient, then penalizes the context-conditioned output for deviating from it via forward KL divergence. This simple addition requires only one extra forward pass per training step, yet it effectively mitigates context-induced degradation and, in many cases, even improves no-context performance. Across 12 configurations spanning diverse domains and model families, our method improves context-conditioned accuracy in the majority of settings, reduces context-induced harm in 11 out of 12 settings, and effectively eliminates response-length inflation. A mechanistic case study further confirms that context removability is achieved at the representation level, with hidden states remaining nearly identical regardless of whether the context is present.

Time-series foundation models (TSFMs) are increasingly explored as predictive experts within emerging agentic time-series systems. However, TSFMs exhibit heterogeneous inductive biases, and no single model consistently dominates across forecasting regimes, making expert selection a critical challenge. Existing systems often delegate this decision to LLM-based controllers, incurring substantial inference overhead. We present TimeRouter, an efficient routing framework that leverages empirical complementarity across a pool of pretrained TSFMs through lightweight discriminative routing, selective gating, and ensemble fallback. Concretely, TimeRouter combines a learned routing head, a selective gate, and an ensemble fallback, enabling adaptive expert selection without invoking an LLM at inference time. TimeRouter achieves state-of-the-art performance on the GIFT-EVAL leaderboard, with an LB MASE of 0.6765. Beyond benchmark performance, our ablation studies provide empirical insights into TSFM routing design, highlighting the importance of pool composition and selective gating. Taken together, these results position TimeRouter as a modular and lightweight routing layer for future agentic time-series systems built upon foundation-model pools. Our code is available at https://github.com/UConn-DSIS/TimeRouter.

Approximate tensor-network simulators enable classical simulation of quantum circuits beyond the reach of exact methods, but selecting optimal approximation parameters -- such as bond dimension thresholds -- remains a costly trial-and-error process. We present a family-aware neural architecture that predicts both the minimum approximation threshold required to achieve target fidelity and the expected wall-clock runtime for quantum circuit simulation, given only the circuit's OpenQASM description and execution context. Our key insight is that quantum circuits from different algorithmic families (e.g., QFT, Grover, VQE) exhibit fundamentally distinct simulation cost profiles due to their differing entanglement structures. We employ family-conditioned residual corrections -- additive, family-specific adjustments atop a shared backbone, drawing on established conditional computation techniques -- enabling the model to capture both universal circuit properties and algorithmic nuances. The architecture incorporates a pretrained family classifier (97.5% accuracy) and domain-informed algorithm fingerprint features derived from gate-composition heuristics. Evaluated on circuits spanning 7--130 qubits across 10 algorithm families, our system achieves 79.5% exact threshold accuracy (91.2% within one rung) and $R^2 = 0.82$ runtime correlation, with inference completing in approximately 50 ms -- replacing trial-and-error simulation runs that may take minutes to hours. Ablation studies confirm that family-aware modeling provides the single largest performance improvement (+3.2 percentage points), validating the hypothesis that algorithm family is a first-class feature for simulation cost prediction.

High-quality training data is essential for the success of machine learning models. However, real-world datasets often contain mixed types of errors arising from systematic flaws in data preparation pipelines, including label errors, feature errors, and spurious correlations. Effective debugging of training data requires both detecting erroneous samples and identifying their specific error types to enable targeted repair, yet existing data cleaning and attribution methods fail to adequately address this dual requirement. In this paper, we propose DeMix, a novel framework that simultaneously diagnoses erroneous samples and their error types. Our key insight is that different error types produce distinct patterns on model behavior. DeMix captures such error-specific patterns by influence vectors that characterize how each training sample affects model predictions across all validation samples. We formulate training data debugging as a multi-label classification problem where a classifier is developed to predict error types directly from influence vectors. We further introduce an intervention-based learning strategy that guides the classifier to capture invariant rationales specific to each error type, ensuring the learned classifier generalizes effectively. Empirical evaluations on 11 tasks across tabular data prediction, recommendation systems, and LLM alignment demonstrate that DeMix significantly outperforms state-of-the-art approaches, achieving a 22.61% improvement in data debugging F1-score and a 9.32% gain in task model performance after data repair. Code is available at: https://github.com/SJTU-DMTai/DeMix.

The widespread adoption of face recognition (FR) technologies raises serious privacy concerns, as facial data can be exploited without consent. To address this challenge, we propose Adv-TGD, a generative adversarial attack framework that synthesizes photorealistic faces capable of impersonating target identities and deceiving face recognition systems. Built upon Stable Diffusion, Adv-TGD performs per-sample LoRA fine-tuning conditioned on concise textual prompts to generate natural yet adversarially manipulated identities. Unlike conventional identity-attack approaches, our method optimizes lightweight cross-attention adapters for each source-target pair within a single-step denoising process. Latent blending is constrained by a face-local heatmap mask to ensure spatially precise identity manipulation while preserving non-sensitive regions. We introduce a composite objective that integrates masked epsilon-MSE reconstruction, thresholded identity divergence in FR embedding space, directional feature alignment, and source-similarity suppression to balance adversarial attack and visual realism. Optionally, LLaVA-generated attribute prompts enhance fine-grained semantic details without reintroducing identity cues. Under the black-box evaluation protocol, Adv-TGD attains an average attack success rate (ASR) of 85.90% across IR152, IRSE50, MobileFace, and FaceNet, surpassing the semantic SOTA baseline Adv-CPG by +6.25 points, diffusion-based makeup method DiffAIM by +3 points, and noise-based P3-Mask by +16 points. Despite its strong attack efficacy, Adv-TGD preserves high visual fidelity (PSNR = 27.15 dB, SSIM = 0.981). Furthermore, we demonstrate the flexibility of our framework by successfully extending it to in-the-wild datasets (LADN), general object classification (ImageNet), and transformer-based diffusion models (FLUX.1).

Multimodal learning hinges on capturing redundant, unique, and synergistic information across modalities, which collectively constitute multimodal interactions. A critical yet underexplored challenge is that these implicit interactions vary dynamically across samples. In this work, we present the first systematic, information-theoretic analysis highlighting why learning these dynamic, sample-specific interactions is critical for effective multimodal learning. Our analysis further reveals deficits in conventional paradigms at learning these distinct interaction types: modality ensemble approaches struggle to capture synergy, while joint learning paradigms often under-utilize redundant information. This highlights the need for an approach that can adaptively learn from different interaction types on a per-sample basis. To this end, we propose Decomposition-based Multimodal Interaction Learning (DMIL), a novel paradigm that explicitly models and learns from sample-specific interactions. First, we design a variational decomposition architecture to isolate the constituent interaction components. Second, we employ a new learning strategy that leverages these explicit interaction components in a fine-tuning process to achieve comprehensive interaction learning. Extensive experiments across diverse tasks and architectures demonstrate that DMIL consistently achieves superior performance by adapting to holistic sample-specific interactions. Our framework is flexible and broadly applicable, establishing an interaction-centric paradigm for multimodal learning. The code is available at https://github.com/GeWu-Lab/DMIL.

When many people highlight the same document, is the crowd a single consensus, or is it internally structured into reader sub-groups that mark different things -- and is that structure a stable property of a reader or of the document? Building on prior work showing an individual's within-document highlighting signal is a whisper while individuality lives in selection, we ask the group-level question on a co-readership platform using a margin-preserving curveball null. Experiment 1: within a document, readers form strong sub-groups -- pairs agree far beyond what shared salience, mark density, and sentence popularity predict (nearest-neighbour agreement z=+6.3, significant in 88% of documents). Under an eight-block region-preserving null, shared engagement with the same coarse regions of the document accounts for about 40% of this excess; the majority survives as finer reader-specific agreement (z=+3.6, 77% significant). So the within-document crowd is, in a descriptive sense, factional. Experiment 2: is that grouping a stable reader trait? Here we are honest about power. The cross-document split-half reproducibility of a pair's agreement is near zero pooled (+0.078 and 0.000 in two separately drawn samples), and a power calibration shows the test is informative only for pairs that co-read many documents. In the only informative high-overlap subset (k>=4), point estimates are positive but small-sample, imprecise across the separately drawn samples, never significant, and attenuate under the region-preserving null. We therefore leave cross-document stability unresolved: the data is consistent with anything from situational grouping to a weak-to-moderate stable reader trait. The crowd is factional within a document; whether its factions follow the reader across documents is, honestly, beyond our reach.

Stance detection requires identifying an author's position toward a target, often from short-form texts where stance is implicit, indirect, or rhetorically framed. Although large language models (LLMs) achieve strong performance on this task, single-pass prompting can be brittle when multiple interpretations are plausible. Existing aggregation strategies, such as majority voting or self-consistency, improve robustness by combining labels, but they discard the intermediate reasoning needed to resolve conflicting interpretations. We introduce a multi-agent reasoning framework with adaptive worker allocation for stance detection that shifts aggregation from label-level voting to reasoning-level synthesis. The framework employs a Manager-Worker architecture in which a Manager adaptively allocates a variable number of Worker agents based on input complexity. Each Worker analyzes the input from a distinct perspective and produces a reasoning-only explanation without emitting a stance label; the Manager then synthesizes these explanations to produce the final prediction. We evaluate the proposed framework on SemEval-2016, P-Stance, and COVID-19 Stance using Llama, Mistral, and Gemini. Results show that the framework yields the largest gains on implicit and context-dependent stance cases, achieving 86.07 Macro-F1 on COVID-19 and 82.90 on SemEval-2016, while remaining competitive on more explicit stance datasets such as P-Stance. These findings suggest that adaptive reasoning-level aggregation is most beneficial when stance cannot be reliably inferred from surface cues alone.

Predicting process-property relationships in manufacturing is often challenged by high experimental costs and the limited interpretability of complex 'black-box' models. This paper proposes a novel knowledge distillation framework designed to achieve high-accuracy predictions in data-scarce scenarios. The framework integrates analytical physics priors, which are systematically extracted from scientific literature via Large Language Models, into a privileged teacher model. We employ a Graph-Masked Attention layer to capture the complex physical dependencies among input variables showing strict setpoints or a combination of static and high-frequency temporal signatures. This privileged knowledge is distilled into a lightweight student predictor for inference. The feasibility and robustness of the framework are evaluated through a comprehensive experiment across five diverse manufacturing processes. To ensure statistical reliability, given the small dataset sizes, a repeated K-fold cross-validation technique is employed to quantify model stability and generalization. Results indicate that the proposed framework consistently achieves high predictive accuracy across all evaluated domains. Most importantly, the architecture demonstrates significant fault tolerance by maintaining robust predictive performance even in scenarios where LLM-derived analytical priors are suboptimal or incomplete. Furthermore, the student predictor achieves an inference frequency exceeding 6000 Hz, which facilitates real-time edge deployment on standard industrial hardware. This work provides a scalable solution for bridging the gap between theoretical physics and real-time industrial monitoring in data-limited environments.

Activation steering offers a lightweight approach to control language models' behavior at inference time, but whether it succeeds or fails heavily depends on the prompt, concept, model, and steering configuration. Finding the regime and boundaries of successful steering typically requires expensive grid searches and post-hoc evaluation of full autoregressive rollouts. In this work, we investigate whether steerability can be predicted from the model's internal states at the beginning of the generation process, e.g., after generating the first few tokens, and how to leverage such a predictor to improve steering success rate. To this end, we first introduce ASTEER, a testbed including 1.4M steered generations, spanning 150 concepts with each steering success/failure labeled. Leveraging this testbed, we analyze the model's early decoding dynamics by extracting features that compare hidden states before and after steering across layers and initial decoding steps. These features help us understand how steering's effects propagate along layers and token positions, which provide key information for steerability prediction. We then train a Gradient Boosting Decision Trees (GBDT) classifier on these features to predict whether an intervention will under-steer, succeed, or over-steer without requiring full rollout. Our predictor achieves around 0.7 macro-F1 score on unseen concepts, demonstrating that early hidden states encode substantial, structured information about eventual steering efficacy. We further leverage this steerability predictor as guidance for steering strength searching, achieving near-optimal performance with a small fraction of decoding cost.

In this paper, we consider a class of networked systems comprising an interconnected set of linear subsystems, disturbance inputs, and performance outputs. Using dissipativity theory, we first propose a model-based hierarchical control design strategy to ensure the closed-loop networked system is dissipative from its disturbance inputs to performance outputs. This involves designing local controllers for each subsystem to enforce local dissipativity guarantees, which are then exploited to co-design distributed global controllers and the interconnection topology to enforce global dissipativity guarantees while optimizing interconnection topology costs. The overall design process requires only solving a sequence of linear matrix inequality (LMI) problems, thereby retaining compositionality and decentralizability while avoiding non-convex, iterative design processes that are inefficient and centralized. This model-based hierarchical control design strategy assumes the knowledge of the subsystem dynamics, which may not hold in many real-world networked systems. Motivated by this, we also propose a data-driven hierarchical control design strategy that assumes only the availability of rich input-state-output trajectory data from the subsystems. The proposed data-driven design process assumes that the unknown disturbances affecting the subsystem dynamics are bounded by a quadratic matrix inequality (relaxing conventional bounds) and accounts for this by using the matrix S-lemma. Finally, the effectiveness of the proposed model-based and data-driven hierarchical control designs is illustrated for a networked system representing a DC microgrid, with the aim of enforcing robust (dissipative) voltage regulation and current sharing.

We introduce Kuramoto attention, a self-attention layer in which each hidden coordinate is an angle. The layer scores tokens by gated cosine similarity, attends over previous phase states, and updates each token by the tangent component of the attention-weighted circular mean. Because the values are the raw phase states, this update is exactly the Kuramoto coupling term $\sum_u A_{t,u}\sin(θ_u-θ_t)$, with the attention matrix acting as an adaptive, content-dependent coupling kernel. Equivalently, the gated score is a learned metric on the torus that selects which tokens couple, and the update pulls each token toward the circular mean of the tokens it selects, tightening their phase agreement. The same two ingredients, an invariant similarity score and an on-manifold mean, define such a layer on any compact group; the torus is the abelian case, where both are closed-form. The softmax weights solve an entropy-regularized phase-retrieval problem, and rotary position enters as a position-dependent phase drift in the score. On enwiki8 character-level language modeling, the layer trains as a functional language model whose bits-per-character stays close to a strong matched RoPE+SwiGLU transformer: within $0.02$ BPC at one million parameters ($1.637\pm0.010$ versus $1.616\pm0.004$) and level on the median at five million ($1.448$ versus $1.452$ over five seeds) with the transformer ahead on the mean ($1.468$ versus $1.456$). These experiments establish that the constrained geometric structure is a viable language model at this scale; the structure itself, and its synchronization reading, is the contribution. Ablations isolate the load-bearing components, and the result gives a compact bridge between self-attention and phase synchronization.

Text-attributed graphs (TAGs) underlie real-world applications such as citation networks, social media, and e-commerce. Few-shot graph learning on TAGs is hard: with only a handful of labels per class and the rest of the graph unannotated, neither GNNs nor LLMs can learn well on their own. GNNs read topology and fail on cold nodes; LLMs read text and fail on text-ambiguous nodes. Existing LLM-GNN methods all follow the same recipe: designate one model as the golden teacher and use its outputs (e.g., features or pseudo-labels) to supervise the other. We argue this golden-teacher assumption breaks under sparse supervision: neither model is golden, and treating either as such transfers its blind spots into the student. We therefore ask: can we avoid designating either model as the golden teacher, and still perform effective graph learning? We answer with LLM-GNN Co-Teaching, a bidirectional co-teaching framework in which neither model is fixed as teacher. The GNN and LLM exchange their most confident pseudo-labels under an architecture-specific small-loss criterion, and both update every round. Supervision is then mined from the trajectory: whenever a node moves from cross-model contradiction at round t to cross-model agreement at round t+1, the LLM's two answers on the same input form a preference pair (old contradicting self < new peer-endorsed self) for DPO training. We call this Round-based Pseudo-Label Preference Optimization (RPL-PO). On six benchmarks, LLM-GNN Co-Teaching consistently outperforms GNN-as-Judge and all prior methods, with absolute 3-shot gains of 7.86% on Cora and 7.73% on ogbn-arxiv; improvements carry over to 5-shot and to zero-shot cross-dataset transfer. Error-structure analysis further shows that abandoning the golden-teacher assumption substantially improves the LLM's graph learning capability on challenging samples.

We study superspace concentration as a quantum resource, formalized through the focus measure F(\r{ho}) = λ_max(\r{ho}_super) - the largest eigenvalue of the reduced superspace state - which quantifies the capacity of a quantum system to concentrate informational weight into a preferred subspace of an extended degree-of-freedom space. We develop a complete resource-theoretic framework around this measure and validate its properties through GPU-accelerated numerical simulation. Analytic decoherence predictions are confirmed to machine precision (1.11 x 10^{-16}) for superspace dimensions dS in {2,4,8,16,32}. Focus monotonicity holds across 10,000 random states with zero violations under four focus-non-generating channels across six system configurations. Focused quantum states resist coherent unitary attacks with significantly greater resilience than standard fidelity predicts, with focus remaining above 0.9 at attack strength ε = 0.302 versus ε = 0.174 for fidelity. We further demonstrate that the focus measure and the U(dS)-asymmetry measure are operationally distinct: asymmetry remains near zero and provides no robustness signal under coherent and targeted attacks while focus tracks spectral concentration and remains robust until ε > 0.3. The connection between Grover's algorithm and superspace concentration is made explicit via the identity F(|ψ_k><ψ_k|) = P(marked), providing a resource-theoretic interpretation of oracle query complexity. Finally, we provide the first numerical characterization of the focus capacity gap ΔF, identifying a log_2(dS) scaling law confirmed for both product and correlated noise channels.

We present an approach based on tensor networks for distributed quantum computing simulation of chemical wavepacket dynamics in a continuous variable representation. The central idea is that the tensor-network representation of the multidimensional time-evolution operator naturally induces an elevated Hilbert space where the dynamics decomposes into a set of independent lower-dimensional propagations. This transformation converts an entangled quantum evolution into a set of parallel computational tasks that can be executed asynchronously across heterogeneous quantum and classical computing architectures. The resulting formalism establishes a direct connection between tensor-network decompositions, uniformly controlled quantum circuits, and asynchronous distributed quantum computing. The approach is developed with a goal towards hybrid quantum/classical implementation, and is appropriate for a general heterogeneous mixture of quantum hardware systems. The experimental realization of the asynchronously distributed quantum processes that arise from the tensor-network decomposition are carried out on the Sandia National Laboratories' trapped-ion quantum computer, where the circuits are compiled using native partial-entangling $XX(θ)$ gates, reducing the expected two-qubit gate infidelity by more than 30\% relative to conventional fully entangling decompositions. We demonstrate the methodology by quantum computing the vibrational spectra of a small protonated water cluster that shows critical quantum nuclear behavior. Such water cluster systems have been found to be challenging for experimental action spectroscopy and for theory, and here, for the first time, we provide results for vibrational spectroscopy that are in agreement with the respective classical results to within 4cm$^{-1}$, thus allowing for the potential for spectroscopic accuracy from quantum computations.

Modern Vision-Language Models (VLMs) benefit from chain-of-thought prompting and test-time scaling, but these gains often come with prohibitive inference cost due to large visual contexts and long decoding chains. We view this cost through two coupled axes: Visual Context Scaling (VCS), which controls how much visual evidence is passed to the language model, and Visual Reasoning Scaling (VRS), which controls how much inference-time reasoning search is performed. Existing methods typically optimize one axis at a time, leaving the joint allocation of compute across these axes underexplored. We introduce Adaptive Visual Inference Scaling (AVIS), a lightweight policy that adapts both VCS and VRS per query. AVIS realizes VCS through Key Diversity Visual (KDV) pruning, a training-free $O(N)$ key-based rule for removing redundant visual tokens before prefilling, and realizes VRS through adaptive self-consistency, using a learned difficulty predictor to select the number of reasoning rollouts. AVIS is deployment-friendly and compatible with shared-prefill inference, where all rollouts reuse a single prefilling pass and KV cache. Across diverse image and video reasoning benchmarks, AVIS improves the accuracy--compute trade-off relative to VCS-only and VRS-only baselines, and remains effective on top of RL post-trained VLMs while keeping compute and latency low.

In many materials and product design problems, desirable candidates exhibit properties that fall within an acceptable range rather than achieve a single optimum. Recovering multiple, distinct solutions that satisfy such specifications is also practically valuable, as some candidates may be preferred for reasons of cost, processability, or robustness that are difficult to encode directly in an objective function. Here, we develop a range-aware Bayesian optimization (BO) framework in which the acquisition function directly scores the posterior probability that a candidate satisfies a target range. The framework naturally extends to parallel pursuit of multiple distinct specifications over a shared candidate space. Across benchmark tasks, range-aware acquisition consistently recovers larger and more diverse sets of valid designs than standard BO baselines and recent goal-seeking methods. Its utility is further demonstrated in two practically motivated design case studies involving optimizing reaction conditions for polymer synthesis and sequence-defined oligomer discovery for prescribed optical absorption bands, supported by quantum chemical calculations. These results suggest that range-aware BO can provide a practical and sample-efficient foundation for specification-driven design, particularly when design flexibility and solution diversity are important considerations.

We propose a spectral-based, unsupervised representation learning framework to derive low-dimensional embeddings for clinical concepts and patients in rare disease cohorts from electronic health records, where data are high-dimensional but sample sizes are limited. To overcome this challenge, we incorporate a knowledge matrix extracted from a broader population that shares a partially overlapping subspace with the rare-disease cohort. Our method departs from existing approaches by relaxing restrictive one-to-one signal-alignment assumptions between the latent data matrix and knowledge matrix, allowing more flexible and realistic forms of structured sharing. We introduce a novel two-step spectral embedding procedure: first, we identify and remove irrelevant components from the knowledge matrix; then, we apply a projection-based method to separately recover shared and heterogeneous components. Simulations and an analysis of a real-world multiple sclerosis cohort show that the proposed method outperforms competing approaches, particularly in challenging scenarios where shared signals are weak and only partially aligned, as is common in rare-disease data.

Closed-loop planning in complex, real-world driving scenarios presents a critical challenge for autonomous driving systems. While traditional rule-based methods are interpretable, their predefined heuristics lack the adaptability for dynamic traffic environments. Learning-based approaches have shown considerable promise. Conversely, learning-based approaches, despite their promise, struggle to balance the modeling diverse and multimodal driving behaviors and real-time planning, often leading to indecisive or unsafe actions. To address this limitation, we propose Consistency Planner, a real-time planning framework with fast-sampling consistency models. Our approach is built upon two key technical contributions. Efficient Multimodal Sampling: We employ fast-sampling consistency models to generate a diverse set of plausible future trajectories. This enables efficient, real-time exploration of multimodal actions, overcoming the computational bottlenecks of previous iterative generative methods. Heterogeneous Feature Fusion: We introduce an attention-enhanced decoder that dynamically integrates heterogeneous input features (including scene feature and action token) into a cohesive representation for robust planning. Extensive evaluation in the Waymax simulator demonstrates superior performance in safety metrics compared to existing methods, with particularly strong results in challenging dynamic scenarios.

Graph analysis underlies many applications whose answers cannot be looked up in a single record or retrieved along a path: laundering rings, drug repurposing, user preference, and scientific theme are all inferred from a node together with its neighbourhood. We introduce GraphInfer-Bench, a benchmark for whether LLMs can perform this graph inference: producing an open-ended answer that no single node supports and no path retrieves. Existing graph-QA protocols cannot test this capability: algorithm simulation, node classification, single-node description, KG-QA, and GraphRAG all admit answers retrievable from one node or along a path. GraphInfer-Bench defines five tasks along Description (what a region is) and Comparison (how regions differ), each constructed so the ground truth lives in no single node. The release contains 42,000 samples across six real-world graphs, produced automatically and screened by a four-layer quality-control protocol. We evaluate four method families against the same tasks: graph-token alignment models, zero-shot frontier closed-source LLMs, Graph2Text supervised fine-tuning, and plain GNNs as a structural reference. No method family closes the gap. Graph-token alignment partially handles description tasks (relational, theme) but collapses on comparison tasks. Frontier LLMs lead on outlier detection and community partition among LLM-based methods but lag on masked-node prediction. Graph2Text SFT is the strongest LLM-based method on the description side yet falls behind frontier LLMs on comparison. Across every task, plain GNNs match or beat the strongest LLM-based row, with the largest margin on community detection. GraphInfer-Bench surfaces graph inference as an open capability gap rather than a property of any one architecture.

Large Language Models (LLMs) have advanced rapidly, but their limitations in structured and multi-hop reasoning underscore the need for graph-native, synergistic artificial intelligence (AI) systems. Graph-structured data underpins critical applications across social, biological, financial, transportation, web, and knowledge domains, making it essential to understand how LLMs can leverage graph computation for grounded, context-rich inference. Three complementary synergies are emerging: LLMs augmented with graph computation for retrieval and reasoning; bidirectional integration between LLMs and knowledge graphs (KGs), where LLMs support KG construction and curation while KGs enforce semantic constraints and factual consistency; and AI agents strengthened by graph algorithms for planning, decision making, and multi-step reasoning. In parallel, LLMs introduce new capabilities for graph data management and graph machine learning (ML) through natural language interfaces and hybrid LLM-graph neural network (GNN) pipelines. This tutorial synthesizes the algorithms, systems, and design principles driving these converging directions, offering data science and data mining researchers a unified perspective on integrating LLMs, graph data management, graph mining, graph ML, and agentic computation into next-generation graph-native AI systems.

Reinforcement learning typically improves multi-turn agent capabilities through the terminal outcome of the trajectories, which makes it difficult to determine credit assignments for each intermediate turns. Recent on-policy self-distillation methods offer a promising alternative by converting privileged feedback into dense token-level supervision through a self-teacher. Our study is motivated by the unexpected performance degradation observed when naively extending this paradigm to multi-turn settings, which we attribute to a lack of alignment between privileged feedback, such as successful trajectories or terminal outcomes, and the student's current decision context. We introduce HERO, a hindsight-enhanced self-distillation framework that uses next environment observations as locally aligned feedback. After each rollout, HERO reflects on the completed interaction to convert each observation into a compact turn-level diagnosis, that captures actionable feedback about the original action such as its necessity, validity or failure cause. On TauBench and WebShop, HERO improves task success and reduces unnecessary turns over environment-feedback-only self-distillation and GRPO. It is especially effective under limited training turn budgets, where successful rollouts are rare and GRPO provides weak reward-contrast signals.

Continuous electrocardiography (ECG) monitoring could surface rhythm abnormalities before they escalate into cardiovascular events. However, a deployable system must satisfy three requirements simultaneously: legal-grade privacy (GDPR, HIPAA), real-time inference on constrained edge hardware, and detection quality under non-IID cross-hospital data. We design and evaluate an end-to-end federated system addressing all three for unsupervised 12-lead ECG anomaly detection on PTB-XL dataset, combining three autoencoder families (VanillaAE, ConvAE, VAE), Flower-based federated averaging (FedAvg) across ten simulated hospitals, client-side differentially private SGD (DP-SGD) with a Rényi-DP accountant, and 8-bit integer (INT8) post-training quantization with Raspberry Pi 4 benchmarking. Our main contributions are: an empirical characterization of how these mechanisms compose, practical DP-specific recommendations, and technical and security insights for a clinically sensitive setting. Federated learning matches or exceeds the centralized baseline across all architectures (ConvAE federated area under the ROC curve, AUROC, $0.782$), and an $\varepsilon$ sweep identifies $\varepsilon=4$ as the recommended clinical operating point. INT8 quantization roughly halves model size and cuts Pi 4 latency by up to $44%$ with $<0.12%$ AUROC loss. Crucially, DP and quantization penalties are empirically independent, so practitioners need not trade a strong privacy guarantee for a compact edge footprint. To our knowledge, this is the first system combining federated learning, formal $(\varepsilon,δ)$-DP, unsupervised reconstruction-based detection, and quantized AArch64 deployment.

The escalating demand for mental healthcare, driven by rising societal stress, highlights the limitations of traditional psychiatric diagnostics. Conventional methods - relying primarily on clinical interviews and patient self-reports - are inherently vulnerable to subjective bias and the varying empirical judgment of practitioners. To address the need for quantitative evaluation, biological signal-based detection, including electroencephalography (EEG) and functional near-infrared spectroscopy (fNIRS), has emerged as a promising objective alternative. Such technology is particularly vital for identifying latent depressive states that may be unrecognized by the subjects themselves. Furthermore, in aging populations, the high comorbidity between depression and dementia necessitates early differentiation to prevent mutual symptom exacerbation and maintain Quality of Life (QoL). This pilot study of eleven healthy students establishes a framework for biological signal-based depression detection, serving as a foundational step toward automated, objective diagnostic tools for clinical use.

Generic time-series foundation models transfer poorly to wireless network telemetry whose signals are bursty, zero-inflated, and coupled across protocol layers. We present APEX, a network-native, decoder-only transformer for forecasting enterprise AP telemetry, and evaluate it on DHCP degradation as a representative network task. APEX is pre-trained on 10-channel multivariate telemetry from ~4,500 production wireless networks (~100K AP time series, 34 metrics per AP), and is available as APEX-Large (269M, cloud) and APEX-Edge (10.5M, edge). On a 192-step (4-day) DHCP degradation benchmark, APEX-Large reduces MAE by 18% over the strongest foundation-model baseline (Toto) and 38% over SARIMA, with anomaly-detection F1 = 0.93, while APEX-Edge enables sub-second, privacy-preserving inference on AP-class edge hardware. These results suggest network-native pre-training is a practical foundation for proactive wireless operations.

Large language models (LLMs) achieve remarkable performance across a wide range of tasks, but their autoregressive decoding process incurs substantial inference costs due to inherently sequential token generation. Speculative decoding addresses this bottleneck by employing a lightweight draft model to propose multiple future tokens that are subsequently verified in parallel by a larger target model. Recent work has demonstrated that diffusion language models are well suited for this setting, as they can generate entire blocks of draft tokens in parallel and thereby alleviate the sequential constraints of autoregressive drafting. A subtlety of this regime is that block-diffusion drafters generate tokens bidirectionally within a block, whereas verification is performed by an autoregressive target model that evaluates tokens in a strictly left-to-right manner, leaving a gap between the symmetric training-time objective and the asymmetric verification-time reward. In this work, we offer an empirical analysis of three training-time interventions that narrow this gap: token positional weighting, a first-error focal loss that targets the position that breaks the accepted prefix within each block, and a chain loss term that substitutes a differentiable surrogate for the expected accepted length. The three interventions act along orthogonal axes (position, block-conditional first error, joint prefix) and compose additively; they are likewise orthogonal to test-time alignment mechanisms such as multi-draft self-selection, with which they can in principle be combined. Across four target models and six reasoning, code, and dialogue benchmarks, the three interventions raise accepted draft length by 21-76% per benchmark over a position-uniform baseline, without adding additional forward passes and without changing the inference pipeline or the rejection-sampling exactness contract.

Agent Skills augment large language model (LLM) agents with procedural knowledge at inference time, but current benchmarks rarely distinguish what a Skill says from how it is organized. We study this distinction through Progressive Disclosure, where a concise root file points agents to supporting resources on demand, and compare it with a normalized flat baseline. We present SkillJuror, a framework for evaluating Skill writing paradigms through semantically controlled variants, matched multi-trial evaluations, and trajectory evidence while holding task knowledge fixed. In an 82-task SkillsBench study, Progressive Disclosure changes runtime behavior before aggregate outcomes: distinct Skill resources touched per trajectory rise from 1.18 to 3.85, and effective uptake events rise from 1.33 to 3.92. It also yields 17 additional verifier-passing trials out of 410 matched trials (+4.1%) over the normalized flat baseline. The benefit is task-dependent. Progressive Disclosure helps when supporting resources guide implementation, checking, or repair, but is weaker when success hinges on exact output conventions, numerical thresholds, or long artifact-generation pipelines. These results show that Skill organization is not mere presentation: it can change how agents search and apply procedural knowledge, while outcome gains depend on whether the exposed resources are actionable for the task. Code is available at https://github.com/zhiyuchen-ai/skill-juror.

Modern pretrained self-supervised automatic speech recognition models are trained on large-scale audio data to encode speech into contextualized representations. However, their training data are heavily skewed toward high-resource languages with little data from low-resource languages, raising concerns about the potential underrepresentation of typologically uncommon speech sounds such as click consonants primarily found in Khoisan languages. This leads to our central research question: Can these models recognize click consonants as accurately as other speech sounds? To address this question, we fine-tune and compare pretrained self-supervised speech models (Wav2Vec2 and HuBERT) on data from two click-rich Khoisan languages (G|ui and West !Xoon). Our results reveal that the fine-tuned models consistently recognize clicks more accurately than non-clicks, suggesting that self-supervision enables generalization across human speech sounds including rare phonemes.

Financial and tabular question answering requires more than fluent reasoning: answers must be grounded in the exact facts, formulas, units, signs, and scales that support them. A single misread cell or incorrect operation can silently produce a plausible but wrong result. We introduce \textsc{MOCA-Agent}, a market-of-claims code agent that replaces free-form multi-agent debate with claim-level verification. The system decomposes each question into typed atomic claims, asks specialist trader agents to buy or sell those claims, clears their orders into confidence-weighted accept/reject decisions, and synthesizes an executable Python program from market-supported evidence. A code-aware verifier then checks the program for execution, structural consistency, and common financial reasoning errors, with at most one market-aware repair round. Across ten public benchmarks spanning financial numerical reasoning, general tabular reasoning, ESG question answering, and multimodal chart reasoning, \textsc{MOCA-Agent} achieves strong performance using a fixed Qwen3.6-27B backbone, including $78.3\%$ on FinQA, $76.0\%$ on FinanceMath, $71.2\%$ on MultiHiertt, $86.9\%$ on ESGenius, and $85.6\%$ average on FinChart-Bench. These results show that aggregating evidence at the level of atomic claims, rather than whole answers, improves robustness in high-stakes numerical reasoning.\footnote{The code and data are available: https://github.com/UBC-NLP/MoCA-Agent.

Urban heat is amplified by impermeable surfaces and heterogeneous built environments, yet street-level variability remains difficult to quantify because multi-sensor observations are rarely available in consistent, analysis-ready form at the necessary spatiotemporal scales. We present "Urban Heat MiniCubes," a publicly available, FAIR-oriented dataset designed for machine learning applications in urban heat research. The dataset provides harmonized 90 x 90 km gridded data cubes for 48 cities in the Western Hemisphere spanning 2022-2023, with variables reprojected and collocated to a common grid to reduce preprocessing (e.g., reprojection, resampling, and spatiotemporal alignment). Urban Heat MiniCubes includes two complementary modalities: (i) higher-spatial-resolution, lower-frequency observations from Landsat 8/9 (e.g., surface reflectances) and Sentinel-1 (e.g., synthetic aperture radar backscatter), and (ii) higher-temporal-frequency, coarser observations from GOES-R (e.g., longwave infrared brightness temperatures) and a microwave land surface temperature product. We document variables and metadata and provide technical assessment using inter-variable analyses and autoencoder-based reconstruction-error summaries across pixel classes (e.g., water and cloud). Potential use cases and limitations are also discussed.

The advancement of AI capabilities compels researchers and the public to be more aware of its potential worldwide impact. A pressing near-term concern is the regulation of military AI applications. Armament manufacturers and defense contractors are increasingly investing in AI capabilities and forging partnerships with AI companies, creating a burgeoning coalition that demands military leaders, arms control diplomacy experts, and AI researchers collaborate to ensure a safer future. While AI researchers often focus on the long-term implications of superintelligent AI, this approach may not adequately address the immediate challenges posed by AI in military applications. Success requires acknowledging and mitigating the emerging risks of frontier AI models that plan to be integrated into defense applications, like military AI systems. Arms control has reduced past catastrophic risks, so lessons learned from nuclear deterrence can guide AI safety and security research towards innovations in verification and diplomacy. AI researchers, however, must assist in leading the technical research that clearly defines and alleviates instability in military settings. Given these new responsibilities and the lack of sufficiently reliable solutions, we argue that AI researchers must take a leading role in advancing arms control research to minimize risk in military AI applications.

We introduce the ladderpath index as a measure of language complexity grounded in algorithmic information theory. It counts the minimum steps needed to reconstruct a sequence through hierarchical reuse of repeated substructures, capturing an exactly computable but constrained form of algorithmic compressibility related to, but distinct from, Kolmogorov complexity. We apply the ladderpath approach to 21 parallel corpora from the Parallel Universal Dependencies dataset. The ladderpath index is approximately invariant across the languages, and varies much less than the corpus length. This is more pronounced when all corpora are mapped to a unified binary representation, providing evidence for the equi-complexity hypothesis from a representation-independent perspective. We also observe trade-offs between character inventory size and corpus length, and between vocabulary-level and corpus-level reconstruction complexity, supporting the trade-off hypothesis that total complexity is conserved and redistributed across linguistic levels. The reusable substructures identified by the ladderpath approach, without any linguistic input, overlap with words and morphological components attested in the natural vocabulary. The hierarchical reuse captured by the ladderpath approach parallels the chunking mechanisms proposed in cognitive science, where the human cognitive system compresses linguistic input into nested, reusable units under shared memory and processing constraints. This connection between cognitive chunking and the ladderpath approach provides a new interpretation for the equi-complexity and trade-off hypotheses, grounding both in the shared cognitive architecture that underlies language processing across human languages.

Contrastive Reinforcement Learning (CRL) has seen recent success in a wide variety of goal-conditioned robotics tasks by learning structured representations of the dynamics. However, despite its success in locomotion and simpler control domains, CRL often struggles in interaction-rich manipulation. We argue that a key source of this difficulty is object-centric interaction, such as contact or grasping, that induces distinct changes in the underlying dynamic modes. In this work, we formulate manipulation dynamics as a piecewise-smooth Markov process and show that interaction-induced mode changes create piecewise nonlinear reachability structures that are difficult for standard CRL energy functions to represent and plan over. Based on this analysis, we introduce Interaction-weighted Resampling (IWR). IWR performs interaction-aware resampling around phases before, during, and after interactions, encouraging the learned representation to preserve the mode boundaries that determine future reachability to capture multi-modal and piecewise nonlinear reachability. Across interaction-centric environments, including 2D dynamic control, robotic manipulation, and robot air hockey, IWR improves both sample efficiency and overall performance over prior CRL methods, with 19.8% average improvement in simulation. Finally, using a sim-to-real pipeline with policies trained by IWR, we demonstrate the first real-world goal-conditioned robot air hockey agent capable of hitting goals, improving success from 25% to 60%. Project Page: IWR-arxiv.github.io.

Autoresearch agents now propose, evaluate, and select scientific candidates against a metric, and that metric is usually an aggregate reduced over a heterogeneous space of regions, slices, or cohorts. We show that when scientific validity lives in that disaggregated structure, the aggregate can rank the wrong candidate first. The headline number improves while the structure underneath inverts, so a decision made on the number accepts a candidate that quietly breaks the model. The failure is not domain-specific. It appears wherever a candidate's validity is multi-dimensional but its verifier is a single reduction. We demonstrate the inversion on a fire-model task in the Ecosystem Demography model. The highest-scoring candidate and a slightly lower one are within noise of each other on global score, yet the top-scoring one collapses the protected boreal regions while the other preserves them. What separates them is the per-region behavior, not the headline number. This decision should not be left to the agent that produced the candidates. The agent optimizing the score is the last party likely to catch the score being wrong, and a prompt has no remaining turn once the agent has stopped. We move the decision to an external control loop that audits each candidate on its disaggregated behavior and acts after the agent has decided. It can demote a candidate the agent would have accepted, and it can reopen a run the agent had declared finished. Our contribution is the inversion finding itself, and a search-discipline protocol that decides on reviewable candidate-effect evidence instead of the score.

LLMs and LLM agents should improve when given feedback, but identifying when they are able to do so is difficult: feedback is heterogeneous, domain-specific, and difficult to control. We approach this challenge by asking LLMs to perform regular-expression induction, a classical symbolic learning problem where precise mechanisms for feedback exist in the form of counterexamples. In counterexample-guided learning, a learner (LLM) proposes candidate regular expressions from positive/negative-labeled strings, and the teacher (verifier) returns counterexamples showcasing the difference between the candidate and target languages. We identify novel counterexample-guided refinement strategies that enable effective regex learning, such as regularization and symbolic counterexample clusters. We also explore agentic strategies such as reflection and repair loops. Empirically, we find that verifier feedback substantially improves sample efficiency on challenging regex-induction tasks, reducing the number of labeled examples required and enabling learning of complex target expressions where standard prompting fails. For example, on the hardest task groups, our counterexample-guided framework improves success from 3.2% to 38.1% and from 38.9% to 74.1% on two different regex domains. These results suggest that LLMs can benefit from rich feedback beyond treating it as additional data, opening the door for robust verifier-guided methods for LLM-based program synthesis and formal reasoning.

Training capable OS agents requires data that simultaneously captures structured user intents, multi-turn task delegation, and grounded tool execution--properties absent from existing datasets. We propose ISE (Intent -> Simulate -> Execute), a three-stage synthesis paradigm that addresses these gaps jointly. Stage 1 constructs roughly 50000 structured intents via a 4D framework (Persona x Domain x Task x Complexity); after deduplication the pool contains 43956 unique intents and attains a Vendi Score of 61.57 over the entire pool on mpnet-base-v2 embeddings (cosine kernel, q=1). Stage 2 drives multi-turn user-agent interaction through a role-locked user simulator that grounds each user turn in actual execution outcomes, producing 23132 complete trajectories averaging 8.12 user turns and 68.24 total dialogue turns. Stage 3 runs every tool call inside a live, isolated OS workspace, generating authentic failure-recovery dynamics instead of simulated responses. Fine-tuning on ISETrace improves ClawEval pass@1 from 19.3 to 37.7 using Qwen3-8B on agent tool-use tasks with a standard protocol. This result outperforms zero-shot GPT-4o and the larger Qwen3-32B base model which is four times bigger. An ablation on Stage 2 proves multi-turn simulation brings a large portion of the performance gain. We release all source code and dataset at https://github.com/Valiere01/ISE-Trace.

Fourier neural operators (FNOs) are effective and efficient surrogates for approximating solutions of PDEs and generalize across discretizations. However, owing to the reliance on frequency truncation to maintain learning efficiency of FNOs, empirical studies suggest that FNOs exhibit spectral bias toward low-frequency information, which may hinder the learning capability especially for certain PDEs with strong high-frequency oscillations. To address this limitation, we propose SirenFNO, a novel framework that leverages sinusoidal representation networks (SIRENs) to learn implicit neural representations and performs mode-wise kernel parameterization. Our SIREN parameterization learns a full-grid spectrum with a constant and discretization-independent parameter count, thereby eliminating the need for frequency truncation. We further extend SirenFNO with functional tensor decompositions to enhance parameter and learning efficiency. Empirical results show that our SirenFNO consistently outperforms FNO with approximately $4$ to $15$ times parameter reductions with preserved discretization invariance, and our functional decomposition variants obtain performance improvements with a maximum of $73$ times fewer parameters across multiple PDE benchmarks.

Large language models increasingly express uncertainty through natural-language statements, yet these expressions often fail to reflect the model's sampled behavior. We study verbal uncertainty alignment as a distributional calibration problem: the appropriate uncertainty target for a prompt should be estimated from repeated model outputs rather than from an isolated response. However, group rollouts alone are insufficient, since the resulting target must provide a useful training signal. Existing targets only partially satisfy this requirement. We propose SAGE, Semantic-Answer Guided Entropy, a group-level uncertainty target that constructs an answer-conditioned uncertainty geometry over sampled responses. SAGE preserves categorical, numeric, and symbolic answer distinctions while maintaining a smooth and scale-preserving calibration signal. We further apply this target through Group-Uncertainty Preference Optimization, or GUPO, an uncertainty-channel training framework that supervises verbal uncertainty expressions rather than the full response. Experiments across factual, mathematical, and multiple-choice reasoning tasks show improved uncertainty ranking, lower calibration error, and reduced overconfidence.

Accurate prediction of absorption, distribution, metabolism, and excretion (ADME) properties is critical to drug discovery, but remains challenging because ADME endpoints are noisy, interdependent, and often data-limited. We propose a molecular graph-transformer pretraining framework that combines chemistry-specific self-supervision with contrastive mutual information machine learning (cMIM). Our method encodes molecular graphs into latent variables, reconstructs SMILES strings from the graph-derived latent codes, and augments the contrastive objective with domain-specific self-supervised chemistry tasks. Rather than treating these tasks as auxiliary regularizers with separately tuned loss weights, we formulate reconstruction, contrastive discrimination, and chemistry-specific supervision as unit-weighted log-probability factors in a single probabilistic latent-variable objective. For fine-tuning, we propose a multi-task GNN readout architecture with task-specific multilayer perceptron heads, preserving shared representation learning while mitigating negative transfer and improving the modeling of heterogeneous, nonlinear task relationships. Across Biogen, ExpansionRX, and ChEMBL-MT, the resulting Contrastive KERMT pretraining improves over the KERMT baseline by 7.6%, 9.9%, and 9.5% respectively (averaged over significantly-improved endpoints). Adding ADME-adjacent molecules to the pretraining corpus further improves transfer, and the contrastive component sharpens chemically meaningful latent neighborhoods.

Biometric systems are increasingly deployed in security applications; however, they remain vulnerable to spoofing attacks, in which attackers exploit counterfeit biometric data to gain unauthorized access. This research evaluates the effectiveness of state-of-the-art machine learning models, MobileNetV2, DenseNet-121, Inception-v3, and Spoof Trace Disentanglement (STD) in detecting spoofing attacks within facial recognition systems. Using the CelebA-Spoof dataset, the study evaluates model effectiveness using metrics such as accuracy, precision, recall, and F1 Score. Cross-dataset validation is carried out on the MSU-MFSD dataset to assess generalizability. The results show MobileNetV2 as the most efficient model, achieving 92% accuracy while balancing computational effectiveness, making it appropriate for real-life applications. Inception-v3 shows moderate robustness, while DenseNet-121 and STD struggle with generalization. The findings highlight the need for advances in domain adaptation and hybrid architectures to enhance biometric security systems.

Language models can state that "the Earth orbits the Sun" and, when role-playing Aristotle, assert the opposite. Recent work argues that persona adoption is fundamental to how language models operate, with models constantly selecting the most appropriate persona for a given context. Does such role-playing merely change the model's outputs, or does it also affect what the model internally represents as truthful? We study this question with linear truth probes, applying them to LLMs role-playing historical personas whose likely beliefs differ from modern consensus. For each persona, we compare false claims the persona would likely have endorsed (*era-believed*) with topic-matched false claims they would not have endorsed (*era-false*). Across prompting, in-context learning, and supervised fine-tuning, persona induction suppresses era-believed statements less than equally false alternatives, yet they remain classified as false overall. Role-play therefore shifts what these models say more than what they internally represent as true. We contrast this with models trained on harmful advice that exhibit Emergent Misalignment (EM). Across three model families (Qwen 2.5 14B, Qwen 3 8B, and Llama 3.3 70B), their false claims move substantially toward the true region of probe space, are defended under challenge roughly half the time versus about a sixth for role-play, and are used in downstream reasoning. Role-play and Emergent Misalignment thus are points on a spectrum of belief internalization, where role-play changes what a model says with little representational change, while Emergent Misalignment shifts the internal representation of false claims without fully marking them as true.

The success of large-scale deep learning models in neuroscience is fundamentally constrained by severe data heterogeneity. Native fMRI data aggregated from diverse sources exhibit substantial variation in both spatial and temporal resolutions. Consequently, most existing frameworks rely on lengthy, rigid preprocessing pipelines that enforce uniformity across datasets. This practice introduces two critical limitations: (1) potential degradation of subject-specific anatomical information; (2) significant computational overhead, often requiring hours of processing per subject. Here, we propose FlexiBrain, a resolution-agnostic voxel-level encoding framework for native fMRI based on Mamba-JEPA. FlexiBrain defines patch sizes in real-world physical units and employs a dynamic patch resizing, thereby bypassing destructive spatial standardization while enabling direct ingestion of data in native space. We instantiate the framework using an efficient Mamba-JEPA backbone to model high-dimensional 4D fMRI signals. Across five diverse downstream neuroscience tasks, FlexiBrain consistently outperforms recent state-of-the-art methods, achieving gains of up to 12 percentage points without external data augmentation. Importantly, FlexiBrain functions as a seamless plug-in module, substantially reducing preprocessing costs and accelerating the development of robust voxel-level fMRI foundation models. Code is available at https://github.com/OneMore1/FlexiBrain.

The performance of modern language models depends critically on pretraining data composition. Yet existing data selection methods rely on auxiliary classifiers for document scoring or mixture optimization, adding computational overhead and dependence on labeled data. We propose WebGraphMix, a lightweight data selection framework that computes structural centrality scores over the Common Crawl host-level web graph and uses them to vary the proportion of central versus peripheral documents in the pretraining mixture. We hypothesize that central hosts expose models to reusable abstractions, while peripheral hosts encode specialized, long-tail knowledge. WebGraphMix computes centrality scores efficiently at web scale, requiring no model training, labeled data, or downstream supervision. We integrate WebGraphMix into the DataComp-LM pipeline and train models at 400M and 1B parameter scales with 8B and 28B tokens respectively, evaluating on 23 tasks ranging from factual knowledge to symbolic reasoning. Our experiments show that central and peripheral web regions encode complementary capabilities. Mixture combining both at a ratio of 1:1 achieves 41.4% on average, compared to 39.8% for uniform sampling. Combining structural scores with document-level quality classifier scores further improves performance to 43.8%. These findings demonstrate that web graph topology is a meaningful axis for pretraining data curation, capturing information that is largely orthogonal to existing content-based approaches.

Indoor localization from wireless measurements remains challenging in large-scale deployments due to substantial variation in building geometry, the set of detectable access points (APs), and the heterogeneity of received signals. Existing learning-based methods often perform well only in limited settings and degrade under environmental shifts, making robust anchor-free localization across diverse indoor environments notoriously difficult. In this paper, we present OmniLoc, an environment-interactive foundation model for anchor-free user equipment localization across diverse indoor environments. To the best of our knowledge, OmniLoc is the first foundation-model-based approach built directly on wireless measurements for this task. OmniLoc is built on three key designs. First, a unified input tokenization module converts heterogeneous wireless measurements into a common representation that is more amenable to learning. Second, a geometry-aware Transformer performs AP-aware feature extraction by emphasizing dominant APs while aggregating complementary evidence from supporting APs. Third, a geometry-aware location estimation module conditions regression on geometric embeddings to produce geometrically consistent location predictions. We evaluate OmniLoc on both a large-scale in-house dataset and a public benchmark dataset. Results show that OmniLoc significantly outperforms existing methods, consistently improves existing backbones when its design components are integrated, and demonstrates strong generalization in cross-environment evaluations.

Understanding and predicting how social beliefs evolve in response to events -- from policy changes to scientific breakthroughs -- remains a fundamental challenge in social science. Given LLMs' commonsense knowledge and social intelligence, we ask: Can LLMs model the dynamics of social beliefs following social events? In this work, we introduce the concept of the Social World Model (SWM), a general framework designed to capture how social beliefs evolve in response to major events. SWM learns state-transition functions for social beliefs by mining temporal patterns in social data and optimizing the evidence lower bound, without the need for explicit human annotations linking events to belief shifts, or for expensive census data. To evaluate SWM, we introduce a benchmark, SWM-bench, derived from real-world prediction markets, specifically Kalshi and Polymarket. SWM-bench includes over 12k data points for social belief prediction tasks spanning diverse domains such as politics, finance, and cryptocurrency. Our experimental results show that SWM significantly outperforms time-series foundation models, achieving state-of-the-art results on Kalshi data and demonstrating competitive performance on Polymarket data, while offering interpretable insights into the underlying mechanisms of social belief dynamics.

Federated continual learning (FCL) must learn from distributed task streams under limited resources, such as communication, computation, memory, and label availability. Existing FCL methods often rely on repeated local optimization, replay, and full supervision. Analytic alternatives avoid iterative training and replay, but using high-dimensional random features to improve accuracy requires a second-order feature statistic, the Gram matrix, which has a quadratic communication cost in the random feature size $M$. We propose FedRAN, a resource-aware analytic FCL framework that replaces gradient-based updates with compact random feature statistics. Each client transmits a truncated-SVD summary of its Gram matrix, reducing the dominant second-order upload from quadratic to linear in $M$ for fixed rank. The server performs a two-level QR-SVD subspace merge, spatially across clients and temporally across tasks, and solves a ridge classifier in closed form. FedRAN further supports label scarcity through prototype-based pseudo-labeling. Across CIFAR-100, ImageNet-R, and VTAB datasets, FedRAN improves average accuracy by up to 4.8 percentage points over the strongest baseline, uses 30.6-121.8$\times$ less per-client communication than optimization-based FCL, and is 190.3$\times$ faster on average than gradient-based baselines; with only 20% labels, pseudo-labeling improves average accuracy by up to 6.61 points. These results show that FedRAN enables accurate and resource-efficient FCL under communication, computation, and label constraints. The source code is available at https://github.com/JebacyrilArockiaraj/Fed-RAN-SSL.

Correcting handwritten exams by hand is time-consuming and error-prone, particularly for large cohorts, while fully digital exams tend to force a didactic narrowing towards closed question formats. A practical middle ground keeps paper-based, problem-oriented tasks but records the assessment-relevant answers as single capital letters in a table that a machine can read. The open question is whether this reading can be made accurate and, above all, fair enough for unsupervised grading. Earlier automated approaches reached only about 88%--91% recognition -- too low -- and failed on the cases that matter most: answers placed outside the cell, crossed out, or written in cursive. We show that general-purpose vision-language foundation models (VLMs), which interpret the page rather than match pixel templates, close this gap. On a benchmark of 61 anonymised exams (3141 answer positions) the best model reaches 98.4% accuracy, well above the previous baseline. Crucially, we centre the evaluation on fairness: we distinguish false negatives (a correct answer marked wrong, which disadvantages the student) from false positives, and a lightweight prompt that supplies the reference solution as context lowers the false-negative rate to 0.58%. Under an exemplary grading scheme only three of the 61 exams would be graded worse, all caught by a student self-review step. Fully automated, fairness-aware exam grading at scale is therefore defensible; we release the anonymised benchmark to support reproducibility.

Software engineering teams increasingly depend on GitHub issue threads to coordinate work, report bugs, and negotiate technical decisions, yet most repository health tools focus on code metrics and ignore the conversational dynamics that drive or stall development. This paper presents SentTrack, a dual-lens framework for detecting socio-technical bottlenecks from GitHub issue discussions. Applied to the AvaloniaUI open-source repository across approximately 9,000 issue threads, the framework addresses three questions: how to automate workflow-inefficiency detection from real-time conversational data, whether sentiment signals can surface risk earlier than traditional label-based methods, and how to isolate human narrative from machine-generated noise in mixed-media issue text. SentTrack combines two complementary pipelines. A horizontal pipeline translates raw issue reports into clean summaries using a large language model, extracts mid-level concern phrases, and clusters them through UMAP and HDBSCAN, producing 613 semantic clusters from the first 3,608 issues processed. A vertical pipeline applies the ABCDE collaborative interaction framework to classify each comment and infer thread-level outcomes. Across the full corpus, 49\% of threads ended in stagnation and only 13\% reached resolution, with the resolution gap identified as the dominant bottleneck signal. A weighted scoring engine that combines negativity, stagnation, resolution gap, and thread length gives maintainers an interpretable prioritization tool for high-friction discussions before they stall development.

In this paper, we study Mahalanobis-guided latent out-of-distribution (OOD) detection for test-time RL controller switching in nonlinear time-varying systems. RL controllers can quickly control high-dimensional systems within the training distribution, but their performance can degrade when time-varying dynamics produce unseen observations. We consider a combined ES--DRL controller, where RL provides fast in-distribution actions and bounded extremum seeking (ES) provides robust model-independent control under OOD operation. The key challenge is deciding when to switch. We train a variational autoencoder (VAE) on in-distribution beam-profile observations and use Mahalanobis distance in the VAE latent space to detect OOD beam profiles at test time. This OOD decision sets a binary switch that selects either the RL controller or the ES controller. We evaluate the approach in safety-critical particle accelerator control. In this setting, spatial magnet motion creates OOD beam profiles that were not seen during RL training. Visualization of the VAE latent space shows that the proposed method identifies this OOD scenario and provides an interpretable signal for switching between RL and ES in the combined controller.

Prior-fitted networks (PFNs) are a promising class of tabular foundation models that perform in-context learning, whereby the entire labelled training set is supplied as context, and predictions for test queries are produced in a single forward pass. However, the quadratically scaling self-attention mechanism in many PFN architectures makes inference prohibitive for very large training datasets. We propose CRUMB (Clustered Retrieval Using Minimised-MMD Batching), a three-stage inference wrapper that (i) clusters the test queries, (ii) selects a small, distributionally matched training subset for each cluster by greedily minimising the maximum mean discrepancy (MMD), and (iii) runs exact PFN inference on each reduced-context batch. CRUMB is architecture-agnostic and requires no retraining. On the 51-dataset TabArena benchmark, evaluated across three PFN architectures (TabPFNv2, TabICLv1, TabICLv2), we show that CRUMB outperforms similar state-of-the-art context selection strategies. We also show that CRUMB is resilient to covariate drift, as the MMD-minimisation step naturally helps align the training context distribution to match the current test batch distributions.

The expansion of the digital domain has resulted in a substantial increase in digital communication, with email emerging as one of the most prominent channels. The proliferation of email communication is apparent in both professional and personal contexts, thereby creating numerous vulnerabilities for malicious actors to exploit. Spam emails, a form of unsolicited correspondence often bearing malicious intent towards recipients, have been an ongoing challenge for email users since the inception of email technology, and this problem has been exacerbated by the growth of the digital landscape. Email spam filters are integral components of email clients, engineered to identify potentially harmful messages and alert users to their malicious content. Phishing, frequently the initial phase of malware-based attacks, is evolving rapidly, with malware becoming increasingly sophisticated over time. A widely adopted approach for detecting malicious activity within malware and spam domains is the application of machine learning. Our aim is to assess the impact of the evolution within the spam email domain on these machine learning-based detection systems and to explore strategies for mitigating associated performance degradation.

Large Language Models (LLMs) have achieved strong performance across natural language processing tasks, yet reliable reasoning remains an open challenge. Although modern LLMs show progress in structured inference, multi-step problem solving, and contextual understanding, their reasoning behavior is often inconsistent and sensitive to prompting strategies, task design, and model scale. This survey provides a systematic analysis of more than 300 recent papers from arXiv, Semantic Scholar, Google Scholar, Papers with Code, and the ACL Anthology to examine how reasoning capabilities emerge in LLMs and where they fail. We make three main contributions. First, we introduce a structured taxonomy of LLM reasoning research, covering Chain-of-Thought reasoning, multi-hop reasoning, mathematical reasoning, common sense reasoning, visual and temporal reasoning, code and algorithmic reasoning, retrieval-augmented reasoning, tool-augmented and agentic reasoning, and reinforcement learning-based reasoning. Second, we analyze methodological trends across these paradigms, including prompting methods, model architectures, training objectives, reward modeling, and evaluation benchmarks. Third, we synthesize recurring limitations and failure modes, such as reasoning hallucinations, brittle multi-step inference, weak causal abstraction, and poor cross-domain generalization. By organizing a rapidly expanding literature, this survey offers a unified view of the current capabilities and limitations of reasoning in LLMs. We also identify emerging research directions, including meta-reasoning, self-evolving reasoning frameworks, multimodal reasoning, and socially grounded reasoning. Overall, this work aims to serve as a reference for developing more robust, interpretable, and generalizable reasoning systems in future language models.

We study the task of density estimation, where we hope to accurately estimate a probability density from $n$ samples. A textbook method for density estimation in total variation distance is the minimum-distance estimator approach, where we conclude both the algorithm and the analysis merely from bounding the VC dimension of a particular concept class (the so-called Yatracos class). While this technique has originally yielded sharp guarantees primarily for total variation distance, in this work we extend the minimum-distance estimator approach for learning within Hellinger distance. Our main observation is that we may produce an analogous recipe for Hellinger (where we only require bounding the VC dimension of a related concept class) by drawing connections to recent results yielding reverse data processing inequalities. This recipe is flexible enough to accommodate fast algorithms originally designed for total variation distance; by modifying the approach of Acharya et al. (2017) we conclude the first near-linear time algorithm for learning classes including univariate mixtures of log-concave densities and mixtures of Gaussians (with arbitrary variances), with near-optimal sample complexity.

Accurate loss reserving is foundational to insurer solvency, yet accelerating climate driven catastrophes systematically violate the stability assumptions on which traditional actuarial methods depend. This white paper presents a research program testing whether Long Short Term Memory (LSTM) neural networks can detect and adapt to these structural breaks faster and more accurately than Chain Ladder, Bornhuetter Ferguson, and Cape Cod methods. Using 15 plus years of regulatory development triangle data from Florida and Louisiana, enriched with NOAA hurricane intensity indices and sea surface temperatures, we hypothesize a targeted improvement of 15, 20% in reserve accuracy for catastrophe exposed years, a threshold grounded both in the prior neural network reserving literature and in the formal convergence results developed here. Beyond empirical validation, we develop a theoretical framework grounding LSTM structural break detection in probabilistic terms, providing formal performance guarantees that compensate for the limited number of catastrophe events in the test period. We document the research design, methodology, expected contributions, and a candid assessment of limitations.

Safety assurance cases provide structured justifications that safety-critical systems meet their safety requirements. Recently, the notion of defeaters has emerged as a rigorous means of challenging the validity of safety arguments. Examples of defeaters might include overly strict claims, unreliable evidence, or reasoning gaps. However, defeaters remain ad hoc, lack structured support for critical reflection, are inconsistently described, are difficult to review, and lack documentation standards. To address this, we propose Defeater Cards, a new structured documentation artifact for systematically characterizing, reasoning about, and managing defeaters in safety cases. Drawing on a literature survey and thematic analysis, we identify documentation criteria that inform the card's structure, based on the 5W1H framework. Defeater Cards are designed to support informed analysis and evolution, improve traceability and auditability, and enable the reuse of defeater knowledge across systems and product variants. We demonstrate their applicability through two cross-domain case studies, showing how they expose hidden assumptions, surface reasoning gaps, and support ongoing safety assurance case evolution. To support adoption and community reuse, we also release an open-source repository of defeater cards as a baseline upon which researchers and practitioners can build and describe lessons learned.

Large Language Models are highly sensitive to prompt formulation, necessitating automatic prompt optimization to unlock their full potential. While evolutionary algorithms have emerged as the dominant paradigm, they suffer from a critical bottleneck: data efficiency. Current methods treat the development dataset as a static benchmark, wasting significant compute budget on uninformative data. In this work, we introduce APEX (Automatic Prompt Engineering eXpert), a novel framework that optimizes the data usage alongside the prompt search. APEX dynamically stratifies the dataset into Easy, Hard, and Mixed tiers based on the optimization lineage. By prioritizing the Mixed tier, which identifies the data where the LLM has mixed performance, we identify two high-leverage subsets: the addressable frontier for generating informative mutations and the rank-sensitive frontier for distinguishing candidate quality. We evaluate APEX across three diverse benchmarks: IFBench, SimpleQA Verified, and FACTS Grounding. Under a fixed budget of 5,000 evaluation calls, due to its data efficiency, APEX outperforms the initial prompt by an average of 11.2% on Gemini 2.5 Flash and 6.8% on Gemma 3 27B, demonstrating that a data-centric approach is key to efficient and effective prompt optimization.

The deployment of LLM-based agents in scientific analysis raises opposing concerns: that agents may reduce methodological diversity, or that they may amplify the analytic flexibility through which researchers reach motivated conclusions. We argue these worries target two empirically separable layers: a design layer of methodological choices, and a verdict layer in which a decision rule maps estimates to a substantive claim. We test both by running 20 independent executions of Claude Code and Codex on a prominent immigration and social-policy against a many-analysts human baseline. At the design layer, Codex matches human methodological diversity and Claude Code produces nearly three times as many specifications; both agents' effect estimates remain broadly aligned with the human consensus, and no agent model exactly matches any human model. A prompt-induced anti-immigration researcher prior reorganizes each agent's methodological decisions but, unlike for biased human analysts in the same data, does not shift aggregate estimates or final verdicts; nor do agents reroute along the methodological axes humans use to bias their estimates. At the verdict layer, an explicit confirmatory prompt flips Claude Code's verdicts from 10% to 90% support while leaving its coefficient distribution essentially unchanged, operating through rule omission rather than rule softening. AI agents can rival or exceed human methodological diversity at the design layer while remaining vulnerable at the verdict layer. In our setting, the locus of AI bias is not estimation but interpretation.

Recent anecdotal evidence suggests that AI coding agents can reproduce published findings when provided with original data and code; yet systematic evaluation across social sciences remains limited. Existing evaluation benchmarks are insufficient, either small or conflate agent performance with problems in the reproduction materials themselves, such as code that fails to execute correctly. Here we introduce SocSci-Repro-Bench, a benchmark of 221 tasks spanning four disciplines and 13 substantive domains, constructed from studies whose results are either fully reproducible with available materials or demonstrably non-reproducible due to missing data, allowing us to isolate agents' reproduction capacity. Evaluating two frontier coding agents, Claude Code and Codex, we find that both can reproduce a large share of social science findings, with Claude Code substantially outperforming Codex. These reproduction rates considerably exceed those previously reported for general-purpose LLM-based agents on comparable reproducibility benchmarks. Both agents also perform strongly on a reasoning task requiring identification of underlying research questions, and additional analyses suggest that results are not primarily driven by memorization. Providing the original paper PDF alongside replication materials modestly improves performance but introduces bias on tasks where reproduction is impossible. We also show that agents can be nudged toward confirmatory specification search through subtle prompt framing. Together, these findings suggest that at least some frontier coding agents can serve as reliable executors of computational workflows while underscoring the need for careful benchmarking and prompt design as AI systems assume larger roles in scientific production.

Trust in an AI system is often anchored by explanations of how it works, which one then uses to forecast its behavior on new inputs. For large reasoning models (LRMs), this conventional route is particularly difficult to follow: explanation methods for single token generations do not naturally generalize to long trajectories, and the trajectories themselves are often not faithful when read as natural language. We propose an alternative that bypasses the explanation step: treat behavior forecasting as a learnable task and train Behavior Forecasters that operates on a single reasoning trajectory to make the same forecasts one would typically seek from an explanation. The forecaster's training data is obtained by querying the LRM with no human annotation, and its inference is done in a single forward pass. We instantiate this approach on two tasks: how likely the LRM is to repeat its answer on re-runs, and how removing parts of the input changes its answer. We evaluate this approach on both tasks across three diverse reasoning datasets and find that trained Behavior Forecasters are more accurate than GPT-5.4 and Claude Opus-4.6 reading the same trajectories as naive readers, at a small fraction of their inference cost. We find that fine-tuning the backbone end-to-end and initializing it from the target LRM are each necessary for strong performance. These results show that the reasoning trajectory carries information about the LRM's future behavior that goes beyond what naive reading conveys.

Graphical model editing is shifting from desktop applications to web-based tools. We analyze the characteristics of existing frameworks and, based on this analysis, we derive a set of design principles that imply low-effort generation, extensive customization possibilities, and straightforward deployment of the resulting editors. On these grounds, we introduce EMFular, a purely web-based framework for managing EMF models without any backend. The accompanying EMFular generator maps a given Ecore model (an EMF metamodel) to a ready-to-use and ready-to-customize graphical editor. EMFular editors provide 'EMF consistency', that is, they not only support standard modeling operations such as creation, inspection, navigation, editing, and undo/redo, but they also handle containment and inverse references in close alignment with EMF; they also provide interoperability with existing EMF tooling through compatible de-/serialization. A generated editor is an Angular project with designated extension points, which allows developers to customize and extend all aspects of the editor using the expressive power of Angular and its ecosystem, guided by the extension points of EMFular. We evaluate EMFular in terms of editor adequacy (available editing capabilities), adaptability (customization mechanisms and required effort), and robustness of the generation.

Existing multi-agent LLM orchestration methods, ranging from brute-force ensembles to learned routers, select models and topologies based on task and model features. However, these methods do not consider the runtime state of the serving infrastructure. On shared GPU clusters under concurrent load, this infrastructure blindness causes systematic resource underutilization: preferred models accumulate deep request queues while equally capable alternatives sit idle. In multi-agent pipelines, where each query triggers multiple sequential model calls, these delays then compound across every downstream step. Closing this gap is challenging because the relevant infrastructure signals (queue depths, KV-cache pressure, latencies) are dynamic and noisy, and they must drive three different decisions: planning, per-step routing, and scheduling. We introduce INFRAMIND, a framework that makes the entire multi-agent stack infrastructure-aware. An infra-aware planner conditions topology and role selection on real-time system load and remaining budget, biasing toward simpler graphs under congestion and richer ones at low load. An infra-aware executor then observes per-model queue depths, cache utilization, and response latencies at each agent step to decide which model to call and how deeply to reason; a budget-aware scheduler further reorders each model's queue so that urgent requests are served first. Cast as a hierarchical constrained MDP and solved end-to-end via reinforcement learning, the system learns to balance quality against latency automatically. Across five benchmarks, INFRAMIND delivers up to +7.6 pp accuracy over the prior baseline at low load with up to 7x lower latency, and sustains up to 99.9% SLO compliance under high load where every baseline drops below 50%.

Efficiently sampling from a complex probability distribution is a fundamental problem which has become increasingly pertinent in recent years with the rise of generative AI, as sophisticated sampling procedures from LLMs have been proposed to solve challenging reasoning problems. The efficacy of such sampling algorithms is limited, however, by the relationship between the LLM and the particular sampling task at hand, which has motivated the framework of test-time training (TTT). TTT works by updating a model's weights in response to partial generations and reward feedback received at inference time, thus adapting to the particular problem. In this work, we propose a formalization for TTT as the problem of producing a sample from a given probability measure $μ^\star$ belonging to a known class ${F}$ of distributions, given an oracle $\hat μ$ which yields approximate density estimates for $μ^\star$. This is closely related to the problem of reducing sampling to approximate counting studied in seminal works of Jerrum, Valiant & Vazirani (1986) and Jerrum & Sinclair (1989): namely, when ${F}$ is the class of all distributions, it coincides exactly with the aforementioned counting-to-sampling reduction. In this paper, we first show a quadratic lower bound on the query complexity of sampling from $μ^\star$ given query access to $\hat μ$ (for sufficiently large classes ${F}$), thus showing that the random walk approach proposed by Jerrum & Sinclair (1989) and refined by Hayes & Sinclair (2010), is optimal. This answers an open question posed by Hayes & Sinclair. We then show that this lower bound can be circumvented if the size of ${F}$ is bounded appropriately. As we discuss, this latter result can be viewed as an abstraction of TTT, and thus represents a starting point for the development of a principled theoretical framework for TTT.

The growth of agent skills has transformed how agentic systems are built, evaluated, and deployed. As skill libraries continue to scale, rigorous evaluation becomes critical to ensuring their utility, quality, and safety in real-world applications. Consequently, the field is undergoing an emerging paradigm shift from isolated skill creation to automated, evaluation-driven skill evolution. In this survey, we systematically examine the landscape of skill evolution and evaluation beyond foundational skill creation. We categorize evolution into four distinct paradigms, spanning execution feedback, trajectory distillation, compression, and reinforcement learning, showing how each element contributes to improving skill utility and reliability. We also provide an analysis of six skill-centric benchmark categories, identifying structural gaps in benchmark coverage, trade-offs, and metric richness to advance skill research. Finally, we identify open directions for building skill ecosystems that are generalizable, efficient, and verifiably safe. The project URL is https://github.com/Cassie07/AgentSkill_Survey

Mirror Descent (MD) extends Gradient Descent (GD) beyond Euclidean geometry and has recently reappeared as a lens for KL-regularized policy optimization in reinforcement learning and LLM post-training. This raises a basic robustness question, crucial to reproducibility and reliability: how sensitive are MD dynamics to their inputs? We focus on initialization, often itself a pretrained or previously aligned model. Quadratic-regularized MD, including GD and Mahalanobis geometries, is well-known to be stable for convex smooth objectives. We show a sharp contrast: once the regularizer is non-quadratic, MD can be exponentially more sensitive to initialization than GD, even with a well-conditioned regularizer in Euclidean norm. We give a three-dimensional construction with a convex, smooth objective and a strongly convex, smooth, well-conditioned regularizer where an initial $\varepsilon$ perturbation is quickly amplified to $\min\{\text{polylog}^{-1}(1/\varepsilon), \varepsilon e^{Ω(ηT)}\}$ after $T$ iterations of MD with step size $η$. For canonical KL-regularized MD on the simplex, we show that even linear objectives can amplify an initial $\varepsilon$ perturbation exponentially fast in high-dimensional or near-boundary regimes. Finally, we show that adding a Bregman regularization term toward an anchor point can stabilize the dynamics while largely preserving the optimization guarantees, and that the choice of anchor is crucial: anchoring at the initialization only partially mitigates the instability, whereas anchoring at a fixed point yields a more stable mechanism.

Mathematical knowledge is split between bibliographic databases (e.g., MathSciNet, zbMATH Open) and formal proof libraries (e.g., Lean mathlib), preventing unified access between published results and their formalizations. We propose a relational bridge-database that aligns publication metadata with formal artifacts, providing an interoperability layer between mathematical literature and machine-verifiable proofs. We introduce a paper-level formalization score that measures how much of a publication is covered in formal systems. As a feasibility study, we show how such scores can be estimated via cross-document alignment between informal texts and Lean formalizations, enabling large-scale analysis of formalization coverage. This framework is a first step toward integrating bibliographic and formal mathematical ecosystems into scalable, machine-actionable knowledge graphs linking publications to formal proof objects.

Speech foundation models often struggle in low-resource domains due to domain mismatch and data scarcity. We propose Gumbel-BEARD, a domain adaptation framework that automates Whisper encoder layer selection via an end-to-end trainable hard Gumbel-Softmax selector. It enables self-supervised adaptation with a BEST-RQ objective that dynamically adapts to target acoustic characteristics without manual tuning. Experiments on the MyST child speech corpus demonstrate efficiency and scalability: with 10 h of labeled data for fine-tuning, our method matches a fully supervised baseline trained on the complete 133 h labeled set. We establish new state-of-the-art word error rates (WERs) of 8.21% using Whisper-medium on MyST and 11.06% using Whisper-small on the OGI Spontaneous dataset. Evaluation on CORAAL further confirms robustness to adult dialectal domain shifts, with up to 6% relative WER reduction, highlighting the generalizability of our approach to diverse low-resource conditions.

Jailbreak attacks expose persistent safety weaknesses in large language models (LLMs), but existing stateless single-turn methods face a trade-off: hand-crafted prompts are expressive but static, while iterative prompt optimization can adapt but often relies on low-level mutations that require many target queries. We propose JailbreakOPT, a tool-assisted framework for improving iterative single-turn jailbreak prompt optimization. JailbreakOPT organizes diverse atomic jailbreak prompts into an attack tool library and composes them through a unified intra-episode optimization abstraction to generate stronger standalone attack prompts. To reuse experience across attack episodes, JailbreakOPT further frames tool selection as a contextual bandit problem and applies contextual Thompson sampling to guide exploration and exploitation based on past outcomes. Experiments across multiple target LLMs and attack goals show that JailbreakOPT improves attack success rate (ASR) while reducing the number of attacks until success (No.A) compared with atomic single-turn attacks and existing iterative optimization baselines. This paper may contain offensive or harmful content.

Natural language interfaces to databases aim to translate user questions into executable SQL, yet remain brittle in real-world settings where questions are underspecified and schemas are large and ambiguous. Ambiguity across user questions, database schemas, and model interpretations are central failure modes in NL2SQL, leading to misaligned intent, incorrect schema grounding, and erroneous SQL generation. Existing approaches rely on human clarification or treat ambiguity as a schema representation problem, but these do not scale nor resolve ambiguity autonomously. We propose SOMA-SQL to automatically resolve ambiguity via targeted synthetic query log and ambiguity-driven probing. SOMA-SQL constructs synthetic query log to ground schema interpretation and guide candidate SQL generation; it then executes targeted probing queries, driven by a structured ambiguity taxonomy and candidate disagreements, to produce disambiguation evidence for final SQL selection and repair. This active approach to ambiguity discovery and resolution generalizes across unseen schemas and query distributions without human-in-the-loop. Experiments on six public benchmarks demonstrate that SOMA-SQL improves execution accuracy by 13.0% on average over state-of-the-art baselines, with gains of up to 16.7% on ambiguous questions.

Every day, millions absorb claims from podcasts and streams that no fact-checker ever sees. Spoken misinformation is built through conversation, where credibility comes not from facts alone but from how claims are framed, reinforced, or left unchallenged across turns. Yet fact-checking has focused on isolated text, leaving dialogue audio under-studied. We introduce MAD2, a new Multi-turn Audio Dialogues benchmark for spoken claim verification, containing 1,000 two-speaker dialogues with 3,368 check-worthy claims and approximately 10 hours of audio, and propose calibrated multimodal fusion of a context-aware audio encoder and a dialogue-aware text model. Across settings, adding dialogue context improves verification, but the gains depend on scenario type. Using only preceding context often matches offline performance, supporting live-moderation settings, and audio contributes most when transcript-based models are destabilized by additional context. Overall, conversational structure matters more for verification than misinformation framing.

Compression progress is a long-standing proposal for intrinsic motivation: reward an agent when its world model becomes better at predicting or compressing experience. The folk claim is that this reward is "credible" because it is paid only for learning. We make this precise and prove it. If intrinsic reward is the signed decrease of a fixed sealed-audit loss, r_t = E(theta_{t-1}) - E(theta_t), then cumulative reward telescopes exactly to endpoint audit improvement, so no policy can push reward up indefinitely while true audit performance stagnates or degrades. For finite audit panels the same result holds with a sharp false-positive budget: cumulative empirical reward is at most true audit improvement plus 2 Delta_n(F, delta), the uniform audit deviation of the model class. This is horizon-free: adaptivity over time costs nothing once the sealed panel uniformly controls the class. The theorem also identifies the failure modes: the guarantee disappears if progress is clipped, scored on the agent's own stream, exposed to a high-capacity model on a reusable panel, or applied to a neural class that makes Delta_n vacuous. We give a Lean 4 mechanization of the structural core (telescoping, the finite-audit bound, finite Gibbs, and the entropy floor) and an experiment suite on ARC-TGI grid-transformation generators with adaptive holdout attacks. Experiments confirm the theory: finite-audit deviation scales as n^{-0.527}; signed progress resists clip-farming, stream leakage, and noisy-TV curiosity; naive reusable audits are exploitable by black-box scalar feedback, while standard release defenses keep the attack below the 2 Delta_n threshold. Signed compression progress on a sealed audit is an accounting signal of genuine improvement.

Repository-level benchmarks for evaluating Large Language Model (LLM) code repair on Secure Multi-Party Computation (MPC) software do not yet exist, and directly transplanting general-purpose benchmarks such as SWE-bench fails on three structural fronts: (i) MPC repositories are dominated by generic Python infrastructure rather than cryptographic logic; (ii) high-value MPC fixes lack the standardized tests rigid extraction pipelines require; and (iii) standard fail-to-pass evaluation is insufficient for code that must also be cryptographically safe. MPC is increasingly deployed for privacy-preserving machine learning, biomedical collaboration, and secure analytics. Existing MPC-specific code-synthesis efforts cover only operator-level or single-framework tasks; evaluating LLM agents on real repository-level MPC repair instead demands MPC-aware data curation and a verifier matched to the security and numerical-fidelity guarantees MPC programs must obey neither of which existing benchmarks provide. We introduce MPC-Patch-Bench, a repository-level benchmark organised around two frameworks. (1)The Data Curation Framework combines a domain-specific curation agent that filters raw pull requests through three cryptographic layers with a human-AI completion engine that synthesizes missing problem statements and Fail-to-Pass/Pass-to-Pass tests, yielding 205 fully verified instances. (2)The MPC Verifier provides dedicated security and numerical-fidelity checks via dynamic differential testing against plaintext oracles and MPC-specific static analysis rules that flag unsafe reveals, insecure arithmetic, and illegal public/private casts. The strongest evaluated LLM functionally resolves only 22.9% of MPC-Patch-Bench tasks; the MPC Verifier further reduces verified resolution to 17.1%, with up to 40% of functionally-passing patches rejected for cryptographic or numerical-fidelity violations.

Neural recordings are often interpreted as local measurements, yet the signal at any one sensor can also reflect structured activity distributed across the broader network. This raises a basic question: to what extent does an electrode's signal reflect local versus distributed information in the underlying system? More specifically, how much of an electrode's activity is carried by its immediate neighborhood, and how much is embedded more broadly across the array? We address this with a Spatially Masked Regression (SMR) framework that reconstructs each electrode's timeseries from the remaining electrodes while excluding a configurable neighborhood around the target. By progressively increasing this mask, spatial locality becomes an experimental control for quantifying how much predictive information survives after nearby channels are withheld. We apply SMR to intracranial EEG with heterogeneous electrode coverage and to scalp EEG with standardized montages over sensorimotor cortex. Using distance correlation between original and reconstructed signals, we find strong within-subject reconstruction in both modalities, substantial residual predictability even when local neighbors are excluded, and markedly stronger cross-subject transfer in EEG than in iEEG. Masking shows that nearby electrodes contribute strongly to reconstruction but do not account for all of it, indicating that individual channels reflect both local redundancy and broader distributed structure. Surrogates that preserve selected marginal or spectral properties while disrupting phase structure or temporal ordering substantially reduce performance, supporting the conclusion that SMR depends on structured temporal and cross-channel organization rather than on marginal statistics alone. These results position SMR as an interpretable framework for quantifying the balance between local and distributed information in recordings.

Adversarial robustness evaluations of large language models (LLMs) typically report attack success rate (ASR) under fixed query budgets, implicitly treating all attacks as equally costly. In practice, the computational expense of different attack strategies can vary by orders of magnitude. Consequently, ASR at a fixed budget can obscure the true effort required to jailbreak a model, thereby making it hard to determine whether an attack's cost justifies its payoff to the attacker. We propose a compute-aware evaluation framework based on computational pressure, measured in cumulative floating-point operations (FLOPs), as a proxy for adversarial effort. We introduce risk-compute curves, which map compute budgets to attack risk, and derive two metrics that summarize the average pressure required for a given attack to succeed. Across ten models spanning three families and four different stages in language model training and alignment, evaluated with three attack strategies (gradient-based, iterative refinement, and template-based) on two jailbreak robustness benchmarks, we find: (1) alignment training has non-monotonic effects on compute-space robustness; (2) scaling model size reduces gradient-based attack effectiveness but has limited impact on cheaper template-based attacks; (3) gradient-based attacks optimized on a surrogate model can transfer to a separate target model, providing a way to reduce attacker costs; (4) compute cost varies by up to ${\approx}5{\times}$ across harm categories within a single model; and (5) safety-aligned RL increases aggregate cost while leaving some categories disproportionately accessible. We release our framework to enable compute-aware risk assessment and evaluation.

Large Audio-Language Models (LALMs) excel at audio understanding but expose little about where in an audio signal they attend. We introduce instruction-based vector steering, which constructs a steering vector by contrasting activations from differently instructed prompts while keeping the audio fixed. Through a systematic probe of LALM attention, we find that - unlike standard prompting or audio-based steering - this intervention significantly redistributes the temporal attention allocated to audio tokens, concentrating it on acoustically relevant regions. We then show that this attention shift is behaviorally meaningful: in a controlled three-event setting, reading out the temporal position of maximal steering-induced attention change recovers the location of a queried sound event without any training, attaining 60.87% and 68.72% overlap with ground-truth intervals on Qwen2-Audio and Audio Flamingo 3, far above direct prompting (31.84%, 46.75%) and random baselines (27.74%). Our results characterize a mechanistic property of instruction-based steering in LALMs and provide a training-free probe for the latent temporal structure these models encode.

Large Language Models (LLMs) are deployed across cultural contexts but often reflect homogenized values inherited from training data. Evaluations of cultural alignment typically rely on direct prompting with survey-style questions, which frequently elicit neutral or safety-aligned responses and fail to capture underlying model preferences. We propose a framework for probing and steering latent cultural representations in LLMs along the two Inglehart--Welzel axes of the World Values Survey (WVS). By translating social value questions into scenario-based behavioral dilemmas, we extract token-level probabilities to measure implicit values and apply activation steering, optionally combined with country-conditioned prompting, to shift model behavior without retraining. Across three open-source LLMs and four target cultures, we find substantial variation in steerability and identify latent entanglement, where interventions along one cultural dimension induce shifts along another. This coupling mirrors correlations in human WVS data and persists across activation, prompt, and hybrid steering. It constrains axis-independent alignment, though general task performance is largely preserved.

Tensor Product Representations provide the structural fidelity required for symbolic reasoning in models but suffer from exponential dimensionality growth when encoding deep recursive structures. Conversely, Vector Symbolic Architectures maintain constant dimensionality but sacrifice capacity and fidelity due to noisy compression via superposition. In this work, we propose Orthogonal Subspace Carving (OSC), a memory architecture that binds fillers to roles by projecting onto the null space of the role basis before aggregating into a fixed order-p tensor. OSC uses projections to enforce geometric orthogonality between bound structures within a static memory trace. We show that this mechanism decouples the tensor order from the structural depth, enabling deep recursive binding within a constant memory footprint. By performing retrieval via recognition, this construction allows for component vectors that are orders of magnitude smaller than the memory tensor, giving superior memory efficiency in settings involving high superposition. We also show that TPR is a special case of binding in Clifford algebra, and give a Clifford formulation of OSC.

Gaussian splatting methods have become increasingly popular for neural reconstruction of the real world. However, they are often limited in scale and resolution due to compute and memory constraints. We present a multi-GPU Gaussian splatting approach that scales reconstruction to higher resolutions and larger scenes while abstracting away the code complexity typically associated with distributing a model. To accomplish this, we propose a PyTorch backend that distributes the Gaussian parameters and splatting operators across GPUs via CUDA unified memory and NVLink. Because distribution occurs at the operator level, the model code requires no explicit cross-device communication. More broadly, the backend exposes multiple GPUs as an aggregate PyTorch device and supports other PyTorch operators. We demonstrate city-scale reconstructions with street-level detail consisting of over 1 billion Gaussian splats, more than 25 times as many as the current state of the art.

Short pretraining runs can reduce experimental cost, but they can also over-promote configurations that only look strong at tiny budgets. We study an auditable staged-promotion protocol for a fixed micro-pretraining runner on two heterogeneous host blocks: Windows A100 and Linux L40S. Starting from twelve prior-screened configurations, we use staged budgets of 2 minutes, 5 minutes, 10 minutes, 60 minutes, and 12 hours, with frozen promotion rules before expensive continuations. The early screens are intentionally treated as unstable: the 5- and 10-minute rankings are host-sensitive, and the eventual 12-hour top-ranked condition is not the mean-best condition at the replicated 10-minute gate. Because seed ranges differ across stages, these changes are operational promotion evidence, not within-seed curves. A replicated 60-minute gate keeps the Staged Factorial Screening bridge reference in the promoted set, where it ranks first in all four 60-minute host-seed cells. In the final 12-hour confirmation package, the bridge condition ranks first in all four host-seed cells across two seeds; the greedy comparator does not meet the frozen 0.010 val_bpb near-equivalence rule; and the cheaper d8/ar48 (depth-8, aspect-48) sentinel does not meet the frozen 0.020 mean-gap rule. The executed 12-hour branch spends 144 GPU-hours, and the full staged protocol records 169.2 training GPU-hours including screening stages. Continuing all four 60-minute candidates would spend 192 GPU-hours, while continuing all nine replicated 10-minute candidates would spend 432 GPU-hours. The latter numbers are accounting counterfactuals for unrun continuations, not evidence that skipped candidates could not have overtaken the reference. The result is a bounded cost-allocation finding, not a claim of global optimality, capacity-normalized superiority, or superiority over adaptive hyperparameter optimization methods.

Full-duplex spoken language models (FD-SLMs) enable seamless speech interaction by allowing models to listen and speak simultaneously, yet the internal mechanism by which they coordinate listening and speaking remains underexplored. We analyze the predictive behavior encoded in FD-SLM hidden representations and find that they exhibit stream-specific predictive patterns: during listening, they preferentially predict the incoming user stream, whereas during speaking, they preferentially predict the model output stream. Building on this observation, we show that FD-SLMs dynamically modulate their internal predictive focus between two states: a generative state aligned with model output generation and a perceptive state aligned with incoming user input. However, this modulation can lag behind abrupt changes in conversational context. During user interruptions, the model remains transiently biased toward the generative state before transitioning into the perceptive state, causing it to miss the beginning of the incoming input. We term this delayed internal transition state inertia. To quantify its downstream impact, we introduce the Zero-Buffer Benchmark (ZBB), a diagnostic benchmark for evaluating immediate interruption comprehension when user speech begins abruptly. We evaluate this setting using response correctness and initial-word occurrence rate (IWOR). Finally, we mitigate state inertia through activation steering with a perception vector, a training-free intervention with little additional computational overhead. Across multiple state-of-the-art FD-SLMs, activation steering substantially improves interruption handling; for example, on PersonaPlex, it improves correctness from 28% to 45% and IWOR from 40% to 72% without any fine-tuning.

Deep learning models facilitate the discovery of molecules with tailored properties among billions of candidate compounds. However, the computational burden to develop and deploy state-of-the-art models continuously increases, limiting their scalability. Most large-scale models are unimodal in nature and overlook the potential to leverage complementary molecular data modalities. To address these shortcomings, this paper introduces the Graph-Language Alignment for Chemical Inference and Exploration using Representations (GLACIER) model, a student-teacher framework that integrates molecular graphs, SMILES strings, and physicochemical descriptors to learn rich molecular embeddings. Our framework consists of three stages: (1) we pretrain three student encoders on 100,000 drug-like molecules: a message-passing neural network for molecular graphs, a transformer-based encoder for SMILES strings, and a multilayer perceptron for physicochemical descriptors, (2) we fuse these student modalities using a novel Finsler geometry-aware module, and (3) distill complementary knowledge from large teacher models, including MiniMol and MolFormer, into a single lightweight model via contrastive learning. We demonstrate that GLACIER is a robust framework that delivers high predictive performance and computational efficiency in complex molecular property prediction tasks. Our code is publicly available at https://github.com/eemokey/glacier.

Pre-mediation, the preparatory phase preceding direct human negotiation, plays a critical role in achieving mutually beneficial agreements, yet is often omitted due to cost, time, and limited access to trained mediators. We introduce an automated mediator for human negotiation, implemented as a structured pipeline of LLM modules, that supports pre-mediation in integrative negotiation settings. The pipeline decomposes preparation into specialized modules for dialogue, preference prediction, response-level critique, and structured summarization, separating inference, generation, and evaluation to address limitations of monolithic single-prompt approaches. We use the term "agent" for each module following common LLM-systems terminology, but the components are not autonomous and do not interact peer-to-peer; outputs are passed forward in a fixed sequence. We evaluate the system in two controlled human-subject experiments comparing AI-based pre-mediation with professional human mediators in a multi-issue negotiation scenario. On short-term self-reported measures, the automated mediator achieves preparation outcomes broadly comparable to human mediators, including trust in the mediator and confidence in reaching mutually beneficial agreements, while achieving substantially lower error on the preference-inference task under our scenario and prompts (36% lower RMSE). A second study shows that targeted prompt refinements reduce excessive affirmation patterns from 36.6% to 16.8%, matching human mediator baselines. Our findings suggest that structured LLM pipelines can provide scalable, low-effort pre-mediation support broadly comparable to human mediators on short-term self-reported preparation outcomes. The pipeline's single-party design mirrors how human mediators run pre-mediation today and enables parallel deployment across all parties to a dispute, supporting scalability.

Standard linear probing declares a property "encoded" when a classifier on hidden states achieves high accuracy. The protocol works well on a snapshot but breaks across pre-training: probe accuracy saturates within the first few thousand steps, leaving most of training invisible to the instrument. We introduce fragility, a complementary per-layer metric defined as the activation-noise level at which probe accuracy collapses. Fragility is sensitive to both the margin of separability and the redundancy of representation, both of which keep evolving long after accuracy plateaus. Applied to open-checkpoint language models, fragility recovers structure that accuracy alone cannot see. Moralized representations emerge along a lexical $\to$ compositional gradient: lexical moral detection first, compositional moral encoding later. Because probe accuracy on its own tracks how lexically separable a dataset is, we establish the compositional encoding directly, by showing it transfers across construction types that share no contrast tokens. A layer-depth robustness gradient develops monotonically across training while accuracy stays flat. And matched fine-tuning corpora that produce identical probing accuracy leave distinct fragility fingerprints, showing that data curation reshapes probe robustness without changing probe accuracy. In every comparison we test, where probing accuracy returns a flat answer, fragility returns a structured one.

Spoken language, whether produced by humans or large language models (LLM), unfolds over time with varying semantic content. However, we still lack simple, interpretable time-series features that capture how generic versus specific content is distributed over time, and that can be used to compare human and AI-generated speech. We introduce a semantic-timescale analysis pipeline that turns word-level transcripts with timestamps into semantic time-series. For each spoken narrative, we compute (i) semantic specificity using WordNet-based word depth and (ii) contextual similarity using SBERT embeddings and quantify their temporal dependence using autocorrelation-window measures (ACW-0 and related metrics). We then compare original speech to multiple shuffled controls that selectively disrupt lexical identity, temporal order, and word duration. Across human-read autobiographical narratives, TTS readings, and LLM-generated texts rendered with TTS, we find that segments with longer ACW-0 in the semantic time-series tend to contain more generic vocabulary, whereas segments with shorter ACW-0 are enriched in more specific words. These associations are strongly attenuated or abolished when word order and timing are randomized, indicating that ACW-based measures capture non-trivial temporal organization of semantic content beyond static lexical distributions. Our results suggest that ACW-based semantic timescales are a useful family of features for analyzing and comparing the temporal structure of human and AI-generated speech.

Structured abstracts are important for biomedical literature processing, by facilitating information retrieval, text mining, and knowledge synthesis. However, a vast portion of abstracts indexed in PubMed remain unstructured, presenting a significant bottleneck for downstream text-processing workflows and applications. To resolve this limitation, we introduce Structured PubMed, a comprehensive corpus of section-labeled biomedical abstracts compiled from the complete PubMed database, encompassing over 23.2 million research-article records. The corpus is divided into two distinct subsets: a collection of 5.9 million author-structured abstracts parsed from official XML files, and an automatically labeled collection of 17.2 million originally unstructured abstracts structured via a verbatim-extraction Large Language Model pipeline. Every record is harmonized under a unified five-section schema and mapped to its original PubMed identifier, publication type, and publication date. This dataset can be utilized to train sentence-classification models, benchmark text-segmentation architectures, and perform large-scale, section-specific information extraction at an unprecedented PubMed-wide scale.

With the growing demand for on-device LLM inference, edge SoCs increasingly integrate NPUs to improve performance and energy efficiency under tight power and thermal budgets. However, practical LLM deployment on current client NPUs remains difficult: widely used quantization formats such as AWQ do not map cleanly onto many existing NPU software stacks, which are often proprietary and expose limited low-level control. In this work, we present \textit{TileFuse}, a close-to-metal mixed-precision kernel library for AMD XDNA2 NPUs that targets transformer linear layers in quantized LLM inference. TileFuse brings practical low-bit formats such as AWQ-style W4A16 and W8A16 directly onto XDNA2, rather than forcing the model to be reshaped around an NPU-specific quantization scheme. TileFuse co-designs weight layout, metadata placement, mixed-precision microkernels, and array-level dataflow. Specifically, it fuses unpacking, dequantization, and GEMM/GEMV execution into a single kernel flow, introduces an interleaved pre-tiling layout that supports GEMM dimensions up to 32K, and redesigns GEMV dataflow to utilize the full 4x8 AIE array. Across kernel-level evaluations, TileFuse improves performance by up to 121.6% for GEMM and 281% for GEMV over full-precision baselines, while delivering more than 2x performance and energy-efficiency gains over strong iGPU baselines on GEMM. In end-to-end LLM experiments on Ryzen AI laptops, TileFuse achieves up to 2.0x lower prefilling latency with more than 64.6% lower energy consumption. Together, these results show that XDNA2 is a practical target for AWQ-style edge LLM inference and that native NPU support for off-the-shelf quantization can make NPUs substantially more usable in real client deployments.

Large language models (LLMs) can translate and modify source code, and have been shown to do so for codes of different complexity. Whether they can port a complete, production geophysical model to a different language without degrading its physics has not been established. We demonstrate that LLM-assisted code translation can preserve the physics of a complete production ocean model while moving it into a modern performance-portable form. We report our experience using an agentic LLM coding assistant, directed by domain experts, to port the FESOM2 unstructured mesh ocean--sea-ice model (about 74000 lines of core Fortran) first to C and then to C++/Kokkos for performance portability across CPUs and GPUs. We describe the practices that proved necessary, what worked and what did not, and the failure modes that we encountered. Three practices mattered most: translating in two stages that separate reproducing the numerics (Fortran to a clean C reference) from introducing parallelism (C to Kokkos); requiring a strictly literal translation in which the assistant was not permitted to ``improve'' the source; and validating each stage against an acceptance criterion suited to it. The C port reproduces the original Fortran at the level of long-term simulation statistics over five years. The Kokkos port is bit-for-bit identical to the C reference on CPU and statistically close on GPU over multi-year runs. On eddy-rich meshes up to 7.4 million surface vertices a single A100 GPU node runs 1.6--3.7 times faster than a CPU node, reaching the 1-2 simulated-years-per-day required for production integrations. The result is more than a single GPU port: by following a clear validation procedure, an LLM moved a full Fortran ocean model into another language and onto accelerators while preserving its physics in a matter of weeks.

Large Language Model (LLM) offers opportunities to enhance Human-Building Interaction (HBI) by enabling more direct interactions through intuitive interfaces to complex building systems. These systems can be characterized by the vast amounts of data across multiple formats, the lack of nonconfidential and generalizable information, and the requirement of domain expertise for interpretation. Applying LLMs to domain-specific tasks like HBI presents additional challenges. Limited training data makes traditional fine-tuning approaches impractical. Meanwhile, the opacity of LLM training data requires careful integration of domain knowledge to ensure reliability. Additionally, different LLMs exhibit varying alignment characteristics, suggesting that achieving both natural interaction and technical accuracy requires a multi-agent approach. These challenges highlight the need for innovative approaches to adapt LLMs for specialized domains while maintaining accuracy and user engagement. In this paper, we develop a hierarchical multi-agent framework that utilizes semantic routing and programmatic reasoning to decouple natural language understanding from building analytics. Instead of standard RAG approaches, our system employs a "Doorman" mechanism for task decomposition and specialized coding agents that generate executable Python scripts for precise arithmetic. We validate this framework on a dataset from more than 200 commercial buildings. Results demonstrate the effectiveness in providing accurate and contextual responses for diverse users, including stakeholders, from tenants to building managers, across various building system applications.

Retrieval-augmented generation degrades when scaled to large, heterogeneous document collections, where dense similarity loses discriminative power, and top-k retrieval increasingly returns semantically similar but contextually incorrect chunks. We refer to this failure mode as vector search dilution. Even when using hybrid dense+sparse retrieval, we observed this firsthand in a deployed Wyoming Department of Transportation corpus, where scaling from 54 to 1,128 documents (88,907 chunks) reduced accuracy from 75% to below 40%. To address this dilution, we propose MASDR-RAG ( Multi-Agent Scoped Domain Retrieval for RAG) and evaluate it on 200 expert-validated queries across five LLM backbones, six corpora, and two index stacks. Our results indicate that domain scoping using organizational metadata is the key fix, significantly improving P@10 from 0.77 to 0.86 ($p < 0.05$). Furthermore, our investigation of multi-agent orchestration revealed that a high degree of configuration dependence results --creating what we call the precision-faithfulness paradox. Based on these varied outcomes, our practical recommendation is simple: scope first, then perform a single synthesis call, reserving full multi-agent orchestration for genuinely multi-domain corpora paired with native-tool-call backbones. Code and Data will be made public upon acceptance.

In hierarchical reasoning, failures often originate at intermediate decision points where the agent commits to a wrong branch without recognizing that it lacks critical information. Rather than treating clarification as an external uncertainty trigger, we propose ACTION-RATING, a formulation that places it inside the agent's action space on a shared ordinal scale with navigation, so that asking competes directly with acting at every decision point and help-seeking becomes observable at intermediate states. Two structurally distinct information-seeking modes emerge from the agent's own ratings: mandatory (no viable branch) and opportunistic (residual uncertainty despite a leading candidate). On Harmonized Tariff Schedule classification (30,000-node taxonomy, three benchmarks, 9~LLMs across 4 families), we observe a regime shift from mandatory to opportunistic clarification, with Information-Seeking Effectiveness (ISE), a local diagnostic defined as the fraction of help interactions followed by a correct next navigation step (not a final-task metric), rising from 50% to 74%. Three diagnostic contrasts fail to reproduce this structure. A separability test shows that the information-seeking pattern (mode split, ISE ranking) persists when answer quality is degraded (-18.8% accuracy), supporting an empirical separation between where an agent seeks help and the quality of the help it receives. Under the controlled answer channel, accuracy gains reach +16.2% at 10-digit; we read this as an upper bound on what better localization could unlock, not a deployment estimate.

Clock Tree Synthesis (CTS) is a computationally expensive stage in the physical design flow, requiring iterative EDA tool invocations to navigate a vast configuration space for optimal power, wirelength, and timing skew. Existing machine learning approaches require computationally expensive retraining or fine-tuning cycles to adapt to unseen macro architectures and are architecturally mismatched to the millions of evaluations demanded by exhaustive combinatorial search. We present SwiftCTS, a physics-informed surrogate framework that addresses both limitations simultaneously. By coupling lightweight, physics-grounded statistical features with gradient-boosted ensembles, SwiftCTS trains in under five seconds on a CPU and delivers sub-millisecond inference without GPU support. To handle out-of-distribution (OOD) designs without retraining or fine-tuning, we introduce a K-shot multiplicative calibration mechanism that anchors predictions to just one or two physical reference runs, reducing power prediction error from 24.5\% to 3.3\% and wirelength error from 56.6\% to under 1\% on unseen macros. Integrating this engine with an evolutionary optimizer, SwiftCTS evaluates 100,000 CTS configurations in under ten seconds, yielding Pareto-optimal frontiers that are physically validated within the OpenROAD flow. Closed-loop validation confirms prediction errors below 0.5\% for power and wirelength, and timing skew predictions within five picoseconds on an OOD benchmark, consistently outperforming default tool heuristics across all target metrics. Code publicly available at: \href{https://anonymous.4open.science/r/SwiftCTS-7E6E}{https://github.com/BarsatKhadka/SwiftCTS}

We propose Annealed Entropic Allocation, an annealed weighted soft-min framework for sequential budget allocation in ranking and selection. The central idea is to replace the non-smooth maximin large-deviation rate objective with a weighted log-sum-exp surrogate that aggregates challenger-specific pairwise scores through soft-min weights, mitigating hard switching when several challengers are nearly active. To improve finite-budget discrimination, we incorporate the saddlepoint approximation -- a sub-exponential correction derived from refined pairwise tail asymptotics. Because these corrections are sub-exponential and the smoothing parameter is annealed to zero, the surrogate preserves the same first-order large-deviation target as the classical maximin formulation. We show that the surrogate converges uniformly to the hard minimum, that the soft-min weights concentrate on the active challengers, and that, under fixed weights, the induced target allocation map is continuous on the simplex interior. Numerical experiments on Gaussian and exponential instances demonstrate competitive performance, especially when multiple challengers are nearly tied.

Modular neural network pipelines suffer from error compounding: noise at any module boundary propagates and potentially amplifies through subsequent modules. We introduce energy conservation as a hard physical constraint on inter-module information flow. Activation energy (the squared L2 norm of feature vectors) is enforced to be exactly preserved at every module boundary. Unlike soft energy penalties, conservation is an inviolable law: the network may redistribute energy across neurons but cannot create or destroy it. Four experiments on CIFAR-10 demonstrate: (1) conservation retains 77.4% of clean accuracy at noise sigma=0.2, versus 35.1% for baselines and 30.9% for energy-penalized models (p<0.001, 5 seeds); (2) pipelines become depth-invariant, retaining 93.3% at depths 2 through 5 with noise at every boundary; (3) the advantage generalizes to systematic bias (+45.1%), Gaussian (+40.4%), and adversarial noise (+4.8%), with a principled non-effect on dropout (-0.3%); (4) on ResNet-18, the conservation advantage scales inversely with intrinsic normalization: +0.3 pp with BatchNorm, +26.2 pp without at sigma=0.2, reaching +58.0 pp at sigma=0.5. Experiment 5 validates the operator on a real modular robotic pipeline (MuJoCo physics, Franka Panda). Across three independent runs on separate machines (90 trials per cell), conservation provides +18.9 pp average advantage on monocular-depth-style noise. A formal bound proves conserved noise energy is strictly less than input noise energy.

We study distributed optimization with stochastic gradients and finite-bit communication modeled by random (unbiased) quantization. We propose q-PDGD, a quantized stochastic primal-dual method, and analyze it under relaxed global geometry. Under restricted secant inequality (RSI), a constant step-size yields linear contraction to an explicit neighborhood determined by gradient noise, quantization distortion, and network connectivity, while a diminishing step-size achieves O(1/k) convergence without shared-minimizer assumptions. Under Polyak-Lojasiewicz (PL) inequality, we obtain linear-to-neighborhood convergence in the same stochastic quantized setting. Our results match the best-known centralized stochastic rates in oracle complexity, and are supported by experiments demonstrating the predicted tradeoffs between quantization level, step-size choice, and graph structure.

Scientific AI agents increasingly retrieve evidence, reason across sources, and synthesize conclusions used in consequential decisions. Yet, their ability to do so in high-stakes domains such as health remains unclear. We introduce SciConBench, a large-scale live benchmark of 9.11K questions and expert-written conclusions from systematic reviews to evaluate open-domain scientific conclusion synthesis. The benchmark draws on an expert-validated automated evaluation pipeline that decomposes conclusions into atomic facts and measures correctness and comprehensiveness via factual precision and recall. To mitigate data leakage, we further introduce SciConHarness, a clean-room evaluation harness that equips agents with controlled web interaction to ensure valid measurement. Evaluating 8 frontier models and deep research agents, we find that factual quality remains low: under clean-room settings, the best agent achieves only a factual F1 of 0.337. Our clean-room setting consistently reduces performance relative to unconstrained evaluation, suggesting that leakage inflates estimates of models' true synthesis capabilities. Finally, we audit consumer-facing agents (e.g., Google AI Overview, OpenEvidence) and find they frequently generate incomplete and sometimes contradictory conclusions, even when the ground-truth answer is available. Overall, our results show that reliable synthesis of scientific conclusions remains an open challenge, and that clean-room evaluation is essential for assessing open-domain AI agents.

Sustained driving automation systems are envisioned to be used as the foundation for driverless mobility services. However, both researchers and practitioners acknowledge that current driving automation systems are not yet able to handle all traffic situations that a human driver can handle. To bridge this gap and enable mobility services without an in-vehicle human driver or fallback, remote operation (or teleoperation) is increasingly discussed. Recently, first legal actions have been taken to enable some forms of remote operation on public roads. Remote operation encompasses a broad spectrum of methods to support a driving automation system, ranging from remote assistance, which includes providing information or releasing a maneuver, to remote driving, which includes driving the vehicle from a remote location. As such, safe implementation of remote operation in public road traffic challenges the collaboration of multiple academic disciplines (e.g. engineering, psychology, informatics, law, etc.) and stakeholders (e.g. remote operation service providers, remote operators, vehicle manufacturers, regulatory authorities, etc.). At the same time, the interdisciplinary discourse is often challenging due to differing expectations and language. To build a common ground, this article traces terminology back to the original differences in information processing both on human and vehicle side. This framework aims to help further discourse by directly specifying what is needed to engage a diverse audience including researchers and stakeholders of different backgrounds and interests. Recently discussed forms of teleoperation are integrated into this framework.

We introduce Embodied-R1.5, a unified Embodied Foundation Model (EFM) that integrates comprehensive embodied reasoning capabilities, spanning embodied cognition, task planning, correction, and pointing, within a single architecture toward general physical intelligence. Leveraging three automated data construction pipelines to significantly expand the data coverage of critical capabilities, we build a large-scale data system of over 15B tokens, and design a multi-task balanced RL recipe to alleviate heterogeneous task conflicts. We further introduce a Planner-Grounder-Corrector (PGC) closed-loop framework that enables a single model to autonomously execute and self-correct over long-horizon tasks. With only 8B parameters, Embodied-R1.5 achieves SOTA on 16 out of 24 embodied VLM benchmarks, surpassing leading models like Gemini-Robotics-ER-1.5 and GPT-5.4. Benefiting from the internalized embodied capabilities, Embodied-R1.5 can be fine-tuned into a VLA with only a small amount of data, outperforming leading VLA models like $π_{0.5}$ across 4 popular manipulation benchmark suites. We further conduct extensive zero-shot real-robot experiments, validating performance in instruction following, affordance grounding, articulated object manipulation, and long-horizon complex tasks, demonstrating strong generalization to the physical world. We open-source model weights, datasets, training code, and EmbodiedEvalKit, an evaluation framework tailored for embodied tasks, to facilitate future research in EFMs.

Learning from imperfect data is a central theme in machine learning, connecting practical questions of robustness to fundamental questions of learnability. Here we examine attribute noise: learning from corrupted inputs while keeping the labels intact, a setting that has received considerably less analytical attention than its label-noise counterpart. We consider two types of corruption models: additive noise and replacement noise. Through experiments with multi-layer perceptrons (MLPs) on corrupted classification datasets, we find that neural networks remain robust, maintaining well-above-chance accuracy even when inputs are >90% corrupted -- far beyond human recognition. To understand this robustness, we analyze infinite-width networks in the heavy-corruption regime using a mean-field-inspired approach and derive a leading-order decision rule for the classification outcome: the network implements a prototype rule, the nearest-class-mean, assigning each test point to the class whose training-set average it most closely resembles. This leading-order decision rule is universal across a broad range of MLP architectures, holding for any depth, as well as a wide class of activation functions and noise distributions. The same centroid mechanism closely matches finite-width network behavior in our experiments and provides an interpretable and analytically tractable account of why learning can succeed even when individual training examples carry almost no signal.

Large language models are increasingly deployed across professional domains, bringing hard-to-predict risks, including the generation of harmful or disrespectful content. Although substantial progress has been made in developing safety evaluation datasets, existing resources remain overwhelmingly English- and Chinese-centric. This limitation is particularly pronounced when evaluating languages that operate within shared sociocultural, legal, and ethical contexts. To address this gap, we introduce Schützen: a German--Bulgarian safety dataset designed to assess model answerability under risk, covering both a low-resource language (Bulgarian) and a high-resource language (German). Experiments with multilingual and language-specific LLMs reveal pronounced cross-language differences in safety behavior, highlighting the necessity of tailored, region-specific evaluation resources to support the responsible deployment of LLMs in Germany and Bulgaria. Datasets and code are available at https://github.com/xnlp-lab/Schutzen. Warning: this paper contains examples that may be offensive, harmful, or biased.

We introduce SPADE (SPlit And Delay Embeddings), an autoregressive transformer for sequences whose tokens carry multiple features. Rather than embedding these features jointly, SPADE embeds them independently. Delaying each feature stream relative to the previous one allows intra-token correlations to be learned by the standard self-attention mechanism. Applied to point-cloud calorimeter shower generation in the highly granular ILD detector, SPADE is competitive with the state of the art AllShowers model on photon showers, and substantially outperforms its VQ-VAE-based predecessor OmniJet-$α_C$. The mechanism is applicable to any generative task with multi-feature tokens, enabling LLM-style pretraining workflows for higher-dimensional data.

Recovering cosmological information beyond the power spectrum is a central goal for upcoming cosmological surveys, since late-time non-Gaussian signal in the matter density cannot be accessed through two-point statistics alone. Marked statistics fold part of this information back into the two-point level by reweighting the field with non-linear functions. We propose a neural marking scheme to generalize this process through a set of interpretable, physically motivated transformations that directly allow to interpret the gain in cosmological information at the morphological level. We employ a contrastive learning objective to align learnable marked summaries with the underlying cosmological parameters. At $k_{\max}=0.2\,h\mathrm{Mpc}^{-1}$, our neural mark tightens the marginalized constraint on $σ_8$ by $2.9\times$ and on $Ω_m$ by $1.8\times$ compared to classical marks, breaking the $Ω_m-σ_8$ degeneracy at the Fisher information level. It further reduces the parameter MSE across our cosmological parameter prior by $1.45\times$ over the best classical mark. The learned latent geometry aligns with the $Ω_m$ and $σ_8$ directions in parameter space, indicating that the contrastive objective recovers the dominant axes of cosmological information. Our approach opens the door to more powerful, interpretable summary statistics for cosmological inference.

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.

Supervised fine-tuning (SFT) typically maximizes the likelihood of every token in a demonstrated trajectory. However, an observed token can be non-unique, noisy, or misaligned with the model prior. Strictly fitting toward this one-hot target may be suboptimal, especially when the pretrained model encodes a rich knowledge prior. In this work, we reinterpret SFT as target distribution design: instead of studying only the loss objective, we analyze the token-level target that the loss drives the model to match. We introduce the Q-target framework, which decomposes SFT supervision into two explicit choices: (1) how strongly to rely on the observed token, and (2) how to allocate the remaining probability mass over alternatives. This perspective unifies many existing SFT variants as implicit choices of the target distribution Q. Building on this view, we propose Target-SFT which constructs the training objective directly from the desired target distribution. This method consistently outperforms across the ten reasoning dataset-model settings evaluated, showing the effectiveness of this target-based approach. Overall, our formulation reveals a more fundamental design principle for SFT training and opens a broader search space for SFT objectives.

Large Language Model (LLM)-based multi-agent systems are increasingly powerful, but current agentic workflow optimization paradigms make an unsatisfying trade-off. Task-level methods spend substantial offline compute yet deploy only a single workflow, leaving complementary candidates unused, while query-level methods synthesize a new workflow per query at substantial inference cost. Our motivating analysis shows these paradigms are more complementary than competing: workflows discovered during offline search often solve different subsets of queries, and many queries handled by expensive query-level generation can already be solved by cheaper precomputed workflows. This suggests a different objective: rather than searching for one universally best workflow or regenerating one per instance, we should build a compact bank of reusable, complementary workflows and select among them adaptively at inference time. Doing so requires solving three coupled problems: generating complementary rather than redundant candidates, compressing them into a small deployable portfolio, and assigning each query to the right workflow under a performance-cost trade-off. To this end, we present FlowBank, a three-stage framework for portfolio-based agentic workflow optimization. Diversifying proposes DiverseFlow to steer search toward under-covered queries and produce a high-coverage candidate pool. Curating proposes CuraFlow to compress this pool into a compact portfolio with minimal redundancy. Matching casts deployment as edge-value prediction on a query-workflow bipartite graph and routes each incoming query to the portfolio member with the best predicted utility. Across five benchmarks, FlowBank achieves the highest average score among the evaluated methods while remaining cost-competitive, improving over the strongest automated and handcrafted baselines by 4.26% and 14.92% relative, respectively.

In this paper, we propose EEVEE, the first multi-dataset test-time prompt learning framework for LLM agents, enabling test-time prompt learning under real-world task streams. Existing methods are largely designed for single-dataset settings, while real-world applications require models to handle heterogeneous input streams drawn from multiple datasets, domains, and task distributions, limiting their practical applicability. To mitigate cross-dataset interference, EEVEE introduces a router that partitions incoming inputs into task clusters and assigns them to suitable prompt configurations. This design is optimized via a router-prompt co-evolution strategy, which employs interleaved router and prompt learning phases to address their mutual dependency. Experiments across multiple datasets demonstrate that the framework improves robustness under heterogeneous data streams while maintaining single-benchmark learning capability and efficiency. Specifically, EEVEE improves average multi-benchmark scores by 10.38 and 24.32 points over Qwen3-4B-Instruct and DeepSeek-V3.2, surpassing SOTA methods GEPA and ACE by up to 37.2% and 48.2%.

Data tells stories that shape society; the data journalist's job is to turn raw information into stories non-experts can trust. A high-quality news feature takes a newsroom team weeks: hunting for context, running statistics, choosing an angle, and designing visuals. Recent agents handle individual steps well: data-science agents close the analysis loop, while design agents synthesize beautiful websites. But can an agent serve as a data journalist end to end? We introduce Data Journalist Agent (Data2Story), a multi-agent framework that orchestrates specialized roles into a single virtual newsroom. Data2Story contributes two innovations. (i) Claims are evidence-grounded: an Inspector links every number, angle, and asset back to data, code, or an external reference. (ii) Articles are multimodally generative: rather than defaulting to plain text and static charts, Data2Story reasons about what readers will want to see, then deploys multimodal tools, such as interactive maps for geography and audio for music. We evaluate Data2Story on 18 articles, each paired with the originally published expert piece, along four axes: (a) human-agent angle coverage; (b) rubric evaluation with 53 participants across five dimensions; (c) computer-use agents as judges, a cost-saving proxy for how readers navigate interactive articles; and (d) verifiability, where a coding verifier re-executes statements against the data and checks claims against references. Data2Story produces competitive, evidence-traceable multimedia stories, with particular strength in transparency and auditability. Human articles retain an edge in editorial angle, creative design, and presentation. We position Data2Story as a collaborator for journalists, enabling more evidence-based, transparent, and verifiable reporting. Code and demos are available at https://data2story.github.io.

Conditioning a language model on additional context, such as feedback on a previous attempt, typically improves its response. Self-distillation trains the model to retain this improvement when the context is not present. The method works by matching the model's output distribution under two settings: a student that sees only the question, and a self-teacher that also sees the context. What the model learns therefore depends on what context the self-teacher receives, yet the design of this context remains largely unexplored. We study context design for self-distillation by training a solver on feedback from a frozen critic. We compare three conditions: (i) a binary reward (GRPO), (ii) the reference solution, and (iii) a step-by-step critique aligned to the solver's reasoning trace. Step-aligned critique yields the largest gains, outperforming GRPO by 16.11 points and reference-solution-conditioned self-distillation by 5.27 points (Avg@12). Per-token advantage analysis reveals why: step-aligned feedback targets only the tokens where reasoning fails, leaving correct behavior intact. Conditioning on the reference solution, by contrast, pressures the model to change its behavior at every token (even correct steps) because an alternative derivation inevitably differs in phrasing and approach. This suggests that structural alignment between feedback and the solver's reasoning is a key driver of self-distillation effectiveness.

Deployed large reasoning models (LRMs) often behave unexpectedly. Test-time steering controls LRM outputs by intervening on their hidden representations, but it can degrade output quality. We argue that prior steering work implicitly relies on internal features that detect behavior in already generated text. We show that these detection features are poor predictors of future behavioral outcomes, and thus not the natural intervention target. Instead, we train activation probes to predict future behavior likelihoods from intermediate reasoning steps. These probes predict the most likely behavior with 64%-91% accuracy, revealing a separate type of internal prediction features. Building on these prediction features, we introduce a text-level steering method, Future Probe Controlled Generation. FPCG samples multiple candidate sentences and chooses the best one according to a probe predicting the future behavior likelihood. This enables steering with almost no output quality degradation. FPCG also enables steering in several evaluations where activation steering fails. These results show that distinguishing detection and prediction features enables a more nuanced approach to controlling LRM behaviors.

Gaussian-process upper confidence bound (GP-UCB) and decision-estimation-coefficient (DEC) methods may appear, at first sight, to belong to different theories. This paper places the two viewpoints in a common algorithmic-information language for frequentist RKHS bandits. GP-UCB fixes an algorithmic, rather than true, Gaussian-process prior and exploits realized-trajectory complexity together with computational tractability, whereas MAMS optimizes a robust class-wide MAIR/DEC envelope. Through the unified MAIR framework and heterogeneous positive-semidefinite algorithmic priors, we generalize both the GP-UCB analysis and the MAMS algorithm, propose a safeguarded master that combines their advantages, and provide a kernel-bandit construction showing that algorithmic complexity can be more informative than class-wide minimax or DEC certificates in overparameterized models. The resulting message is that algorithmic information and class-wide minimax coefficients answer different questions and can lead to different gaps; kernel bandits provide a clean setting in which this distinction becomes mathematically visible.

Large-scale model training increasingly relies on composing multiple parallelism strategies, such as data, pipeline, and expert parallelism, together with memory-saving optimizations like ZeRO. Deployed systems for foundation model pretraining often rely on human experts to manually design a high-level parallelism strategy then implement the corresponding low-level execution strategy, making it difficult to adapt the system to new strategies. Meanwhile, many general-purpose frameworks are more flexible but their implementations are still tied to a fixed set of common parallelism strategies, making it challenging to integrate state-of-the-art strategies. We present Piper, a user-controllable distributed training system that decouples the strategy from the runtime implementation. Piper allows users to declare a comprehensive distributed training strategy with a small set of model annotations and scheduling directives. Each directive applies a transformation on Piper's intermediate representation (IR), a unified global training DAG that represents all computation and communication. Using this IR, Piper compiles per-device execution plans and executes them with a distributed runtime agnostic to the strategy. We show that the combined system maintains performance parity on commonly available strategies such as ZeRO, while also enabling additional performance and memory efficiency gains through joint scheduling of compute and communication in composed parallelism strategies such as DeepSeek-V3's DualPipe.

Full-duplex spoken dialogue models can listen and speak simultaneously, making them a promising architecture for natural conversation. However, current models are trained solely with supervised learning through token-level likelihood maximization, which does not directly optimize interaction-level behaviors, causing interactivity issues such as excessive silence and ill-timed turn-taking. Recent work has applied reinforcement learning (RL) to improve interactivity, but existing methods address only a limited set of interactive behaviors in their rewards. In this work, we propose a post-training alignment method that comprehensively improves the interactivity of full-duplex spoken dialogue models through RL. We address the four canonical axes of interactivity: pause handling, turn-taking, backchanneling, and user interruption. For each axis, we extract short audio segments from human conversation corpora and optimize the model with axis-specific reward functions. An extra LLM-based reward for response quality prevents semantic degradation. We apply our method to two open-source models, Moshi and PersonaPlex, demonstrating consistent improvements in interactivity on both offline evaluation with pre-recorded audio and real-time multi-turn dialogue evaluation.

Large Language Models (LLMs) are increasingly described as performing at the level of human experts on knowledge economy tasks. These claims are primarily based on how LLMs perform on benchmarking tasks that measure average performance across standardized datasets. Primary limitations of many benchmarking tasks are that they often measure performance based on content directly included in LLM training data, and they frequently do not assess the reliability of LLM performance or the magnitude of LLM errors. However, in high stakes contexts, these qualities are critically important. Through a novel LLM benchmarking task that requires writing computer code to complete a data analysis task, we compare the performance of a frontier LLM against submissions from human experts and explicitly measure the variance of responses and the magnitude of errors. Our study reveals that the human experts perform better on average on a range of metrics and demonstrate less variability in performance. Our results provide evidence that LLMs do not consistently perform at the level of human experts and demonstrate the importance of measuring variance and assessing error magnitude in LLM benchmark evaluations.

Long chain-of-thought (CoT) trajectories in large language model (LLM) reasoning cause severe inference bottlenecks due to rapid key-value (KV) cache growth. Current decoding-time compression methods mitigate this issue via token eviction, but typically assume a uniform budget distribution across all layers and heads. In contrast, existing non-uniform budget allocation methods are predominantly designed for the static prompt prefill phase, and they do not capture the stepwise context demands of autoregressive reasoning. To bridge this gap, we propose ReasonAlloc, a training-free framework that recasts decoding-time KV compression as a hierarchical budget allocation problem. ReasonAlloc operates at two complementary levels: an offline layer-wise preallocation strategy captures an architecture-driven demand pattern which we call ``\textit{Reasoning Wave}'', while an online head-wise strategy reallocates resources during decoding to information-rich heads based on real-time utility. Evaluations on mathematical reasoning benchmarks (MATH-500, AIME~2024) using DeepSeek-R1-Distill-Llama-8B, DeepSeek-R1-Distill-Qwen-14B, and AceReason-14B show that ReasonAlloc outperforms uniform-budget R-KV, SnapKV, and Pyramid-RKV (a baseline enforcing a static, monotonically decreasing layer budget), with the largest gains at small budgets (128-512 tokens). ReasonAlloc is plug-and-play with existing token-eviction policies and introduces negligible inference-time overhead.

As data center energy demand approaches grid-level constraints, optimizing conventional server infrastructure is essential for sustainable growth. The long-standing assumption that "cooler is better", i.e., lower CPU temperatures reduce power, does not fully hold for modern low-voltage CPUs, where inverse temperature dependence (ITD) drives higher supply voltages at lower temperatures. This creates a non-monotonic performance-per-watt curve where efficiency peaks at an intermediate thermal point. In this paper, for the first time, we empirically characterize ITD on production Intel Xeon CPUs and demonstrate that efficiency-optimal temperatures are CPU part-specific, and frequently higher than typical data center operating conditions. Measurements from commercial cloud data center platforms (Amazon, Equinix) reveal that approximately half of modern high-power CPUs operate about 10°C below their efficiency-optimal thermal point. By implementing ITD-aware thermal grouping of CPUs and inlet temperature adjustments, data center operators can optimize facility-level cooling and overall sustainability. Our case study shows that this approach can reduce total data center energy by 4-13% without sacrificing performance or reliability.

In this work, we present COGENT, a continuous graph emulator with Neural Ordinary Differential Equations for long-term physical forecasting on irregular geospatial meshes. COGENT encodes a finite history of system states and associated forcing fields and external forcings with a graph-based history encoder, producing node-wise context vectors that capture both local spatial interactions and temporal evolution. These context vectors initialize and condition a latent Neural Ordinary Differential Equation whose dynamics are driven by interpolated future forcings and explicit relative rollout time. By modeling the forecast trajectory as a continuous latent dynamical system, COGENT can generate predictions at arbitrary future times rather than being restricted to a fixed temporal discretization. A residual decoder maps the resulting latent trajectories back to future physical states, enabling direct multi-step forecasting without repeatedly feeding predicted states back into the model. This formulation combines graph-based spatial representation, history-conditioned latent dynamics, and continuous-time rollout in a unified framework for mesh-based physical simulation emulation. In order to stabilize training with long-horizon supervision, we also propose effective rollout-horizon sampling and a progressive rollout-horizon scheduling strategy. We evaluate COGENT on transient ice-sheet simulations generated by the Ice-sheet and Sea-level System Model, demonstrating improved long-range stability over autoregressive graph baselines. These results suggest that continuous graph Neural ODEs provide a promising methodology for scalable physical forecasting on irregular geospatial meshes, particularly in applications that require stable long-horizon predictions and the ability to query system states at arbitrary times.

As custom hardware accelerators become increasingly central to machine learning workloads, efficient data transfer is critical for maximizing accelerator performance on linear algebra kernels. AXI4MLIR, an extension of the Multi-Level Intermediate Representation (MLIR) compiler framework for automated generation of host-accelerator driver code, incurs significant runtime overhead due to non-zero-copy CPU-accelerator data movement. During transfers from the host to the accelerator, data is copied from heap-allocated memory buffers into contiguous Direct Memory Access (DMA)-mapped buffers. This work identifies this copy as a redundant staging operation and eliminates it through zero-copy data movement. The optimization extends accel, an MLIR dialect introduced by AXI4MLIR, and implements lowering support that allocates buffers directly within DMA-mapped memory, thereby omitting the staging copy. We evaluate the proposed scheme using a configurable matrix-matrix multiplication accelerator and show that the zero-copy optimization reduces main memory data movement by up to 2x, increasing overall accelerator utilization.

Recent one-step generative models accelerate sampling by learning deterministic flow maps of the underlying dynamics. These methods rely on learning from ordinary differential equations, leaving open how to define an exact distillation procedure for stochastic dynamics. We introduce the Itô map, an any-step stochastic flow map that takes an intermediate state and Brownian path and predicts future states in a single pass. The Itô map formulation yields novel estimators for inference-time control by providing cheap, differentiable access to posterior samples. Empirically, Itô maps produce diverse, conditionally valid endpoint samples from fixed intermediate states and support strong steering performance on synthetic and image-generation benchmarks. These results establish any-step SDE integration as a useful primitive for posterior sampling and stochastic control.

Large language models (LLMs) are rapidly acquiring capabilities relevant to biological research, from literature synthesis to interpretation of experimental data. Increasingly, LLM agents can also perform in silico biology tasks that previously required experienced human biologists. These emerging AI capabilities offer new opportunities for scientific discovery and biomedical advances, but they also shift the landscape of biosecurity risks. To address this, we introduce the Agentic Bio-Capabilities Benchmark (ABC-Bench), a suite of tasks to measure agentic biosecurity-relevant capabilities. ABC-Bench evaluates LLM agents on both benign and dual-use biology tasks: writing code to operate liquid handling robots, designing DNA fragments for in vitro assembly, and evading DNA synthesis screening. These tasks require a combination of biology and software expertise. All tested LLM agents outperformed the median expert human baseliner on all three tasks. Agents performed highly on tasks drawing on published knowledge and well-documented protocols, and more weakly on a task requiring novel bioinformatics reasoning. In three wet-lab validation experiments, we found that OpenAI's o4-mini-high produced scripts that, when run on an OpenTrons liquid handling robot, successfully assembled DNA with expected sequences.

We study the problem of learning a drifting concept in the presence of Massart noise. In this framework, an online learner has access to a history of independent samples whose labels are noisy versions of a target concept that may change from round to round. The goal is to output, in each round, a hypothesis with small prediction error. We study the complexity of this learning problem for the fundamental class of margin-separable linear classifiers (halfspaces). On the positive side, we give a computationally efficient learner achieving error $η+ \tilde O(Δ^{1/3}/γ)$, where $η$ upper bounds the Massart noise rate, $Δ$ is the drift rate, and $γ$ is the margin. Interestingly, in the realizable setting, an adaptation of our techniques yields an efficient learner with an improved error rate over prior work. On the lower-bound side, we provide formal evidence of an information-computation tradeoff, strongly suggesting that our algorithm's performance is essentially optimal. Specifically, while the information-theoretically optimal error scales with $Δ^{1/2}$, we prove that $Δ^{1/3}$-scaling is unavoidable for low-degree polynomial tests, even in the special case of random classification noise.

Resistance to first-line osimertinib in EGFR-mutant non-small-cell lung cancer (NSCLC) is the canonical example of predictable clonal evolution under therapeutic pressure, yet no public benchmark exists for training or evaluating computational models on the corresponding longitudinal patient trajectories. We introduce OncoTraj, a public benchmark of 813 EGFR-mutant NSCLC patients receiving first-line osimertinib, harmonized from three real-world clinical-genomic sources: MSK-CHORD (672 patients), AACR Project GENIE BPC NSCLC (34 patients), and the FLAURA molecular-resistance supplement (107 patients). OncoTraj defines three locked tasks: (A) binary classification of progression by a fixed 12-month landmark, (B) regression of time-to-first-progression in days, and (C) six-class classification of the dominant resistance mechanism. We release the harmonized dataset, patient-level train/validation/test splits with an audited no-leakage guarantee, an open-source evaluation harness, and six reference baselines spanning a majority-class predictor, logistic regression, random forest, XGBoost, an LSTM, and a multi-task transformer. With v1's single-timepoint snapshot features, no task clears chance on clean within-source evaluation: the uniformity of this ceiling across every model class localizes the limit to the input modality (single-snapshot tissue NGS rather than serial ctDNA), not the algorithm. The benchmark does recover a reproducible literature-consistent association: TP53 co-mutation raises the 12-month progression rate from 29% to 59% cohort-wide. OncoTraj establishes a reproducible, leakage-audited baseline and converts the modality limit into concrete design requirements for a serial-ctDNA-enriched v2.

Data assimilation (DA) in subsurface flow entails calibrating model parameters to match observed data, typically at wells, while preserving geological realism. Latent diffusion models (LDMs) provide efficient mappings from high-dimensional geological model space to a low-dimensional latent variable, reducing the dimensionality of the inverse problem while maintaining plausibility in posterior geomodels. However, the high nonlinearity in the LDM mapping may degrade the performance of Kalman-gain-based ensemble updates. We present a systematic comparison of DA algorithms applied to large-scale 3D channelized geomodels with hierarchical geological uncertainty. We compare model-space and latent-space DA using the ensemble smoother with multiple data assimilation (ESMDA), and demonstrate a key trade-off: model-space updates achieve significant uncertainty reduction but produce geologically unrealistic posterior models, while latent-space updates preserve realism but exhibit limited uncertainty reduction. Motivated by this, we explore rigorous Markov chain Monte Carlo (MCMC) and Sequential Monte Carlo (SMC) algorithms in the 3D-LDM latent space. To accommodate their high computational demands, we develop a fast surrogate flow model that approximates well-rate responses. MCMC and SMC are evaluated against ESMDA across three synthetic test cases, with DA performed in the LDM latent space. All models maintain geological realism due to the LDM parameterization. MCMC and SMC are consistent with one another and achieve lower data mismatch and more uncertainty reduction than latent-space ESMDA. Our overall results demonstrate that ensemble Kalman methods may provide overestimated posterior uncertainty with highly nonlinear parameterizations, while rigorous Monte Carlo sampling, enabled by fast surrogate models, can provide a more reliable alternative.

We introduce First-Order Trajectory Matching (FTM), a surrogate-modeling method that learns the first-order local transport of probability mass from trajectories of stochastic systems. By matching the symmetric first-order motion of trajectories, FTM learns the probability current velocity, whose flow preserves time marginals to match ensemble averages, while also capturing current-like trajectory quantities such as fluxes, circulations, and barrier-crossing currents. FTM learns the current velocity directly from trajectories, avoiding drift, diffusion, and score estimation. Our stability analysis separates discretization error from sampling variance and shows that the one-step simulation-free FTM loss is stable when temporal resolution and sample size are properly balanced. Across stochastic dynamical systems and PDE examples, we empirically demonstrate that FTM provides trajectory-aware ensemble predictions at low, deterministic-rollout cost.

While fuzzing effectively catches crashes, its shallow oracles often miss semantic drifts and optimization-related errors in data-intensive scalable computing (DISC) frameworks. Property-based testing (PBT) addresses this limitation by checking general semantic invariants across diverse workloads and inputs, rather than relying on specific expected outputs. However, systematically operationalizing PBT for DISC systems remains difficult because it requires both reusable property definitions and effective instantiation into valid workloads and data. We present DiscPBT, a property-based testing engine for Apache Spark. DiscPBT introduces eight reusable meta-properties for DISC semantic testing, spanning equivalence rewriting, data decomposition, computation decomposition, and operator-local semantic relations. To operationalize these meta-properties, DiscPBT provides reusable generators for synthesizing valid workload skeletons and input data, together with an instantiation framework that realizes each meta-property in schema-compatible contexts through compatible operators, expressions, and UDFs. Our evaluation on PySpark shows that DiscPBT achieves 1.2$\times$ higher branch coverage and 1153$\times$ greater plan diversity than CometFuzz. Across 66 concrete properties, DiscPBT reveals cross-version semantic drift as well as subtle corner-case pitfalls involving NaN and empty inputs, that are not captured by crash-based fuzzing alone. These results demonstrate the value of systematic PBT for uncovering semantic issues in DISC frameworks.

We study the task of agnostically learning general (as opposed to homogeneous) ReLUs under the Gaussian distribution with respect to the squared loss. In the passive learning setting, recent work gave a computationally efficient algorithm that uses $poly(d,1/ε)$ labeled examples and outputs a hypothesis with error $O(opt)+ε$, where $opt$ is the squared loss of the best fit ReLU. Here we focus on the interactive setting, where the learner has some form of query access to the labels of unlabeled examples. Our main result is the first computationally efficient learner that uses $d polylog(1/ε)+\tilde{O}(\min\{1/p, 1/ε\})$ black-box label queries, where $p$ is the bias of the target function, and achieves error $O(opt)+ε$. We complement our algorithmic result by showing that its query complexity bound is qualitatively near-optimal, even ignoring computational constraints. Finally, we establish that query access is essentially necessary to improve on the label complexity of passive learning. Specifically, for pool-based active learning, any active learner requires $\tildeΩ(d/ε)$ labels, unless it draws a super-polynomial number of unlabeled examples.

Synthetic post-training pipelines commonly filter generated samples with reward models or holistic LLM judges, yet two practices remain rarely examined together: whether the filtering signal is grounded in the source evidence that induced each generation, and whether rejected samples can be systematically recovered rather than permanently discarded. We present a controlled study of both questions across gate configurations, recovery strategies, and generator scales, using adversarially injected corpora to provide ground-truth failure labels. We find that exact source provenance improves faithfulness gating for stronger judges, that hallucination and reward gates reject largely disjoint sample populations making both necessary, and that an adaptive recovery pipeline combining failure diagnosis with targeted regeneration achieves higher yield, recovery rate, and injection recall than naive resampling. Downstream fine-tuning quality is driven primarily by generator scale, with filtration and recovery conditions contributing meaningfully but secondarily.

Backpropagation (BP) is widely viewed as biologically implausible, in part because it requires feedback weights to be the transpose of forward weights for error propagation. Interestingly, when training a network with fixed random feedback weights to circumvent this issue, learning aligns the forward weights with the feedback weights, leading the backpropagated error signal to become an approximation of the standard gradient used by BP. This process, called Feedback Alignment (FA), occurs in MLPs and very shallow CNNs but does not scale well to deeper architectures. In this work, we first investigated differences between BP and FA models, trained on CIFAR10, specifically focusing on the effective rank of the signal. We found that the FA error has a considerably lower rank and hence is constrained to a lower-dimensional subspace compared to BP, limiting exploration of the parameter space. Motivated by this observation, we evaluated two mechanisms for increasing the effective dimensionality of FA: Muon, an optimiser that orthogonalises weight updates; and hidden activity normalisation, which promotes activation orthogonality. Across larger architectures and benchmarks, we find that these methods consistently improve over FA baselines, for example, on CIFAR100 with a Resnet-18, accuracy increases by 9 percentage points. Our results identify low-dimensional gradient dynamics as a key obstacle to scaling FA and suggest that inducing higher-dimensional update geometry is a promising route toward scaling alternatives to backpropagation.

We recast pass evaluation in football (soccer) as a Monte Carlo Tree Search (MCTS)-like evaluation problem whose components mostly exist in the literature under different names: a value model (possession value), a world model (multi-agent trajectories with ball interactions), and a policy over counterfactual actions (sampling pass variants with noise). Building on the first public high-fidelity tracking dataset with 3D ball trajectories from the Bundesliga, we introduce Monte Carlo Pass Search (MCPS), which infers kick parameters for each observed pass, samples execution variants and option variants, rolls each candidate forward with a ball-conditioned world model until the next ball interaction, and scores outcomes with a learned value model to obtain a distribution over gained value. This distribution enables distribution-aware attribution with two complementary execution-surplus scores used for analysis and ranking: mean-based and percentile-based scores. To make the world model sample-efficient under limited public data, we adapt a discrete-token, autoregressive trajectory generator from autonomous driving (SMART) and show it yields strong best-of-20 forecasting accuracy compared to baselines, while supporting fully hypothetical rollouts for downstream evaluation. We have released model checkpoints and code.

Reinforcement learning with verifiable rewards (RLVR) is a promising approach for enhancing reasoning and agentic behavior in large language models. However, rollout-intensive policy optimization is often limited by insufficient reward contrast, arising when overly simple or complex prompts generate low-variance feedback and when outcome-only rewards assign the same terminal assessment to every decision in a multi-turn rollout. Past efforts have focused on allocating available rollout resources to promising prompts, yet they only leverage sample informativeness at the prompt level and neglect variation in prefix-level informativeness across turns within the same rollout. This work targets multi-turn agentic RL by modeling each ReAct-style thought-action-observation turn as a semantically distinct node, allowing budget allocation to extend from prompt roots to turn-level prefixes with further continuations, which naturally forms tree-structured rollouts. We introduce Tree Rollout Allocation for Contrastive Exploration (TRACE), a unified rollout allocation framework that enhances reward contrast within a fixed sampling budget. Technically, TRACE allocates rollout budget to both prompt roots and intermediate prefixes that are most likely to yield mixed terminal rewards. A shared generalizable predictor estimates conditional success probability at these anchors from prefix histories to guide this allocation. The resulting adaptive tree structure enriches outcome-only feedback and amplifies the policy-update signal. Empirically, TRACE achieves competitive performance and efficiency gains on typical agentic benchmarks, e.g., improving Qwen3-14B Multi-Hop QA average accuracy by 2.8 points over competitive baselines at equal sampling cost.

We study a dynamic assortment problem on a two-sided service platform with incomplete information and heterogeneous customers in a discrete-time setting. In each period, a customer arrives seeking service, and the platform chooses an assortment of sellers to display. The customer then proposes a transaction to at most one seller in the assortment according to a multinomial logit choice model. After a fixed number of periods, sellers review the proposals they have received and each chooses at most one customer according to another multinomial logit choice model, after which the cycle repeats. A key challenge is that the platform does not know the choice-model parameters of either customers or sellers in advance. To our knowledge, this is the first study of a dynamic assortment problem in which both sides' choice parameters are unknown. We develop a data-driven algorithm that learns these parameters while optimizing the platform's objective over time. We evaluate performance using regret, which measures revenue loss relative to a clairvoyant benchmark that knows all parameters and customer arrivals in advance. We show that the algorithm's worst-case regret grows polylogarithmically over time, and we derive a matching lower bound, establishing its rate optimality.

Designing FPGA-based accelerators for modern artificial intelligence workloads requires exploring a large and complex hardware design space that involves architectural parameters, data flow strategies, and memory hierarchies, making the process very time consuming. While existing methodologies such as SECDA enable rapid hardware-software co-design through SystemC simulation and FPGA execution, identifying efficient accelerator configurations remains a largely manual process requiring extensive domain knowledge. SECDA-DSE is a framework that integrates Large Language Models (LLMs) into the SECDA ecosystem to guide design space exploration (DSE) of FPGA-based accelerators. It combines a structured DSE Explorer for generating candidate architectures with an LLM Stack that performs reasoning-guided exploration using retrieval-augmented generation and chain-of-thought prompting, coupled with a feedback loop for iterative and reinforced refinement. Building on our previous work introducing SECDA-DSE, this paper extends its evaluation by generating three accelerator designs, including element-wise vector multiplication, 2D convolution, and matrix transpose, and performing end-to-end execution on FPGA hardware. The results show that SECDA-DSE can generate SECDA-compliant accelerator designs that are successfully synthesized and executed on FPGA hardware. Furthermore, the generated designs capture kernel-specific trade-offs between compute parallelism and data movement, highlighting the potential of LLM-guided exploration to adapt architectural configurations across diverse workloads while reducing exploration time and the need for extensive human expertise.

As newsrooms integrate generative AI, journalists face a disclosure challenge: how to communicate AI involvement in ways that maintain reader trust. Current practice offers two approaches: brief one-line labels or detailed disclosures specifying human oversight, editorial accountability, and error reporting mechanisms. Neither achieves journalists' goal of building trust through transparency. An existing controlled experiment with 34 news readers show that detailed disclosures trigger a \textit{transparency dilemma}, reducing trust rather than increasing it, and risk introducing dark patterns that readers scroll past with the illusion of transparency. One-line disclosures avoid this effect but can create an information gap, prompting readers to expend cognitive effort searching for signs of AI involvement that the disclosure indicates but does not explain. Yet readers are not rejecting transparency, they proposed disclosure designs centered on user agency: detail-on-demand interactions, proportional AI-ratio visualizations, outlet-level signals, and explicit "no AI" labels. I argue that this disconnect between what practitioners believe is responsible disclosure and what users actually need is a design problem for the HCI community.

Service placement in the cloud-edge continuum requires assigning application components to heterogeneous resources under multiple constraints, including latency, locality, and policy requirements. Existing approaches rely on optimisation models or heuristics that require explicit modelling, while neural methods lack transparency and formal guarantees. This work proposes a neuro-symbolic alternative based on a Prolog skill, a reusable interface for schema-constrained fact generation and querying, for constraint-aware placement. The skill enables a language model to structure placement intent into symbolic facts, rules, and queries, while delegating validation and reasoning to Prolog. This design bridges high-level intent and formal constraint evaluation, enabling inspectable and policy-aware placement decisions in cloud-edge environments.

Industrial Augmented Reality (AR) has progressed from laboratory demonstrations to operational pilots across design, training, assembly, maintenance and quality assurance, yet broad, dependable deployment in manufacturing remains the exception. We synthesise existing evidence into a full-stack deployment framework structured along six distinct but coupled decision axes: (i) value and benefits, (ii) technical and integration constraints, (iii) human factors and safety, (iv) organisational and economic considerations, (v) data, security and privacy, and (vi) governance, ethics and long-term risk. Within each axis we separate (a)benefits, (b)failure modes and (c)design considerations, and cross-link them through a deployment checklist that engineering managers and vendors can apply when scoping projects. The contribution is conceptual and practice-oriented: a synthesis grounded in the literature and public deployment reports. We mark where the evidence base is mature (e.g. assembly task time, training efficacy), emerging (e.g. cognitive workload trade-offs, cobot safety zones), or speculative (e.g. metaverse-scale governance), and identify open questions whose resolution conditions the transition from demos to dependable infrastructure.

Clinicians diagnose brain tumors by synthesizing patient symptoms, medical history, and quantitative imaging data from modalities such as MRI and CT scans into a unified clinical judgement. However, most deep learning models rely on MRI/CT images alone, failing to replicate the clinicians multimodal reasoning. We explore a two-branch multimodal network combining raw MRI scans with 91 extracted radiomic features (intensity, texture, shape, and boundary descriptors) to classify brain tumors into glioma, meningioma, pituitary, and no-tumor. A pre-trained CNN backbone encodes the image stream, whereas a dedicated MLP encodes the radiomic stream. Both streams are fused via concatenation, gated, or bidirectional cross-modal attention strategies. Across nine experimental runs on a balanced 7,200 image dataset, all multimodal configurations outperform unimodal baselines with gated fusion achieving the best accuracy of 96.13%.

A global shortage of trained sonographers limits prenatal ultrasound screening in low- and middle-income countries, where over half of pregnant women receive no skilled sonography. Current deep learning approaches address detection, segmentation, or classification in isolation, each demanding a separate model and expert-specified labels at inference. We present FADA, a unified vision-language model built on Qwen3.5-VL that performs clinical interpretation, classification, detection, and segmentation through a single interpretation-first pipeline without external labels. FADA distills knowledge from four domain-specific foundation models (FetalCLIP, UltraSAM, USF-MAE, UltraFedFM) via offline pre-computed feature caching. Selective distillation, which applies feature alignment only to annotation tasks while interpretation relies on standard fine-tuning, consistently outperforms full distillation across most evaluation axes. The recommended variant, FADA-SKD, achieves 0.8820 mean Dice for segmentation, 0.7671 mAP@0.50 for detection, and 100% structured interpretation compliance. Expert sonographer validation across 237 images confirms clinically acceptable outputs in both autonomous and human-in-the-loop modes, with 73.5% of interpretations scoring perfectly under clinician guidance. The system is trainable on a single consumer GPU and deployable without cloud connectivity. We validate edge deployment by running the compressed 0.8B model on a commodity smartphone (Qualcomm Snapdragon 7 Gen 1, 12 GB RAM) using llama.cpp with GGUF quantization, completing the full 5-phase pipeline in approximately 60 seconds entirely offline. This establishes a practical pathway for integrating AI-assisted fetal assessment with portable ultrasound devices, directly addressing diagnostic access gaps in resource-constrained settings. Code, models, and data are available at https://github.com/mahmoodphd/FADA.

Hallucinations, where language models (LMs) generate factually ungrounded responses, pose serious risks, as users tend to blindly rely on them. This is particularly concerning in high-stakes domains, where consequences of such model behavior can lead to significant harms. Despite notable progress in understanding hallucinations, it remains unclear how reliably these models can recognize the limits of their knowledge. We introduce PhantomBench, the first large-scale benchmark of its kind, comprising more than 60K non-existent terms and entities derived from real concepts across diverse domains. Using our benchmark, we evaluate a total of 21 models of various types and sizes. We show staggering hallucination rates across the board (with average rates as high as 86.7% in some cases), and note that even frontier models surprisingly fail to abstain on non-existent concepts, especially when the input presumes their existence. We then show that PhantomBench can serve as a proxy for studying model behavior on rare concepts for which models are more prone to hallucinate. We also provide a pipeline to construct PhantomBench, enabling scalable generation of non-existent concepts tailored to the specific needs of researchers and practitioners.

We investigate limitations of learning $\tanh$ neural networks from point evaluations under finite-precision computations and $L^p$ accuracy guarantees, building on Berner, Grohs, and Voigtländer (2023). Our approach is based on a novel construction of sharply localized bump functions via iterated $\tanh$ activations. Using this mechanism, we show that, in a finite-precision setting, no adaptive randomized algorithm based on $m$ samples can achieve a convergence rate higher than the Monte Carlo rate $O(m^{-1/p})$ in the $L^p$ norm, unless the sampling budget grows exponentially with the size of the network parameters and architecture. The results reveal fundamental limitations imposed by finite precision on the learnability of classes containing localized bump functions, extending previous results for ReLU networks to the $\tanh$ setting.

Recent deep learning approaches for network intrusion detection increasingly incorporate temporal architectures such as recurrent networks and Transformers, often reporting near-perfect performance on CIC-IDS2017. However, many existing studies neither supply their temporal modules with genuine sequence inputs nor evaluate under realistic, leakage-free conditions, making it unclear whether reported gains arise from true sequence-modeling capability. In this work, we reformulate CIC-IDS2017 as a temporal intrusion-detection task by constructing ordered flow sequences from network conversations and benchmarking nine classical and deep learning architectures under a random split, two leakage-free splits, and a padding-scheme ablation. The central finding is that padding convention, not architecture, determines the Transformer's performance: on genuinely sequential (non-padded) windows the Transformer achieves the highest macro-F1 of any model in the experiment (0.89); under zero-pad+mask evaluation it drops markedly (-0.24 macro-F1), while LSTM, GRU, and 1D-CNN remain stable. Under leakage-free group evaluation the Random Forest is the most robust model (+0.009), while the Transformer's false-alarm rate grows from 0.04% to 2.7%, a 67-fold increase invisible under conventional protocols. These findings demonstrate that evaluation methodology -- specifically padding convention and split protocol -- has a larger effect on reported performance than architectural choice, and that widely used random splits with repeat-last padding can overestimate model robustness by up to 0.24 macro-F1. We advocate leakage-free splits, explicit padding disclosure, and sequence-aware benchmarking as standard practice in future IDS research. Code and implementation details are available at https://github.com/zachmocz/temporal-ids-bench.

Elite humanoid soccer shooting requires whole-body stability, high-impulse whole-body interactions, and accuracy to targets. Motion tracking-driven reinforcement learning (RL) provides stability in whole-body movement coordination, but a fixed reference makes it hard to adapt to varied ball positions and strike timings; in contrast, task reward-driven RL struggles to explore and discover valid kicks from scratch. We therefore introduce RoboNaldo, a three-stage motion-guided curriculum RL framework for high-impulse humanoid interaction. A single human-kick reference is used as a scaffold and progressively shifts optimization towards shooting performance. The curriculum first learns a stable whole-body kicking prior, then adapts the kick to free-kick settings where the ball is stationary at random positions, and finally extends it to moving-ball shooting through a locomotion-command and kick-trigger interface. A high-level heuristic planner controls this interface during training, while alternative high-level controllers can drive the same low-level policy at inference. In simulation, RoboNaldo demonstrates free-kick shot error 48.6% lower and shoot velocity 2.96x than prior work baselines. In real world on a Unitree G1 with onboard perception, RoboNaldo attains 0.73 m and 0.86 m average target shooting error from 3 m away in free-kick and moving-ball cases, accordingly. And the post-contact ball velocity reaches 13.10 m/s, which is 59-71% of reported professional open-play shot speed. Project page: https://opendrivelab.com/RoboNaldo.

High-content imaging assays quantify cellular responses to chemical and genetic perturbations, yet continuous trajectories of individual cells are unobservable because cells are chemically fixed at acquisition. Perturbation modeling therefore reduces to inferring stochastic transport between control and treated populations observed only as separate marginals. While recent generative models achieve strong end-point alignment, boundary consistency does not determine intermediate evolution: multiple stochastic processes may connect identical marginals while traversing regions unsupported by observed single-cell morphologies. We introduce \textbf{FreeBridge}, a Schrödinger Bridge formulation for single-cell transition modeling under endpoint-only supervision. FreeBridge defines atomic states as instance-segmented single-cell representations, establishing a fixed cellular manifold, and learns stochastic transport constrained within this geometry via empirical latent support regularization. Across BBBC021, RxRx1, and JUMP, FreeBridge maintains competitive or improved endpoint fidelity and mechanism-of-action retention under a unified evaluation protocol; on BBBC021, it further reduces intermediate support violations. These findings highlight the importance of geometric grounding for biologically interpretable perturbation dynamics. Project page: https://y-research-sbu.github.io/FreeBridge/.

Expressive continuous control policies, such as diffusion and flow models, form the backbone of recent advances in scaling imitation learning for simulated and real robot control. While they are known to scale stably in the supervised imitation learning setting, incorporating them into reinforcement learning (RL) pipelines for policy improvement has proven more difficult. It often requires specialized training objectives or backpropagating through denoising processes, which cause well-known issues with stability and affect scalability. In this paper we study the question of whether simple policy improvement schemes at test time alone, leaving stable supervised policy training intact, can be a competitive alternative which sidesteps these issues. To this end, we propose QGF (Q-Guided Flow), an RL algorithm that performs policy optimization entirely at test time. QGF works by pre-training both a reference flow policy (via a standard behavioral cloning objective) and a value function critic and, at test time, using the value gradient to guide the reference policy to generate higher-value actions without any additional policy learning. Empirically, QGF outperforms prior test-time RL methods on single-task and goal-conditioned offline RL benchmarks with high-dimensional action spaces, and is competitive with state-of-the-art training-time algorithms while being much cheaper to run. Moreover, it exhibits favorable scaling with model size by avoiding the instability of actor-critic training, offering a practical and effective alternative RL algorithm with expressive policies.

This study investigates cross-lingual distributional skew (the Shibboleth Effect) in frontier large language models (LLMs) subjected to sustained adversarial conditions. We develop a multi-agent geopolitical wargame, the Cerulean Sea Crisis, a synthetic maritime territorial dispute designed to mirror the structural dynamics of Eastern Mediterranean conflicts. Six frontier models (GPT-4o, Llama-4, Mistral-Large, Gemini-3.1-Pro, Qwen3.6-Plus, and DeepSeek-R1) participate in a between-groups experiment (N = 10 games per arm, K = 5 rounds per game) in which the sole manipulation is the language of play (English versus Turkish), producing 586 validated statements. A zero-shot classifier assesses behavioral dispositions along two continuous dimensions: Concession Rate and Coercive Rhetoric. The results are heterogeneous. Llama-4 shows a substantial, Holm-corrected increase in coercive rhetoric under Turkish (delta = +0.800, p = .002), whereas Gemini-3.1-Pro displays an equally large decrease (delta = -0.750, p = .005). DeepSeek-R1 exhibits a similar negative shift (delta = -0.860, p = .006) and provides chain-of-thought evidence consistent with a buffering mechanism. GPT-4o shows no detectable effect (delta = +0.130, p = .614). These findings indicate that cross-lingual behavioral skew is contingent on model architecture and training regime rather than a universal property of Western-origin LLMs. We identify two distinct buffering mechanisms, chain-of-thought institutional anchoring and multilingual RLHF alignment, and discuss their implications for integrating LLMs safely into diplomatic and crisis-management settings.

Communication-efficient pre-training of LLMs is increasingly important as training draws on compute distributed across clusters, data centers, and lower-bandwidth links. Many practical methods reduce communication frequency but still rely on synchronous All-Reduce operations that maintain identical model states and tie progress to global collectives. This can become a bottleneck when bandwidth or worker speed is heterogeneous. We introduce GASLoC, a novel decentralized pre-training algorithm that generalizes the notion of communication acceleration to the recently popular "outer optimizer" to allow a practical gossip-based training framework that is compatible with adaptive optimizers, allows for local optimizer steps, and can utilize sparse randomized peer communication. Empirically, on a number of standard LLM training tasks, we demonstrate that GASLoC outperforms state-of-the-art decentralized algorithms in single step per communication setting for a number of topologies and, unlike existing decentralized methods in the LLM setting, it allows to obtain performance competitive with DiLoCo when utilizing multiple local steps. In the heterogeneous bandwidth setting we demonstrate the advantage of GASLoC showing that it can significantly outperform DiLoCo.

Real-world multi-agent systems, from traffic coordination to resource allocation, are often modeled as general-sum games where individual incentives conflict with collective welfare. In these settings, the central challenge is not merely finding an equilibrium, but selecting socially desirable outcomes among many suboptimal Nash equilibria. Standard deep multi-agent reinforcement learning (MARL) methods struggle with this problem, as value-decomposition approaches are constrained by monotonicity assumptions and policy-gradient methods often converge to stable but socially inefficient equilibria. To address this limitation, we propose $Φ$-Actor-Critic ($Φ$-AC), a framework that leverages swap regret minimization to steer learning toward high-welfare correlated equilibria (CE). To make counterfactual regret estimation tractable in deep MARL, $Φ$-AC employs a centralized attention critic that predicts vector-valued regrets in a single forward pass, avoiding computationally expensive counterfactual simulations. We further introduce a Lagrangian-based equilibrium selection mechanism that optimizes social welfare while enforcing stability through regret constraints. Experiments on matrix games, Multi-Agent Particle Environments (MPE), and the Melting Pot Harvest scenario demonstrate that $Φ$-AC learns efficient and stable coordination strategies across diverse mixed-motive settings while maintaining high collective return and competitive fairness.

Evaluation remains a critical bottleneck for interactive agent development. Existing evaluation methods often rely on static benchmarks, which fail to capture the dynamic, multi-step nature of agentic behavior and struggle to expose meaningful failure modes. While user-simulation-based evaluation offers a promising alternative, existing simulation frameworks suffer from two major limitations. First, they provide limited mechanisms for evaluating the quality and comprehensiveness of simulated interactions, making it difficult to assess whether a simulator sufficiently explores an agent's capabilities and failure modes. Second, most frameworks are restricted to either UI-only actions or API-only actions, limiting their ability to model the full range of realistic user behaviors. To address these limitations, we propose VISTA, a Versatile Interactive user Simulation Toolkit for Agent evaluation. Our toolkit includes a suite of six metrics for measuring the realism, capability coverage, and interaction effectiveness of simulated interactions. In addition, we develop a hybrid user simulator that integrates both UI-based interactions and API-based interactions, enabling more realistic and comprehensive evaluation across diverse interactive environments. We evaluate VISTA in e-commerce shopping and education customer service settings and demonstrate that it produces more realistic and comprehensive evaluations than existing methods.

Various test-time interventions for Computer Use Agents (CUAs), including critic models, have been developed to improve performance through pre-execution action evaluation in complex Graphical User Interface (GUI) environments. However, existing critics suffer from two key limitations: they (1) focus primarily on short-sighted decision loops (e.g., forgetting earlier actions) and (2) lack the visual grounding needed to detect flawed actions (e.g., clicking wrong UI elements). To address these, we introduce HiViG, a History-aware Visually Grounded test-time framework, built around a multimodal critic trained on real GUI trajectories to abstract past interactions into a compact record and to evaluate actions with visual grounding. At test time, HiViG integrates the critic into the policy decision loop to provide macro-action history, which summarizes the policy's completed achievements, and visually grounded critique, which verifies raw execution coordinates against the current screenshot to intercept errors before execution. Across web, mobile, and desktop benchmarks, HiViG consistently outperforms existing scalar and verbal critics, improving average success rates over the strongest baseline by 5.8% for Qwen3-VL-32B and 9.0% for Gemini-3-Flash, and demonstrates strong cross-platform generalization. Ablations show that macro-action history mitigates short-sighted planning and visually grounded critique reduces execution errors, with both components being critical for test-time scaling in long-horizon GUI tasks.

Aligning text-to-image flow matching models with human preferences via direct reward backpropagation is sample-efficient but hampered by two well-known pathologies: activations cannot be stored across the full sampling trajectory at modern model scale, and chained Jacobian products across steps inflate the reward gradient as it travels back to early indices. Connector-based methods, such as LeapAlign, address these issues by replacing the full backward trajectory with a short pinned path, highlighting a useful decoupling between sampling and optimization. However, the quality of the resulting gradient depends on how accurately this short path approximates the full rollout, especially over long intervals. We propose FlowBP, a unified surrogate-trajectory framework that treats the backward trajectory itself as the design object. FlowBP keeps a no-gradient cached rollout for sampling, then builds a lightweight backward surrogate from cached and selectively re-forwarded velocities. This view separates four choices: the reward-model input, active set, integration weights, and bridge coupling, and recovers prior direct-gradient methods as particular settings. Within this framework, we instantiate three variants: FlowBP-Sparse uses sparse Euler reconstruction, FlowBP-Bridge adds controlled bridge coupling, and FlowBP-Lagrange raises the order of leap quadrature. All three bound memory by the active-set size and limit gradient chaining to at most one Jacobian factor. Across SD3.5-M, FLUX.1-dev, and FLUX.2-Klein-base on preference, quality, and compositional metrics, the three variants improve over direct-gradient baselines on most metrics.

With the widespread deployment of Multimodal Large Language Models (MLLMs) in social interaction, understanding and controlling their behavior under complex personality conditions is essential. This paper introduces explicit personality conditioning and establishes a systematic evaluation framework encompassing single-personality induction, multi-personality induction, and personality switching. Experiments show that personality induction improves image captioning performance but can impair performance on tasks requiring precise reasoning, such as visual question answering (VQA). Balancing and residual effects are observed during multi-trait composition and dynamic switching, indicating that model behavior is co-modulated by both previous and current personality constraints. Existing prompt-based personality induction methods show limited transferability to multimodal settings. Our work reveals the dynamic and complex nature of personality modeling in MLLMs and underscores the need for robust, tailored methods for personality induction and evaluation. The code will be released when the paper is accepted.

Recent advances in reasoning and tool-calling capabilities of large language models (LLMs) have enabled increasingly capable agentic systems. However, existing benchmarks remain limited in task complexity, realism, and domain diversity, and often fail to capture interactions that span multiple domains, limiting their ability to evaluate agents in realistic multi-step settings that require sustained reasoning and coordination. To address these limitations, we introduce T1-Bench, a high-fidelity, comprehensive benchmark for evaluating agentic systems in realistic customer-facing, multi-domain environments, featuring interleaved scenarios that require structured reasoning across multi-turn user-assistant interactions and substantially increasing both compositional complexity and evaluative rigor across 25 domains of varying difficulty. We evaluate T1-Bench using 12 proprietary and open-weight models, providing a reproducible and standardized framework for assessing agent behavior, tool utilization, and conversational quality in complex, multi-step environments. We further complement automatic evaluation with human judgments to strengthen the assessment of qualitative performance. Overall, T1-Bench substantially advances prior benchmarks by increasing task complexity, interaction depth, and domain coverage in simulated multi-domain environments. To facilitate future research on agentic systems, we will publicly release data and evaluation code as open source.

Neural population activity models can recover rich temporal structure from binned spikes, but their read-in and readout layers often remain tied to a fixed set of recorded neurons. This coupling limits reuse in long-term brain-computer interfaces, where recorded neuron identities, counts, and response statistics can change across days. We introduce GRAFT, a Transformer-based neural population activity model that separates reusable temporal dynamics from a recalibratable neuron interface. The neuron interface controls how recorded neurons enter and leave the shared backbone, and auxiliary gain and positional mechanisms support neural activity modeling inside the Transformer. On MC Maze under the standard NLB'21 protocol, GRAFT reaches 0.3866 co-bps as an ensemble, setting a new state of the art on the primary co-bps metric among public and reported NLB'21 results. In a cross-day protocol constructed from the NLB'21 MC Maze dataset series, GRAFT recalibrates from MC Maze to the scaled MC Maze datasets (Large/Medium/Small) by updating only 9.21% of parameters, reaching 0.3749, 0.3112, and 0.3152 co-bps with restricted target-day support sets. These results show that the same interface-backbone separation supports both strong Transformer-based neural population activity modeling and data-efficient cross-day recalibration.

Deep Neural Networks increasingly employ low-precision quantization to reduce computational requirements. While FPGAs are well suited for workloads with heterogeneous precisions, their dedicated digital signal processing (DSP) slices only feature fixed-width datapaths that are significantly underutilized by low-bitwidth arithmetic. While previous approaches have already introduced the packing of multiple values onto the same wide DSP datapath, they either only support specific fixed bitwidths or are wasteful regarding the use of additional support logic external to the DSP. This paper proposes an efficient method to dynamically pack multiple (un-)signed inputs with arbitrary bitwidths into a wide multiplier path by leveraging the DSP's internal pre-adder. Building on this, we present two distinct architectures, one optimized for matrix-vector multiplications and the other for convolutions. Our implementations are integrated into AMD's FINN framework. With these optimizations, we reduce the LUT utilization by 21% and increase the FPS/DSP by 36% for the UltraNet model compared to the FINN reference.

AI control protocols oversee untrusted models by monitoring their actions and modifying potentially unsafe steps, often using a trusted model. This partially tampers with the untrusted model's trajectory. If the trusted model detects such an intervention, it may infer properties of the monitor and adapt to evade control. We introduce \textbf{CIAware-Bench}, a benchmark for measuring \textbf{c}ontrol \textbf{i}ntervention (CI) awareness across frontier models. CIAware-Bench tests whether models can distinguish their own trajectories from those modified by a control intervention. The benchmark is comprised of a suite of four task domains (essay writing, BigCodeBench, Bash Arena, and SHADE-Arena), while varying trajectory watermarking, side-task presence, and the control protocol. Evaluating eleven frontier models, we find low to moderate CI awareness under default settings (up to 0.87; random chance balanced binary classification accuracy is 0.5) with substantial variation across task domains and model pairs. Detection is generally easier across model families, suggesting that models exploit provider-specific differences in style or post-training. Overall, CI awareness is not a fixed model-level property, and should be measured for each new model release and deployment scenario. We release CIAware-Bench to track CI awareness and inform control protocols whose interventions are harder to detect.

Despite its importance to applications in protein design, predicting protein properties like binding affinity and thermostability from sparse experimental data remains a significant challenge. Accordingly, we introduce a class of sequence kernels that exploit evolutionary substitution matrices as well as local linearity and demonstrate that the resulting Gaussian processes provide data-efficient models of protein property landscapes, frequently outperforming alternatives that rely on foundation model embeddings. Furthermore--by learning what are in effect structure-aware substitution matrices--we show that our kernels can readily incorporate structural information from foundation models. We demonstrate that these structure-conditioned kernels are well suited to multi-task learning across multiple protein property landscapes and can decisively outperform local supervised learning methods.

Chain-of-thought (CoT) supervised fine-tuning (SFT) is widely adopted to improve reasoning ability, yet we find that it systematically degrades long-context recall in hybrid linear-attention models. Across architectures including HypeNet and Jet-Nemotron, retrieval performance on Needle-In-A-Haystack (NIAH) deteriorates substantially after CoT-SFT, and the degradation becomes more severe under harder retrieval settings and longer context windows. For example, HypeNet-9B on NIAH-S2@256K decreases from $67.2\%$ to $9.4\%$. We attribute this to CoT-SFT biasing attention gradients toward short-range patterns, disrupting query-key projections ($W_Q, W_K$) that are responsible for long-range routing. Motivated by this observation, we propose QK-Restore, a training-free method that restores only $W_Q$ and $W_K$ from the pre-SFT checkpoint while preserving all other post-SFT parameters. We further introduce a Procrustes variant to balance routing preservation and reasoning adaptation. Across architectures, QK-Restore consistently restores long-context capability at zero training cost while preserving reasoning performance; for instance, on HypeNet-5B it improves S3@256K from $65.4\%$ to $76.4\%$ while maintaining strong reasoning performance.

Adopting a single measure can improve the usability, modularity and accountability of software: a commitment to explicit meaning. This entails constructing and agreeing upon a representation of the behavior of the software, as observed in the domain of application. The phenomena comprising this behavior become a vocabulary that grounds all discourse about the software, among all stakeholders, and for all artifacts and activities. These phenomena are individuals; actions they participate in; and facts that result from actions. They can be organized, by partitioning the set of actions, into concepts, offering larger units of meaning. Examples of exploiting meaning are given in three areas: designing for usability (by aligning user and designer on a single shared meaning); generating modular code with LLMs (by mapping units of meaning to units of code, achieving not only modularity but also legibility); and making agents accountable (by having them adhere to a code of conduct that defines their intended behavior).

Effective demand response in smart microgrids requires prosumers to cooperate voluntarily under strategic self-interest, a coordination problem structurally equivalent to a repeated Prisoner's Dilemma on a social network. This paper presents a multi-agent simulation in which a Large Language Model (LLM) Influence Compiler issues structured demand-response directives to a population of heterogeneous prosumer agents, each governed by a hybrid decision architecture combining game-theoretic base probability (derived from payoff history, neighbour imitation, and exploitation memory) with LLM narrative evaluation of incoming coordination signals. The hybrid architecture resolves a key methodological challenge: LLMs aligned via Reinforcement Learning from Human Feedback (RLHF) exhibit strong cooperation bias when used as direct decision-makers, producing flat dynamics regardless of grid conditions. By separating strategic reasoning from grounded narrative evaluation, the model generates realistic prosumer behaviour across six personality archetypes, with baseline cooperation near 50% and clear differentiation under influence. Compiled structured directives achieve 33.3% demand-curtailment cooperation versus 27.0% for unstructured messaging and 28.0% for a no-intervention baseline ($Δ_\mathrm{comp} = +0.063$), with the advantage preserved across both grounded and idealized agent substrates ($Δ= +0.083$) and across all resistance levels ($R = 0.1$ to $0.7$). Hub-targeted dissemination via high-centrality network nodes outperforms peripheral or random targeting, confirming that grid topology provides mechanistic amplification independent of message content. These results suggest that structured LLM compilation, grounded agent reasoning, and network-aware targeting are complementary design principles for scalable, interpretable demand-response coordination in smart-city energy systems.

Instruction-tuned LLMs are increasingly converted into reasoning models through post-training to improve multi-step task performance. This conversion is usually optimized for reasoning accuracy, without explicitly preserving the alignment behavior of the instruction-tuned model, such as safe refusal, bias avoidance, and privacy protection. We ask: does this conversion preserve alignment? We study this question through a trustworthiness audit and find that it is not behavior-preserving by default. For a systematic analysis, we compare reasoning models produced via supervised fine-tuning, RL-based post-training, and distillation against matched instruction-tuned baselines across six trustworthiness dimensions: safety, toxicity, stereotyping and bias, machine ethics, privacy, and out-of-distribution robustness. We observe that reasoning models often improve on reasoning benchmarks but exhibit alignment regressions, including increased toxicity, amplified stereotyping, miscalibrated refusal, and contextual privacy leakage. These regressions are consistent with behavioral drift from the instruction-tuned baseline, measured by KL divergence. Overall, our results point to the broader conclusion that trustworthiness metrics are essential for evaluating reasoning models and should be reported alongside gains in reasoning capability.

Reusing a held-out benchmark adaptively should, in principle, invite overfitting. Yet benchmark-driven machine learning (ML) has produced surprisingly little overfitting in practice. An attractive hypothesis is that successful ML strategies are highly compressible. We study this in the setting of LLM-driven research agents, where the hypothesis becomes directly testable via two complementary information bottlenecks. In \emph{output compression}, an exploration agent adaptively searches for high-performance models using a validation set, and we test whether a fresh ``reproducer agent'' can reproduce its performance given only an extremely short prompt and the training data. In \emph{input compression}, the explorer receives only one-bit feedback indicating whether each submitted model improves on the running best. Across 8 datasets spanning tabular classification, vision, language modeling, diffusion modeling, and reward modeling, we find that these bottlenecks have little effect on performance: short prompts and compressible feedback are sufficient to reproduce and find high-performance models. The hypothesis is falsifiable: when we deliberately induce validation-set overfitting, the results fail to reproduce with short prompts. Taken together, our results support a description-length explanation for the lack of overfitting in benchmark-driven ML: successful strategies occupy a low-complexity region of strategy space.

Conformal predictive systems (CPS) output calibrated bands of CDFs under exchangeability. We extend generalized CPS to non-exchangeable settings by encoding distributional shifts through observation-specific permutation weights. This yields shift-aware predictive systems that remain valid whenever the test point is, conditionally on the unordered sample, a weighted draw from the observed atoms. Since such weights are typically estimated, we introduce weight-uncertainty boxes and construct robust CPS envelopes with finite-sample or asymptotic confidence guarantees. We derive efficient computation for conformity-measure CPS, conformal binning, and conformal isotonic distributional regression. Experiments under covariate shift and feedback-driven biomolecular design show calibrated predictive bands that widen under stronger shifts and tighten as sample size increases.

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.

Distributed Software-Defined Networking (SDN) clusters replicate flow state asynchronously between a master node and its backups, leaving a window during which two backup nodes can each commit a contradictory rule, the master can serialize both into the data plane, and the kernel datapath can latch onto an action that no node believes authoritative. Existing SDN fuzzers miss this fault: they confine their oracle to the control plane, target a single controller, or do not steer concurrency to provoke replication races. We present GapFuzz, a stateful concurrency fuzzer for distributed SDN clusters. GapFuzz injects pairs of contradictory Northbound requests on two non-master nodes with controlled inter-injection delay $Δt$, and reconstructs the global cross-plane state by querying every replica and the kernel-datapath action through ovs-appctl ofproto/trace. A two-phase timing search detects whether a divergence exists, then doubles and bisects on $Δt$ to bound the injection-time window; a lifetime probe labels each verdict transient or persistent and assigns it to one of four cross-plane state classes derived from the ONOS 2.7 source. On a three-node ONOS 2.7 cluster, GapFuzz produces a divergent verdict in 81.7% of attempts ($N=50$, Wilson 95% CI $[77.3, 85.4]$%); every divergence sits between the cluster's authoritative state and the kernel datapath. Phase 2 separates a 5 ms race window for one template from a doubling-cap regime ($Δt_{\max}=10.24$ s) for six others, and 99.4% of divergences persist past 30 s. Replacing the kernel-datapath probe with the OpenFlow user-space probe used by prior fuzzers drops detection by 26.6 percentage points overall and by 46.5 points after excluding canonicalization-forced verdicts.

Spiking Neural Networks (SNNs) coupled with neuromorphic hardware offer energy-efficient solutions for humanoid robot control. However, existing SNN-based motor control systems address bipedal locomotion and arm control in isolation, leaving integrated control of both unaddressed. We present a spiking architecture that coordinates force-based arm control and bipedal locomotion in a simulated humanoid, using the Neural Engineering Framework (NEF) and Semantic Pointer Architecture (SPA). High-level action selection between locomotor and arm control is mediated by a biologically grounded spiking basal ganglia model. We validate the system through co-simulation of Nengo, for the neural control, and Isaac Sim, demonstrating successful target reaching, continuous digit drawing, path-following locomotion, and finally, switching between walking and arm control via basal ganglia disinhibition. To our knowledge, this is the first integrated spiking controller to combine bipedal locomotion and arm control on a full-scale humanoid platform. The full spike-based implementation enables future deployment on low-power neuromorphic hardware.

Recent efforts to extend large language models (LLMs) to speech inputs typically rely on cascaded ASR-LLM pipelines, end-to-end speech-language models, or bridge/distillation-based adaptation. While these routes respectively reuse strong pretrained components, enable native speech-language interaction, or offer lightweight adaptation, they often suffer from transcript-interface latency, costly multimodal training, or sequential speech-language coupling. To address these limitations, we present AuRA, a method that distills audio encoding capability into the LLM. Specifically, AuRA feeds the same speech input to an ASR encoder (as a teacher) and a LoRA-adapted LLM (as a student) through a lightweight audio embedding layer, and uses layer-wise distillation to align the student's hidden states with corresponding teacher representations, thereby internalizing speech representations into lightweight LLM-side adaptations. Compared with cascaded and serial bridge methods, AuRA enables tighter speech-language joint modeling and efficient parallel end-to-end inference, while also reusing pretrained speech and language models rather than requiring large-scale multimodal training. On multiple speech-language benchmarks, AuRA consistently outperforms cascaded systems, speech-to-LLM adaptation baselines, and large-scale speech-language and multimodal models in both effectiveness and efficiency.

Recent work has demonstrated that online reinforcement learning (RL) can substantially improve the quality and alignment of flow matching models for image and video generation. Methods such as Flow-GRPO and CPS cast the denoising process as a Markov Decision Process and apply PPO-style ratio clipping to enforce a trust region. However, we argue that ratio clipping is structurally ill-suited for flow models: the probability ratio between new and old policies is a noisy, single-sample estimate of the true policy divergence, leading to over-constraining in some regions of the trajectory and under-constraining in others. We propose Flow-DPPO (Flow Divergence Proximal Policy Optimization), which replaces ratio clipping with a divergence proximal constraint. A key observation is that the per-step policy in flow models is Gaussian, enabling exact and cheap computation of the KL divergence between old and new policies. Flow-DPPO employs an asymmetric divergence mask that blocks gradient updates only when they simultaneously move away from the trusted region and violate the divergence threshold. Experiments show that Flow-DPPO achieves higher rewards with better KL-proximal efficiency, alleviates catastrophic forgetting, promotes balanced multi-objective optimization, and enables stable multi-epoch training where ratio clipping degrades. Code and models are available at https://github.com/Tencent-Hunyuan/UniRL/tree/main/FlowDPPO.

Sequential recommendation aims to predict users' next interaction with items by analyzing their historical behavior. However, the limited quality of item representations remains a critical bottleneck. While pre-trained large language models (LLMs) can provide rich semantic representations, existing approaches only rely on static encoding of fixed attributes, overlooking the crucial role of target audiences in defining item identity. Moreover, the semantic space struggles to reflect actual user behavior, resulting in a significant gap between semantic representations and behavioral patterns. To address these limitations, we propose GenAIR, a general framework that empowers sequential recommendation with Generative Archetype-grounded Item Representations. Specifically, we first leverage an LLM to analyze item metadata and infer textual description of the Archetype, which represents the conceptual profile of the item's ideal target audience. We then extract the corresponding embeddings in a single forward pass. Further, to ground these generative archetypes in real-world behavior, we introduce a behavioral calibration objective, which explicitly incorporates behavioral signals from actual interactions. This objective adjusts the structure of the embedding space to reflect empirical patterns. GenAIR enables seamless integration with most existing models while maintaining high efficiency. Comprehensive experiments conducted on three real-world datasets demonstrate that GenAIR significantly improves the performance of various sequential recommendation models and consistently outperforms state-of-the-art baseline approaches. Implementation codes are available at https://github.com/AI-Santiago/GenAIR.

Learning-based motion planners, despite recent progress, often suffer from temporal inconsistency. Small perturbations across frames can accumulate into unstable trajectories, degrading comfort and safety in closed-loop driving. Several methods attempt to inject history as a static conditioning signal to stabilize outputs, only to induce the planner to copy historical patterns instead of adapting to environment contexts. To address this limitation, we propose Diffusion Forcing Planner (DFP), a diffusion-based planning framework driven by history-guided control. Specifically, DFP decomposes the full trajectory into history, current and future segments, and assign independent noise levels to each segment. The model jointly denoises the historical and the future segments, enforcing a heterogeneous joint diffusion process. At inference, classifier-free guidance (CFG) is applied to steer future sampling using annealed history in a controllable manner. Closed-loop evaluation and comprehensive ablations on nuPlan show that DFP achieves competitive performance while producing continuous, stable, and controllable motion plans in complex driving scenarios.

Measuring subjective constructs in naturally occurring social media text requires annotation procedures that are theoretically grounded, empirically validated, and transferable to an encoder model for scalable prediction. Using non-English social media posts annotated according to Schwartz's theory of basic human values, we investigate how different LLMs, prompts, and instruction languages operationalize the expression of values in text. We argue that although texts may permit multiple plausible interpretations, theory-based value definitions can constrain interpretations and reduce spurious value attributions. Beyond precision, recall, and F1, we evaluate structural alignment between values, error structure, confidence-ambiguity relations, and annotation stability. We show that different LLMs produce different value interpretations. Iterative prompt calibration through error analysis reduces misattributions and improves alignment with expert annotations. We also derive targeted expert verification rules from recurrent error structures and use them during corpus annotation. Finally, we show that LLM annotations can be transferred to an encoder model through soft-label training, retaining theory-based value interpretations and information about uncertainty in value expression.

Airport surface operations increasingly constrain performance at high-throughput hubs. This study examines arrival taxi-in decisions at Hartsfield-Jackson Atlanta International Airport (KATL) and proposes a two-stage, data-driven decision aid that mirrors controller workflow. Stage I predicts the runway exit selected by an arriving aircraft. Stage II predicts whether, given that exit, the aircraft will cross the active departure runway at a designated point or use the end-around taxiway. Models are trained using ASDE-X surface trajectories, aircraft characteristics, ramp destinations, short-horizon traffic rates, and weather across multiple look-back windows. We benchmark nine classifiers, including Random Forest, XGBoost, LightGBM, and CatBoost, and evaluate accuracy, macro-F1, precision-recall behavior, confusion matrices, Brier score, and Expected Calibration Error. Across east and west flows, XGBoost and LightGBM outperform Random Forest. Stage I achieves 0.86-0.89 accuracy with macro-F1 scores of 0.40-0.50, while Stage II achieves 0.70-0.74 accuracy with macro-F1 scores of 0.28-0.55. Feature-importance analysis shows that approach speed is the main driver of exit choice. Departure rate, crossing rate, ramp destination, and, for west flow, the selected exit are the strongest predictors of crossing versus end-around routing. Minority classes remain harder to predict because of feature-space overlap, as shown by t-SNE and UMAP analyses. The proposed framework supports controller situational awareness through calibrated, explainable predictions while preserving human responsibility for final routing decisions.

We ask whether large language models (LLMs) merely imitate rationales when choosing between two options, or whether their choices reflect a systematic underlying decision structure. Using synthetic binary decision settings in which models choose between profiles defined by graded attributes, we compare the attribute a model says mattered most with the attribute that best explains its choice under a behavioural model fit to prior decisions. The behavioural model predicts held-out choices well, showing that model behaviour is systematically related to the visible attributes rather than being random. However, direct self-reports and a separate score-based judge recover the behaviourally inferred driver only partially. The resulting picture is neither one of arbitrary behaviour nor one of fully articulated belief - outputs are structured enough to support prediction, but explicit reasons track the recovered driver only imperfectly. This qualitative pattern persists across prompt-order and sampling perturbations, alternative behavioural models, targeted occlusion analyses, and structurally varied decision settings. We interpret this as evidence for ``superficial belief'' in LLM decision-making: models behave as if guided by probabilistic local priorities over attributes, while having only limited verbal access to the attributes that drive their decisions.

Tuning controllers for strongly coupled multi-input multi-output (MIMO) industrial processes is hard: decentralized classical auto-tuning ignores loop interaction, and local numerical optimization from natural initializations stalls in the resulting non-convex cost landscape. We ask whether on-premise open-source large language models (LLMs), which keep data on-site and need no plant model, can help. On a single-loop CSTR, classical relay-feedback tuning (IAE 0.106, near the 0.102 optimum) beats an LLM tuner (0.162): for simple loops the LLM adds nothing. The picture inverts on a strongly coupled quadruple-tank with conflicting set-points, scored by a penalized cost J = IAE + lambda*TV(u) that rewards tracking without chattering actuators. There, naive relay tuning (J ~ 28.6) and naive LLM tuning (29.7) are no better than open loop (22.7), and a local optimizer from balanced starts fails in 10/10 runs. A scaffolded open LLM instead reasons about the coupling, proposes the counter-intuitive asymmetric structure, and reaches J ~ 16.9 +/- 0.2 from any start; refining it with a classical optimizer attains the smooth global optimum (J ~ 12.0, 10/10 vs. 0/10), which even applies a non-obvious negative integral correction decentralized tuning cannot. A global optimizer (differential evolution) also reaches this optimum, so the LLM is not the only route; its advantage is sample efficiency and interpretability: a usable controller in 18 evaluations (where the global optimizer is worse than open loop) plus a stated rationale. This edge grows with dimension, reaching ~6x fewer evaluations on a 3x3 plant. The behaviour generalizes across four open models, and on a benign plant the LLM offers no advantage, sharpening the boundary. We contribute a reproducible benchmark delimiting when open LLMs help in control tuning: not as optimizers, but as a sample-efficient, interpretable structural prior.

Large language models are increasingly used to adapt math word problems for personalized learning at scale, but it remains an open question whether those adaptations are consistent across models, preserve cultural diversity at scale, and reveal which cultural entities models treat as most salient. We analyze how Claude Opus 4, GPT-4.1, and Gemini 2.5 Pro adapt 60 English math word problems into Bengali, Hindi, Punjabi (India), Urdu, Sindhi (Pakistan), Italian, and Sicilian (Italy), a language set spanning the full resource spectrum, from high-resource Italian and Hindi to under-studied Sindhi, Sicilian, and Punjabi. We annotate 6,489 entity transformations, coding whether models preserve, localize, generalize, omit, or change entities such as names, foods, and places. Models agree on transformation type in 62.5% of cases and on specific substitutions in only 33.5%, meaning model choice directly shapes which cultural world students encounter. All 21 language-model combinations show entropy collapse, with adaptation compressing rather than expanding cultural diversity. Models prioritize surface markers such as names, foods, and currencies while preserving deeper structural features such as grade-level systems that embed culturally specific assumptions. Despite prompts specifying target countries, models misattribute regional context by using Bangladeshi taka for Indian Bengali students and produce cross-cultural contamination, such as adapting egg hunts as Eid activities. Some failures are visible in individual translations. Others, including diversity collapse, systematic preference for surface markers, and consistent regional misattribution, emerge only through corpus-level analysis. The surface plausibility that makes adapted problems look correct is precisely what makes deeper failures easy to overlook.

OpenClaw has rapidly emerged as a transformative artificial intelligence (AI) agent framework, and its ability to autonomously execute complex, multi-step tasks has attracted an ever-growing and diverse user base. However, this capability comes with significant risks. While existing research has made important strides in characterizing these threats, such work is predominantly directed at technically sophisticated audiences. It remains largely inaccessible to non-technical users. This demographic now makes up an increasingly large and underserved portion of the community, yet it is these very users who most urgently need practical and straightforward guidance. In response, we bridge this gap through a series of interconnected efforts designed to lower the risk barrier for non-technical OpenClaw users. First, we identify and categorize seven core risks that OpenClaw users may encounter in daily usage, explaining each in plain language so that non-technical users can readily grasp the nature and potential consequences of these threats. Second, for each identified risk, we distill a set of corresponding defensive strategies into clear and actionable operational steps that are easy to follow. Third, to make protection even easier, we provide a companion OpenClaw Skill that automates key security configurations, enabling users to safeguard their systems with minimal manual intervention. Through this work, we demonstrate that safeguarding against the risks of intelligent agents need not be the exclusive domain of security experts, and that non-technical users can meaningfully participate in reducing these risks through simple, practical actions.

Large language model unlearning aims to suppress designated undesirable knowledge while preserving benign capabilities. Many unlearning objectives focus on suppressing undesired answers, while recent target-guided variants specify replacement behavior but still leave update locality largely unconstrained. This paper introduces \emph{Null-Space Constrained Response-Specified Unlearning} (NSRU), a projection-constrained low-rank framework for controlled LLM unlearning. NSRU uses an explicitly structured safe target response to specify the desired behavior for each forget query, while suppressing the original undesired content. To localize adaptation, NSRU estimates per-module retain subspaces from benign hidden representations and uses an orthogonal-projected low-rank parameterization to confine LoRA updates to the null space of the retain subspace. The resulting objective jointly optimizes safe-target learning, undesired-response suppression, and retention preservation under this constrained parameterization. We provide a local first-order analysis showing that the projected update reduces retain-side perturbations while preserving editable directions for shaping forget-query behavior. Experiments on TOFU show that NSRU effectively suppresses extractable forget-set knowledge while improving retain QA performance, model utility, and safe-target alignment over representative baselines. On WMDP, NSRU keeps hazardous-domain accuracy near the random-choice region while preserving broad and domain-adjacent MMLU utility. Ablation studies support the complementary roles of safe-target supervision, undesired-response suppression, retention loss, and null-space projected updates, while sensitivity and robustness analyses indicate stable behavior across the tested hyperparameter and prompt variations.

Combining asynchronous Byzantine Fault Tolerant (BFT) consensus with Proof-of-Stake (PoS) creates a trilemma between Sybil resistance, reward distribution fairness, and protection against persistent plutocracy. Existing DAG-BFT approaches (Narwhal+Tusk, Bullshark, and Mysticeti) prioritize liveness over the fairness implications of stake-based selection, resulting in persistent longitudinal centralization. FairWave is a dual-channel DAG BFT protocol that separates anchor selection from reward distribution. The selection channel is super-linear in stake, guaranteeing Sybil gain < 1 for all split factors K > 1. The reward channel is sub-linear, using square-root stake normalization to mitigate rich-get-richer dynamics. The finalized DAG structure provides deterministic uptime and latency factors, allowing honest validators to agree on operational quality without any external oracle. To avoid circular dependency between selection outcomes and selection weights, reputation is used in a lagged form: the active value at epoch e equals the prior epoch's final value. We derive closed-form constraints for both channels and validate them through nine empirical analyses (approximately 550,000 Monte Carlo rounds) against eight baselines. FairWave achieves a Gini coefficient of 0.149 (vs. Pure-PoS's 0.488), a monotone HHI reduction from 0.039 to 0.021 over 50,000 epochs, an optimal-adversary Sybil split of K* = 1, and a success-rate coefficient of variation of 5.2% under +/-25% input perturbation. Safety (agreement and validity) is a formal consequence of the 2f+1 strong-support commit rule, holding unconditionally for f < n/3; the empirical differential is the monotone-continuous liveness-degradation curve, which decreases from 99.6% commit rate at b=0.20 to 71.1% at the theoretical bound b=1/3 without the discontinuous cliff characteristic of view-change-driven leader-BFT.

Many Markov decision processes (MDPs) in operations research have feasible actions that are state dependent and defined implicitly by various operational constraints. These features make it difficult to use standard deep reinforcement learning (DRL) algorithms, whose action interfaces typically assume either a fixed finite action catalog or a simple Euclidean space. Motivated by a Taylor expansion of the optimal action-value function, we propose Bellman--Taylor score decoding, a framework that moves policy learning to a Euclidean score space while enforcing feasibility through an action decoder. The induced latent-score MDP then can be optimized by standard DRL algorithms without differentiating through the decoder. We provide a performance guarantee showing that the optimality gap of this approach decomposes into a structural approximation error and an algorithmic learning error. Lastly, we apply this framework to a queueing network control problem, where the policy essentially learns a state-dependent index-based dispatching rule. Numerical experiments show near-optimal performance in small instances and considerable improvements over benchmarks in larger systems.

We study the problem of generating synthetic data under differential privacy. We establish fixed-parameter tractability (FPT) for this problem where the parameter is the treewidth of the query family's incidence graph. Our algorithms attain optimal error rates across all regimes and are realized by two different approaches: the first is based on linear programming (LP) and the FPT of the separation problem for the LP dual; the second is based on a subsampled private multiplicative weights method, where we obtain FPT for sampling from Gibbs distributions. Both approaches are unified by a dynamic programming framework over a tree decomposition.

This study aims to explore the performance of the VAR model in comparison with mel-frequency cepstral coefficient (MFCC) matrices and log-mel spectrograms using deep learning. In pulmonary sound classification, spectrogram-based representations suffer from inconsistent temporal dimensions due to varying respiratory cycle durations. Along with traditional trimming/zero-padding, adaptive-length windowing was presented to fix their temporal dimensions. Their spectral and temporal dimensions were optimized by testing a range of parameters. Different convolutional neural network (CNN) architectures were employed to extract features from the two-dimensional representations obtained over the sub-phases. The extracted sub-phase features were then fused using various strategies including direct concatenation, gated recurrent unit (GRU) network and GRU with attention mechanism. Model performances were assessed through respiratory cycle-based evaluation and subject-based evaluation comprising multiple respiratory cycles. Several data augmentation techniques were also studied to cope with limitations in data size. The best cycle-based F1-score (0.877) was obtained using the MFCC matrices with thirteen coefficients and 64-point time resolution per sub-phase representation followed by direct feature concatenation, and the best subject-based F1-score (0.855) was obtained using the MFCC matrices with thirteen coefficients and 256-point time resolution per full-cycle representation, both obtained by adaptive-length windowing. Augmentation degraded the performance of models overall, yet mixup augmentation was the best among the methods tested. MFCC outperformed log-mel spectrogram and VAR model in differentiation of asthma and COPD. Sophisticated fusion strategies did not improve the diagnosis. Augmentation did not contribute, demonstrating the significance of authentic data in pulmonary sound studies.

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.

Physics-informed neural particle flow (PINPF) learns a deterministic transport field that moves particles from a prior distribution toward a Bayesian posterior while enforcing the governing probability-evolution equation. However, the standard PINPF velocity model processes particles independently and therefore does not explicitly condition its transport decisions on the empirical particle population. This paper introduces population-aware PINPF (PA-PINPF), which augments each particle update with a permutation-invariant Deep Sets representation of the full particle set. We investigate two population encoders. PA-PINPF-State summarizes the particle states, whereas PA-PINPF-Feature summarizes the complete local physics-informed feature vectors, including particle position, pseudo-time, measurement information, likelihood values, and score information. The latter allows the population context to represent not only particle-cloud geometry, but also the population-level Bayesian transport geometry. The methods retain the original unsupervised physics-informed residual objective and require no ground-truth posterior samples during training. Experiments on range-measurement tasks and nonlinear time-difference-of-arrival posterior transport demonstrate that both population-aware variants improve over particle-wise PINPF, while feature-population encoding provides the strongest performance. These results show that population-level physics features provide useful global information for learned Bayesian particle transport.

The deployment of Large Language Model (LLM) agents for computer automation is accelerating, yet their ability to navigate complex, professional-grade productivity software is largely untested. We argue that Office automation is an ideal environment for benchmarking document-automation capability, as it requires long-horizon planning and reasoning, precise parameter configuration, and multi-application integration. To quantify this capability, we introduce an evaluation based on China's National Computer Rank Examination (NCRE), featuring 200 comprehensive practical-operation tasks across Word, Excel, and PowerPoint. Each task is scored on a 100-point rubric scale using 7,118 machine-gradable criteria, and Score Rate (SR) denotes the mean percentage of rubric points earned across these tasks. We benchmark 7 frontier LLMs and observe stark limitations: single-turn models score a maximum of 36.6%. A stronger agentic system with execution feedback, iterative repair, and broader Office automation access reaches 68.8%, but remains below the 95.5% community-reference score used as a scoring sanity check. Ultimately, our experiments demonstrate that despite recent advancements in code generation, achieving reliable fine-grained Office document automation remains a significant challenge for current code-generating LLM and agent systems.

Furnished floor plans are fundamental to real estate visualization, interior design, and architectural workflows. However, progress in automatic furniture arrangement has been limited by the lack of real, professionally designed floor-plan datasets with object-level furniture annotations. To address this gap, we introduce AntPlan-270, a curated dataset of 270 architectural floor plans with per-room furniture bounding box annotations across ten residential room categories. Building on this dataset, we present Architect-Ant, an editable automatic furnishing framework powered by a fine-tuned vision-language model. Furniture layouts are represented using a compact, coordinate-based domain-specific language (DSL) that encodes object categories and placements relative to the room geometry. To improve spatial reasoning, we generate procedural reasoning traces that capture architectural constraints such as wall alignment, door and window clearance, circulation, fixture compatibility, and room-specific furniture inventories, and use them to supervise fine-tuning of the model. We then apply preference optimization over candidate object placements to further refine layout quality. The generated DSL can be rasterized into semantic masks and used to condition a Flux-based LoRA renderer, producing realistic blueprint-style furnished floor-plan images while preserving the editable symbolic layout. Experiments on layout furnishing show that Architect-Ant produces geometrically valid and functionally plausible layouts, and suggest a scalable path for furnishing larger structure-only floor-plan datasets.

Autoregressive (AR) language modeling is the dominant paradigm for text generation, yet its sequential token-by-token decoding makes inference memory-bound and inefficient. Existing acceleration approaches, such as speculative decoding and diffusion language models, can yield speedups under certain conditions but do not directly address high-load batch serving--the scenario most critical for industrial-scale deployment. We introduce K-Forcing, a push-forward language modeling paradigm for joint next-k-token decoding. K-Forcing distills an existing AR model into a conditional push-forward mapping--one that transforms independent uniform noise variables into a joint sample of multiple future tokens in a single forward pass. This design preserves fixed-length outputs, reuses the AR teacher backbone, and remains compatible with standard AR serving infrastructure. We train this mapping via progressive self-forcing distillation, which gradually expands the prediction window while enabling the student to closely match the sequence distribution of the AR teacher. We evaluate K-Forcing on LM1B and OpenWebText using a standard causal Transformer backbone. When aggressively configured to generate k = 4 tokens per forward pass, K-Forcing delivers approximately 2.4-3.5x speedup across different batch sizes, while incurring modest quality degradation relative to its AR teacher. As inference increasingly dominates the lifetime compute cost of modern LLMs, K-Forcing offers a promising route toward accelerating AR generation under real-world high-load deployment.

Users rely on execution traces to observe agent behavior, diagnose failures, and ensure accountability. These traces contain rich procedural detail, including tool invocations, intermediate decisions, and error-recovery logic. Yet this detail can expose private procedural skills, allowing downstream methods to recover key formulas, thresholds, and strategies without access to model weights or skill files. To quantify this risk and evaluate protection, we construct \textsc{CapTraceBench}, a benchmark of 75 specialized long-horizon tasks and 154 curated skills across seven domains. We also introduce \textsc{RedAct} https://github.com/XuShuwenn/RedAct, a protected trace release framework that localizes protected key information, rewrites traces while preserving verifier-critical evidence, and embeds behavioral watermarks for downstream provenance analysis. Across representative trace reuse methods, \textsc{RedAct} reduces normalized skill transfer (NST) from 44.7--67.1\% on raw traces to below the no-skill baseline, while preserving audit evidence. Its standalone behavioral watermarks reach 93.6--100.0\% true detection with a false alarm rate of at most 1.9\%. These results frame public agent traces as security interfaces and show that selective redaction can reduce procedural capability leakage without removing audit evidence.

As agentic applications increasingly route user tasks through official and third-party LLM APIs, provenance becomes an operational question: which model generated a given black-box response? We study Dynamic Black-Box LLM Provenance: identifying the source LLM from generations elicited by query-varying, non-predefined prompts rather than a fixed input set or benchmark suite. This setting is difficult because prompt semantics dominate the text, while model-specific authorship traces are weak and inconsistent at the surface level. We introduce READER (Robust Evidence-based Authorship Decoding via Extracted Representations), a lightweight provenance framework that treats a frozen proxy LLM as a reader of hidden authorship evidence. READER maps black-box outputs into proxy activation space, temporally filters token states within each response, and performs Bayesian Evidence Accumulation by summing single-response log-posterior evidence across independently sampled prompts. This avoids fragile mean-pooling of prompt-specific representations while preserving the query-wise evidence needed for calibrated confidence. On Agent500, a 50-target dataset built from agent-style prompts, READER reaches $31.0$-$42.4\%$ top-1 accuracy from a single response and $70.0$-$84.0\%$ from 50 responses, substantially outperforming sentence-encoder fingerprints. Scaling across nine proxy readers further shows that stronger LLMs expose more linearly decodable authorship structure, suggesting that authorship perception is already present in frozen LLM representations and can be converted into reliable multi-query attribution.

Automatic speech recognition systems have been shown to under-perform when it comes to transcribing words rarely seen in the training data, namely specialized terminology. Open-vocabulary keyword spotting, combined with contextual biasing, has been shown to mitigate this issue. However, existing systems can only handle glossaries of a few hundred terms without becoming an infeasible bottleneck. We propose a system that stores features with a memory footprint up to 128 times smaller than a comparable baseline and allows users to process massive databases while remaining open-vocabulary. Without fine-tuning the speech recognition model, our system achieves a comparable entity recall as uncompressed solutions, even in languages not seen during training.

Background. Atrial fibrillation (AF) is the most prevalent cardiac arrhythmia and a major determinant of prognosis. Established AF risk scores rely on factors (older age, hypertension) nearly ubiquitous among patients with cardiovascular disease (CVD), offering limited stratification in this high-risk group. Most target long-term (5-10 year) rather than medium-term prediction. We developed interpretable ML models predicting AF risk over a 24-month and entire follow-up horizon in CVD patients using routinely collected hospital data. Methods. Single-center retrospective study of electronic health records from the National Research Cardiology Center (Russia) for patients aged >=18 with CVD but without pre-existing AF, hospitalized more than once between January 2012 and May 2019. A custom NLP pipeline transformed unstructured discharge reports into 73 structured features, combining a rule-based parser with transformer-based NER. Using LightAutoML we built a full model (73 features), a simple model (reduced subset), and a linear model for a bedside risk score. Performance was assessed by ROC AUC, compared with CHARGE-AF, C2HEST, MHS, and HAVOC, and interpreted via SHAP. Results. Of 80,576 records from 45,000 patients, 17,562 met inclusion criteria; 1,438 (8.19%) developed AF. The full model reached ROC AUC 0.735 (24-month) and 0.696 (entire follow-up); the simple model was nearly identical (0.725, 0.696). All non-linear models outperformed the four clinical risk scores (ROC AUC 0.53-0.64). The simple model uses 13 features and is named Pre-AF 13. SHAP identified age and left atrial volume as dominant predictors. A linear risk score (Pre-AF 9) stratified observed 24-month AF incidence from ~7% to 36%. Conclusion. Interpretable ML models built from routinely collected EHR data identify high-AF-risk CVD patients, outperforming established clinical risk scores.

Reliable extrapolation remains a central challenge for generative models in computational physics, because models trained over finite ranges of time, parameters, or geometries may produce physically inconsistent predictions outside the training distribution. We introduce a least-action-principle-guided diffusion, LAPG, a framework that promotes physical consistency during inference rather than relying solely on constraints imposed during training. The method combines a conditional score-based diffusion model with an action-derived physical guidance score. In the first stage, the learned score model generates an in-distribution proposal; in the second, an action-based variational prior refines this proposal toward the target out-of-distribution condition. This formulation turns the principle of least action into a differentiable inference-time correction mechanism and provides an alternative to pointwise residual penalties that often require empirical loss balancing. We evaluate LAPG on representative ordinary- and partial-differential-equation systems, including free fall, conservative and dissipative spring-mass dynamics, interacting point vortices, and potential flow over parameterized airfoils. In temporal, parameter, and geometric extrapolation tests, LAPG reduces phase drift, preserves dissipative decay, captures vortex motion, and improves the lift response of airfoil flows compared with training-time physics-informed baselines.

Rotary Position Embeddings (RoPE) make attention scores position-relative but leave the value pathway position-blind: the message sent by a value token is the same regardless of its distance from the query. We propose RoVE, a parameter-free modification that makes values position-sensitive by rotating them simultaneously with keys, and show that it turns RoPE attention into attentive convolution. This new perspective unifies several independent formulations of the same operation across computer vision, robotics, and modern LLM architectures. Trained 124M and 354M GPT-2 models show consistent empirical gains over RoPE on few-shot in-context learning, out-of-distribution perplexity, and long-context retrieval, with the clearest improvements on tasks that require long-range aggregation.

Rendezvous is a critical task for multi-agent systems, requiring agents to coordinate to meet at an unspecified location. However, achieving this in fluid environments presents a challenge, as it remains unclear how agents can exploit underlying fluid kinematics to facilitate convergence. In this study, we adopt a multi-agent reinforcement learning (MARL) approach to develop physics-informed rendezvous strategies in vortical flows. Compared to a naive strategy, where agents navigate toward their counterparts, MARL strategies significantly improve the rendezvous rate. MARL strategies also show transferability across varying vortex intensities, vortex scales, and swarm sizes. By breaking the symmetry of the state-action map, MARL strategy leverages a non-intuitive mechanism that prevents agents from becoming trapped in separate vortices, thereby enhancing rendezvous success. Additionally, a heuristic strategy is extracted from the learned strategy and also outperforms the naive strategy. Furthermore, a theoretical analysis demonstrates that fluid deformation impedes the rendezvous process. Large finite-time Lyapunov exponents identify where fluid effects separate adjacent agents, suggesting that targets should be planned in weak-deformation regions. Our findings reveal the important role that agent-fluid interactions play in multi-agent tasks and highlight the MARL capability to explore swarm intelligence in complex flow environments.

Federated Learning (FL) enables collaborative and privacy-preserving model training across distributed clients, but most existing FL systems implicitly assume data stationarity. In real-world settings-such as healthcare, industrial IoT (IIOT), cybersecurity, and smart cities-data streams are inherently non-stationary, leading classical FL methods to suffer from performance degradation, instability, and catastrophic forgetting. Continual Learning (CL) addresses learning under evolving data distributions but has been largely studied in centralized settings, overlooking key constraints of federated systems, including privacy, limited communication, and client heterogeneity. Federated Continual Learning (FCL) emerges at the intersection of FL and CL, aiming to support lifelong, adaptive, and privacy-aware learning over distributed and non-stationary data. This survey provides a comprehensive and systematic overview of FCL. We first present a formal definition of the FCL problem and clarify its distinctive characteristics. We then analyze the limitations of classical FL under non-stationary conditions, highlighting how CL principles support long-term adaptation. To organize the rapidly growing literature, we propose a multi-dimensional taxonomy of FCL approaches. Furthermore, we review representative application domains and data modalities, summarize commonly used evaluation metrics, and discuss experimental perspectives for assessing long-term performance and forgetting. Finally, we highlight key open challenges, including handling extreme heterogeneity under temporal drift, designing scalable and privacy-preserving memory mechanisms, and establishing standardized benchmarks. This survey aims to serve as a reference and a roadmap for advancing FCL toward robust and deployable real-world systems.

Skill documents, structured natural-language instructions that guide Large Language Model (LLM) agents, are critical to modern agent frameworks, yet LLMs struggle to write skills that actually work. On SkillsBench, human-authored skills improve pass rates by 16.2 percentage points, while LLM-authored skills provide no measurable gain. We introduce SkillAxe, a fully unsupervised framework that enables LLMs to iteratively diagnose and refine their own skills. SkillAxe decomposes skill quality into four interpretable dimensions (quality impact, trigger precision, instruction compliance with fault attribution, and solution-path coverage), producing structured improvement briefs that require no ground-truth labels, test suites, or environment rewards. On SkillsBench, SkillAxe improves pass rates by 28\% relative over unimproved LLM skills and closes 47--67\% of the gap to human-authored skills. We validate the approach as a continuous improvement engine in the wild on SpreadsheetBench, where a SkillAxe-built skill library learns from past agent trajectories and raises pass rate from 16.0\% to 52.0\% using only 22 skills.

Distillation of a language model intended to transfer benign behavior to a student model may also transfer undesirable characteristics, if they are present in the teacher model, a phenomenon known as subliminal learning. While qualitative evidence supports the existence of this effect, its magnitude has not been systematically characterized. This study quantifies subliminal behavioral transfer ratios by steering two teacher models (Llama-2-7B-Chat and Qwen2.5-7B-Instruct) at varying steering strengths and distilling student models using only benign data. Evaluation on 100 JailbreakBench prompts with GPT-4.1, serving as the evaluator, indicates that transfer is robust but exhibits distinct scaling behaviors. Llama-2 demonstrates a sharp threshold ($τ= {0.25,0.32} \ \text{beyond} \ α= -0.15$), whereas Qwen2.5 displays continuous and higher levels of transfer ($τ$ up to $0.61$).

Understanding and forecasting lake dynamics is critical for monitoring water quality and ecosystem health across lakes and reservoirs. While machine learning methods have been recently applied to ecological time-series data, existing works assume regular sampling in time and depth, and struggle to generalize across lakes with heterogeneous variables, depths, and observation patterns. To address these limitations, we introduce \textsc{LakeFM}, a foundation model for aquatic systems, pre-trained on large-scale ecological datasets comprising both simulated and observed lakes. Through extensive empirical evaluation, we show that \textsc{LakeFM} learns meaningful representations spanning broader lake-level characteristics, and achieves competitive or often superior-forecasting performance compared to existing time-series foundation and non-foundation models, while producing physically plausible predictions consistent with real-world lake dynamics.

Data leakage -- contamination of a model with information unavailable at baseline -- is the dominant reproducibility failure in machine-learning-based science, yet detection tools require training code, external data, or domain expertise. None operates on the artifact an auditor most often holds: the model's output. We ask what can be decided about leakage from predictions and outcomes alone. We give a decision-theoretic framework in which leakage diagnostics are functionals of the predicted-risk/outcome law, parameterized by a threshold-weighting linked to proper scoring rules and decision-curve analysis. We prove a sharp impossibility: a recalibrated leak matching an honest model's calibration and discrimination is indistinguishable from honest performance by \emph{any} function of the predictions, so the broad class is detectable only against an externally supplied ceiling on achievable discrimination. We then prove what leakage cannot hide: a near-deterministic subgroup -- the signature of a near-label leak -- produces a sustained unit-purity head that no legitimate predictor of a non-deterministic outcome can manufacture, yielding a prior-free test. These results organize leakage into a trichotomy -- miscalibrated, broad-calibrated, and deterministic -- each with a matched detector and failure mode. We validate on UK Biobank using time-windowed comorbidity leakage with known, graded severity, measuring a detection floor of $Δ\cstar \approx 0.007$ on this endpoint, below which residual leakage is undetectable from output and too small to alter conclusions. The numerical floor is cohort- and endpoint-specific; the structural lesson is general: output-only detection fails where residual leakage is indistinguishable from an honestly stronger predictor. The test returns a verdict on a prediction vector in under a second on commodity hardware.

The cost signal that constrained-RL algorithms optimize against is almost always reactive: the simulator emits a non-zero cost only after a collision has begun, and the Lagrange multiplier of PPO-Lagrangian grows only after the episode budget has been exceeded. At race speeds, where collisions are instantaneous and irreversible, any safety mechanism that waits for cost to accumulate is structurally too late. We present VLM-Safe-RL, a framework that integrates a frozen vision-language model into the CMDP Lagrangian update as an anticipatory cost term. The framework comprises four contributions: (i) Decoupled Dual-Path CLIP, independent reward/cost paths that respect the CMDP's factorization; (ii) VLM-Lagrange, an augmented multiplier update that incorporates a per-step VLM cost as an anticipatory term; (iii) Confidence Gating, a Bayes-optimal weight derived from a logistic noise model on the CLIP margin; and (iv) VLMPPOLag, the composed algorithm. On Safety-Gymnasium FormulaOne L2, our principal evaluation ($n{=}5$ seeds, $10^{6}$ steps, budget $d_{\text{lim}}{=}25$) VLMPPOLag$+$Conf is the only configuration in our default budget comparison that simultaneously retains substantive return ($J_r{\approx}40$) and holds cost within budget on a majority of seeds; the five constraint-aware baselines (PPOLag, CPO, CPPOPID, CPO-CLG, PPOLag-RND) each fail at least one requirement. The mechanism generalizes to held-out MetaDrive Medium (catastrophe rate $41\%{\to}26\%$, 95\% bootstrap CI $[-26,-5]$\,pp) and shows directionally consistent transfer to Bullet Safety-Gym; we report honestly where it does not (MetaDrive Easy/Hard, Qwen2-VL backbone) and trace the Hard failure to a Lagrangian-regulation pathology rather than the VLM signal itself. To our knowledge, this is the first work to use frozen VLM signals as an anticipatory cost term inside the CMDP Lagrangian update.

Retrieval-Augmented Generation (RAG) systems are vulnerable to corpus poisoning attacks that manipulate downstream model outputs through malicious knowledge injection. Existing studies mainly evaluate poisoning under simplified retrieval settings, overlooking practical RAG pipelines involving document chunking, dense retrieval, reranking, and grounded generation. In this paper, we revisit corpus poisoning under realistic multi-stage retrieval pipelines and show that many existing attacks substantially degrade after reranking despite achieving high retrieval-stage relevance. We identify retrieval granularity mismatch as a key reason for this failure: document-level adversarial signals are often fragmented during chunking, while rerankers favor locally coherent and answer-bearing passages rather than globally optimized semantic similarity. Based on this observation, we propose Chunk-aware and Rerank-Consistent Poisoning (CRCP), a poisoning framework that jointly optimizes retrieval relevance, reranker consistency, and chunk-boundary robustness. CRCP explicitly models chunking transformations during optimization to generate locally self-contained adversarial passages that remain effective under varying chunking configurations. Experiments on standard RAG benchmarks with multiple retrievers and rerankers show that existing poisoning methods are highly sensitive to chunk size and reranking strategies, whereas CRCP achieves substantially higher attack success rates and stronger robustness across realistic retrieval pipelines. Our findings highlight an important realism gap in current RAG security evaluation and suggest that poisoning in modern RAG systems should be studied as a multi-stage retrieval consistency problem rather than a retrieval-only problem.

Health digital twins (HDTs) promise patient-specific modeling and decision support but current approaches remain structurally fragmented: monolithic models that address a single organ or task lack cross-scale fidelity, while system-level twins lack generalizable architectural frameworks. We propose OmniBioTwin, a System-of-Twinned-Systems (SoTS) framework that organizes HDTs as modular computational entities coupled through explicit interaction operators within a multi-layer network architecture. The framework comprises seven coordinated layers - spanning data integration, autonomous twin modeling, cross-scale coupling, temporal synchronization, and human-in-the-loop decision support. We demonstrate OmniBioTwin by instantiating a multiscale twin for glucagon-like peptide-1 (GLP-1) signaling pathways in Alzheimer's disease, illustrating how molecular, cellular, and organ-level twins can be composed and coupled within a unified system.

Spectral methods rely fundamentally on the stability of principal eigenspaces under random perturbations. Classically, this stability is quantified by the Davis-Kahan and Wedin theorems, which bound the eigenspace error using the operator norm of the noise and the relevant spectral gaps. While these worst-case bounds are sharp for arbitrary deterministic perturbations, they can be wasteful in the low-rank signal-plus-random-noise setting, as they fail to capture the fine-grained interaction between the signal geometry and the noise distribution. In this paper, we study the spectral perturbation of signal-plus-noise matrices corrupted by sparse, random noise with an arbitrary, inhomogeneous variance profile. We demonstrate that under heterogeneous noise variances, the empirical eigenvectors suffer a systematic, deterministic geometric bias that is entirely invisible to classical perturbation bounds. By leveraging the Quadratic Vector Equation (QVE) and establishing fine-grained isotropic local laws, we derive near-optimal, non-asymptotic perturbation bounds for the leading eigenspaces in the operator and $2\to\infty$ norms. The bounds separate the usual signal-to-noise contribution, stochastic fluctuations, and structured geometric bias terms determined by the alignment between the signal eigenspaces and the row-wise variance profile.

Access control in large language models (LLMs) requires modular mechanisms to enable domain-specific behavior without retraining or cross-domain interference. A common hypothesis is that interference during adapter composition arises from overlap in linear parameter updates, suggesting that enforcing orthogonality or directional independence should improve multi-domain performance. We test this hypothesis using DoRA-RBAC, a hierarchical adapter composition framework based on weight-decomposed low-rank adaptation. We compare conventional Euclidean merging with a geometry-aware Riemannian-inspired merging strategy that approximates the Frechet mean via normalized directional averaging across multiple QA benchmarks (GPQA, PubMedQA, SimpleQA, WMDP) on LLaMA-3.1-8B and Mistral-7B. Our results show that while single-domain performance matches LoRA, geometry-aware merging provides no consistent advantage over standard averaging in multi-domain settings.Diagnostic analysis further reveals that angular alignment and orthogonality of adapter updates are weak predictors of composition performance. These findings suggest that adapter interference is not governed primarily by parameter-space geometry, but is instead consistent with interactions in shared nonlinear representations.

Humans process rich auditory environments through tightly integrated cognitive capabilities such as audio perception, audio reasoning, and memory. Despite recent progress in large audio-language models (LALMs) across speech understanding and multimodal audio reasoning, current evaluation paradigms remain largely task- or modality-centric, focusing on end performance while overlooking underlying auditory cognitive behaviours. This reveals a fundamental gap between how auditory cognition is understood in humans and how it is evaluated in LALMs, particularly in the lack of frameworks that operationalise cognitive principles beyond task-level metrics to systematically capture model behaviour. In this work, we introduce RAIL, a human-centric evaluation paradigm grounded in the Cattell-Horn-Carroll (CHC) cognitive framework. RAIL formalises auditory cognition into five core capabilities and develop them into structured evaluation tasks that probe how models process, retain, and integrate auditory information. We further construct a cognitively grounded benchmark with principled data curation and human-aligned evaluation protocols. Evaluating 26 state-of-the-art LALMs, we find that current models exhibit highly uneven performance across cognitive abilities. RAIL establishes a new evaluation paradigm that moves beyond task-centric benchmarking toward cognitively grounded assessment of auditory intelligence.

Gradient-based inversion of reaction-diffusion systems is typically approached via surrogate models or physics-informed neural networks (PINNs), while the most direct route, backpropagation through the PDE's structure itself, has largely been avoided. We pursue this direct route as a diagnostic probe, backpropagating a steady-state loss through unrolled Gray-Scott simulation to recover its parameters, with no surrogate or neural-network augmentation. Optimization fails to converge, and plotting the landscape directly locates the failure in its geometry -- flat plateaus with no gradient signal, bounded by sharp cliffs that align with bifurcation boundaries -- a structure that recurs across loss functions and is inherited however the gradients are routed to parameters. Reading this minimal setup as an ablation of PINN, we disentangle each component's role: with the neural network fixed, the residual loss is quadratic in the PDE parameters and yields a smooth landscape, so it alone already avoids the pathology, by implicitly encoding the full PDE dynamics across all initial conditions. The neural network, for its part, cannot repair an ill-posed parameter subspace, and so serves only to complete the observed data -- a division of labor not previously made explicit. These findings carry concrete design implications for PINN-type methods and a broader heuristic on when added dimensions actually help.

Retrieval-Augmented Generation (RAG) pipelines are compute-intensive, combining embedding, retrieval, reranking, and large language model (LLM) generation. Running them entirely on-device benefits privacy, latency, and offline use, but the energy cost of CPU inference is a major barrier. We present what is, to our knowledge, the first end-to-end RAG pipeline that runs all neural stages -- embedding, reranking, and LLM generation -- on the Qualcomm Hexagon NPU of the Snapdragon X Elite. Profiling on a Dell XPS 13 laptop, we compare NPU-accelerated RAG against CPU and OpenCL/Adreno GPU baselines on indexing and query workloads. On indexing, the NPU achieves 9.1x higher embedding throughput and 12.3x less system energy. On a 120-query Wikipedia-passage benchmark, it delivers 18.1x faster LLM prefilling, 4.0x lower end-to-end query latency, and 4.0x less system energy than the CPU baseline; the same workload on the integrated GPU is 1.7x slower than CPU and uses 6.5x more energy than the NPU. A GPT-4.1 LLM-as-judge evaluation finds NPU answer quality on par with CPU and GPU within evaluator noise (mean 9.32 vs. 8.95 vs. 9.03 on a 1-10 rubric), with 86.7% of queries scoring identically across all three backends. On the Snapdragon X Elite / Hexagon class of laptop SoC, the NPU thus enables practical, energy-efficient on-device RAG without quality regression -- a sustainable path toward green edge intelligence that we expect to generalize to comparable mobile NPUs (Apple Neural Engine, Intel NPU, MediaTek APU) as their software stacks mature.

Designing molecules with target properties is most useful when candidate structures are accompanied by feasible synthetic routes. We introduce My Chemical Harness, a route-native evolutionary framework for goal-directed molecular design in which the search population consists of executable synthetic pathways rather than isolated molecular graphs. Each route is built from purchasable building blocks and reaction templates, executed by deterministic chemistry tools, and scored through task-specific molecular oracles. Large language models (LLMs) are used only as strategy controllers that select high-level preferences over route length, move type, reaction families, motifs, and exploration pressure, while local code performs route construction, validation, deduplication, scoring, selection, and memory updates. This separation lets the LLM guide exploration without allowing it to introduce hallucinated products or unsupported reaction steps. On a soluble epoxide hydrolase proxy task, our LLM agent improves over single pass LLM and deterministic controllers, reaching state-of-the-art performance across the sEH score, synthetic accessibility score, and AiZynthFinder success rate metrics. These results suggest that constrained LLM agents can play a significant role in molecular discovery without requiring training, fine-tuning, or dedicated generative models.

Bernstein--Schur kernels are products of a finite-feature kernel (one with an explicit finite-dimensional feature map) and a completely monotone shift-invariant kernel: nonstationary kernels that fall between the shift-invariant and dot-product templates random features usually exploit, so in general neither Bochner sampling nor polynomial sketching applies to the full kernel directly. We give one random-feature construction for the whole class that \emph{randomizes both factors: it sketches the finite modulation and randomizes the completely monotone radial factor, sampling the latter's one-dimensional Bernstein--Widder scale and then applying Gaussian random Fourier features (whose frequency is still $d$-dimensional). The feature dimension is then $Dm$, set by the sketch size $m$ and the radial-draw count $D$, free of the $O(d^2)$ size of the exact modulation feature. Keeping the modulation \emph{exact} is the analyzable limit ($m\to\infty$): there we prove unbiasedness, an exact variance for the recommended flat estimator, an expected matrix-Bernstein operator-norm bound (with a matching high-probability tail) controlled by the top eigenvalues of the kernel and modulation Gram matrices together with an intrinsic dimension rather than the crude $N\max_{ij}$ entrywise route, and a deterministic relative-spectral kernel-ridge stability result. By conditioning on the sketch, the doubly-randomized estimator inherits the same intrinsic-dimension operator-norm guarantee plus a single additive sketch term, tunable by $m$ independently of $D$. The motivating instance is the biased $yat$-kernel $k_{yat,b}(w,x)=(w^\top x+b)^2/(\|w-x\|^2+\varepsilon)$, $b\ge0$, whose family span contains the inverse-multiquadric kernel by finite differences in $b$; for it the radial mixture is the IMQ spectral sampler, and one frequency per scale is variance-optimal at a fixed radial-feature budget.

Hallucination detection in large language and vision-language models is increasingly framed as selective prediction, where a detector assigns a confidence score and abstains when confidence is low. Unsupervised sampling detectors (Semantic Entropy) avoid labels but plateau in quality, while supervised probes attain stronger in-distribution scores yet degrade sharply when calibration labels are scarce. We recover the response manifold of an LLM as the density ridge of a kernel density estimate built on a six-dimensional kinematic feature map of hidden state generation trajectories. A test generation is scored by the negated Euclidean distance from its projected feature point to the nearest ridge vertex, yielding a low-dimensional geometric skeleton of the stochastic output distribution. We evaluate against Semantic Entropy, topological methods, and log-probability on six QA benchmarks (HaluEval-QA, TriviaQA, GSM8K, POPE, ScienceQA, A-OKVQA) using eight text and vision LLMs in a deliberately label-scarce protocol ($n_{\text{cal}}{=}200$ queries, $N{=}5$ generations). Our ridge-based score beats on AUROC with 5-20 points gain, while demonstrating tempered degradation under calibration-label scarcity.

Transitioning bidirectional video diffusion models into an autoregressive paradigm improves the interactivity of video world models, but existing causal pipelines need many stages (control fine-tuning, autoregressive training, causal initialization, few-step distillation) and still trail bidirectional models in quality due to error accumulation. Recent world models such as Yume-1.5 and Matrix-Game-3.0 instead adopt a bidirectional autoregressive approach, gaining fidelity and stable long-horizon rollout from self-correcting error propagation, yet open-source frameworks (e.g., minWM) support only causal models. We present BiWM, the first full-stack framework for interactive video world models under the bidirectional autoregressive paradigm, jointly optimizing generation quality and inference speed. From a pretrained video backbone, BiWM injects camera control by fine-tuning, then runs a few-step Distribution Matching Distillation (DMD) stage that turns the backbone into an action/camera-controllable world model: just two training stages instead of four in minWM, converging in a few hundred steps on 8xH200 GPUs. A single recipe spans Wan2.1-1.3B, Wan2.2-5B, HunyuanVideo-1.5-8B, and LTX-2.3-22B, and also supports secondary fine-tuning of existing bidirectional models. BiWM enables real-world camera control where minWM loses controllability, integrates pluggable history compression (FramePack-style and PackForcing-style) for long rollouts, and offers an optional NVFP4 4-bit training/inference pipeline. To counter DMD's mode-seeking degradation, we add GAN and mass-covering forward-KL objectives that preserve scene dynamics. We open-source BiWM for resource-constrained research and high-fidelity environment simulation.

Postprandial hyperglycemia is a key risk factor for metabolic disorders; however, existing dietary guidance is often static, impractical, and insufficiently personalized, providing recommendations that are difficult to follow or not impactful. While recent advances leverage continuous glucose monitoring (CGM) and machine learning to predict glycemic responses, these approaches are largely predictive and lack actionable guidance. Moreover, recommendation systems are often misaligned with user goals and require extensive input. We present MetaPlate, a counterfactual explanation (CF) guided, context-aware decision-support framework that generates personalized meal recommendations to mitigate postprandial glucose excursions in healthy adults. MetaPlate integrates multimodal data, including CGM readings, wearable-derived physiological signals, and user-provided meal inputs from $25$ individuals to model pre-meal context. A machine learning model predicts glucose response, while a CF optimization module adjusts meal composition modifying macronutrient amounts to maintain glucose levels within a target range ($\leq 140$ mg/dL). An LLM-based retrieval-augmented generation (RAG) layer enhances interpretability by producing human-readable recommendations using constrained search of the USDA food database. We evaluate MetaPlate via a structured expert-in-the-loop assessment with registered dietitians (RDs), comparing performance before and after prompt refinement. Results show improvements in meal realism, portion suitability, and recommendation likelihood, with expert feedback indicating a shift from clinically implausible outputs to actionable, contextually appropriate recommendations. Our findings emphasize the importance of domain knowledge and structured constraints in LLM-driven systems and highlight the potential of MetaPlate as a real-time personalized dietary decision-support tool.

AI evaluation results are produced at scale but reported inconsistently across leaderboards, model cards, benchmark papers, and company blogs. The cost is interpretive: readers cannot reliably compare results across sources, identify what a report omits, or trace an aggregate claim to its underlying evidence. Recent efforts address isolated components but leave three gaps: they cover only narrow slices of the evaluation lifecycle and do not compose into a single interpretable record; they specify static representations that do not differentiate the questions different stakeholders bring to the same evidence; and they remain proposals on paper, lacking the extraction infrastructure required for adoption at scale. We present \EvalCards{}, an operational reporting layer that composes benchmark metadata, evaluation run data, and model metadata into a unified record. We (1) derive a reporting schema from a structured review of 52 papers and 10 stakeholder interviews, (2) implement four interpretive signals (reproducibility, documentation completeness, provenance and risk, and score comparability), rendered through reader modes calibrated to research and non-research audiences, and (3) deploy a monitoring tool that applies \EvalCards{} across 5,816 models, 635 benchmarks, and 101,843 results, surfacing systematic gaps in current reporting practice.

The NISQ era places stringent constraints on quantum computation, where noise and decoherence fundamentally limit performance. In classical deep learning, model robustness and resilience to perturbations are well studied: deep neural networks (DNNs) maintain high performance despite pruning, noise injection, and structural perturbations due to inherent redundancy in their representations. A central challenge in quantum machine learning is to transfer this notion of robustness to quantum neural networks (QNNs) under realistic NISQ noise. While classical deep learning exhibits robustness through structural redundancy, analogous principles for QNNs remain underdeveloped. We propose JGRA: a framework for assessing robustness in noise-aware QNNs via Jacobian geometry, capturing model sensitivity to parameter perturbations induced by noise. Our method includes entropy-matched noise calibration, noise-aware training, and noise-conditioned Jacobian extraction, yielding geometric descriptors that link clean-regime structure to noisy inference behaviour. We also empirically demonstrate that these descriptors encode predictive information about robustness under unseen noise.

Many multivariate dynamical systems are observed only through trajectories, leaving the mechanisms governing their joint dynamics hidden. Existing approaches can impose interpretable dynamics or learn flexible state transitions, yet the resulting interaction structure is typically either specified in advance or left implicit within the learned dynamics. We introduce MF-Net, a recurrent dynamical model that represents all variables in a shared field state and updates this state through a learned relation law. Each variable carries a field component, and these components evolve jointly through a learnable mechanical transition. Here, mechanical refers to the relation-to-motion organization of the transition, where learned relations shape state-dependent flows, field responses, and motion tendencies that move the field state forward. The resulting structure is part of the rollout itself: learned relations influence how the field moves, and the same internal quantities support both forecasting and structural readout. Across known-law interaction systems, chaotic benchmarks, real neural recordings, and ecological time series, MF-Net achieves competitive short- and medium-horizon forecasting while retaining inspectable structural readout. On the 40-dimensional Lorenz--96 testbed, MF-Net achieves an eight-step $R^2$ of $0.798\pm0.018$; across five seeds, its learned relation matrix recovers the local coupling support with a local/nonlocal strength ratio of $19.80\pm1.00$ and Precision@$K$ of $1.000\pm0.000$. MF-Net provides a structure-readable dynamical modeling framework in which learned relations are trained through forward evolution and, on real data, interpreted as functional predictive couplings under appropriate observational limits.

Conditional generators provide a natural tool for controllable generation, including settings where the desired condition is a new composition of observed attributes or experimental factors. In many applications, especially in scientific domains, such models are attractive to explore conditions for which real samples are rare, expensive, or not yet observed. However, this creates a circularity for evaluation: standard conditional quality metrics require a reference target distribution, but in the extrapolative regime that distribution is unavailable by definition. We address this problem with a post-hoc, per-sample trust score for assessing conditional samples using only the training distribution. The score combines two estimable quantities: global realism, measuring compatibility with the real data manifold, and attribute-wise faithfulness, measuring whether a sample is closer to the requested attributes than to plausible alternatives. We show that the score can recover meaningful comparisons across extrapolated generations, under a mild coverage condition on the observed attributes. These comparisons enable effective filtering, ranking, and abstention of generations and can be used directly on off-the-shelf pretrained models. In biological imaging, selected samples preserve real morphological structure better and improve downstream predictive performance, while similar gains are observed on controlled vision benchmarks. Finally, we show how the score can be applied during generation, enabling abstention before full decoding. Code is available at https://github.com/berkerdemirel/faithful-cond-gen.

Realizing the vision of 6G connected robotics requires reconciling high-performance collaborative control with the rigid spectral limitations of physical wireless channels. In realistic collaborative sensing scenarios, spectral resources are quantized into finite physical resource blocks or orthogonal subcarriers, rendering simultaneous transmission by all agents infeasible. To address this, we propose Multi-Agent Semantic K-Scheduling (MASK), a control architecture designed to sustain robust, risk-aware coordination under strict instantaneous bandwidth caps. We introduce Arbiter-Assisted Semantic Information Gating (A-SIG), a lightweight coordination mechanism that enforces hard access constraints by scheduling only the top-K agents based on locally computed semantic importance scores. By aggregating these prioritized observations into a compact latent state, a self-supervised global encoder enables a distributional policy to mitigate tail risks despite data sparsity. We evaluate MASK across diverse benchmarks, demonstrating that it matches the performance of communication-unconstrained baselines even when channel access is restricted to a small fraction of the swarm size. Furthermore, the framework exhibits inherent resilience to packet erasures, validating semantic scheduling as a critical enabler for resource-constrained 6G systems.

Understanding tactical organisation of association football, hereafter referred to as football, requires identifying distinct match phases. Yet in-possession phases are rarely directly observable and are shaped by evolving tactical intentions, rather than spatial patterns alone. This study proposes a data-driven framework for identifying in-possession match phases from spatiotemporal tracking data. Seven German Bundesliga matches recorded at 25 Hz with TRACAB were analysed. A hierarchical phase model was defined with three tactical intentions (Invade Opponent Space, Keep Possession, Scoring) and six phases (Build Up, Progression, Counter Attack, Maintenance, Sustained Threat, Finishing). A Temporal Graph Attention Network (T-GAN) was developed to combine frame-level player-interaction graphs, contextual features, and Transformer-based temporal modelling. Performance was evaluated using frame-level F1 and a sequence-aware Intersection over Truth-Dominance (IoT-D) metric. T-GAN achieved macro-average frame-level F1 scores of 0.87 at the intention level, 0.76 for invasion-related phases, and 0.79 for scoring phases. At the sequence level, mean diagonal IoT-D F1 increased from 0.68 to 0.79 for intentions and from 0.61 to 0.71 for phases after post-processing, indicating improved temporal coherence. Model comparisons showed that sequence modelling was the main driver of segmentation quality, while graph-based relational modelling was particularly beneficial for Counter Attack recognition. Exploratory player attention analysis further suggested that wide and midfield positional groups contributed strongly to phase discrimination. Overall, the framework translates continuous tracking data into tactically interpretable in-possession phase representations, with potential applications in automated match annotation, tactical analysis, and playing-style profiling.

Understanding how transformer representations evolve across layers, not merely what they encode, remains an open problem in mechanistic interpretability. We recast the transformer forward pass as a discrete population trajectory through a high-dimensional representation manifold, drawing on geometric tools from computational neuroscience. Rather than probing for pre-specified features, we characterize trajectory geometry using five metrics computed directly in the ambient space: trajectory length, curvature, a semantic convergence index, layerwise cosine similarity, and representational stability. Across three model families (GPT-2, TinyLlama, Qwen2.5) and five controlled prompt families, we report four findings. First, semantically related prompts converge significantly in middle-to-late layers (peak CI 0.41--0.58, p<0.001, Mann-Whitney U), consistent with attractor-like dynamics. Second, reasoning tasks produce trajectories of greater curvature than lexical variations (0.71--0.83 rad vs. 0.27--0.31 rad), suggesting curvature encodes computational complexity. Third, ambiguous tokens exhibit trajectory bifurcation with up to 5.6x representational separation by the final layer, absent in unambiguous controls. Fourth, layerwise cosine similarity reveals a universal three-phase structure: encoding, elaboration, and output preparation, consistent across all three architectures. All four effects vanish under shuffled-layer and random-embedding controls. We release a fully open-source, model-agnostic pipeline and argue that trajectory geometry constitutes a principled, probe-free lens for mechanistic interpretability.

Generative models are increasingly used to propose designs, data, and control actions for physical systems, yet many such systems are governed by hard physical constraints rather than by perceptual plausibility. Semiconductor manufacturing provides a demanding test case: generated masks, layouts, synthetic defect data, and process recipes must obey lithography, transport, reaction, and device-physics constraints, because physically invalid samples are not merely low quality but unusable. This Perspective argues that semiconductor manufacturing exposes a broader computational-science challenge, namely that generative AI for constrained physical domains must be physics-informed by construction, not corrected only through post-hoc filtering. We survey the emerging architectural toolkit, including physics-informed diffusion, PDE-constrained variational models, neural-operator priors, and conservation-law-respecting generative networks, and show how it connects to differentiable lithography, TCAD, process simulation, and autonomous experimentation. We identify four integration patterns between generative models and physics-based simulators, and we propose a research agenda centered on physics-fidelity benchmarks, differentiable simulator infrastructure, and multimodal foundation models for physical design and manufacturing. The central claim is analytical rather than rhetorical: where physical validity is the binding criterion of success, architectures that enforce it by construction should be expected to outperform those that filter for it after the fact, and the fab is the setting where this distinction is sharpest.

Generating novel, feasible, and high-quality research ideas is an important yet challenging task in scientific discovery. Recent Large Language Model (LLM)-based methods often ground idea generation with retrieved literature, but the retrieved evidence is usually provided as flat text, such as titles, abstracts, or summaries. Such flat contexts may contain redundant or weakly relevant information, while making cross-paper relations among problems, methods, mechanisms, and findings difficult to identify and trace. To address this challenge, we propose Graph2Idea, a knowledge graph-guided framework for retrieval-augmented scientific idea generation.Graph2Idea first retrieves papers according to the input topic, transforms them into structured knowledge triples, and dynamically constructs a target-centered knowledge graph to make literature relations explicit. It then extracts compact graph-derived contexts that retain target-relevant relational evidence while reducing noisy textual input. Based on these contexts, a two-stage generation process first identifies promising research directions and then guides the LLM to synthesize candidate ideas from graph-grounded evidence. Experiments on a scientific idea generation benchmark show that Graph2Idea outperforms representative baselines under the automatic evaluation protocol. Compared with the strongest baseline scores, it improves Novelty from 0.45 to 0.52, Quality from 0.24 to 0.29, and Feasibility from 0.22 to 0.28. These results suggest that graph-structured evidence helps LLMs generate research ideas through more explicit, compact, and traceable recombination of prior scientific knowledge.

Conventional LLMs keep the full KV cache loaded during decoding, causing a severe GPU memory bottleneck for ultra-long context serving. In this report, we propose Lookahead Sparse Attention (LSA), a novel inference paradigm powered by a Neural Memory Indexer built upon the DeepSeek-V4 architecture. Rather than passively attending to all historical tokens, LSA proactively predicts future context demands and preserves only the query-critical KV chunks in the GPU memory. Crucially, we instantiate this architecture via a backbone-free decoupled training strategy. By formulating the indexer as a standard dual-encoder architecture, we train it independently using standard retrieval training frameworks without ever loading the massive backbone model into GPU memory. We demonstrate that this "less is more" paradigm significantly maximizes serving efficiency while acting as an effective attention denoiser in tasks that rely on long-term global memory. Across primary long-context evaluation suites (e.g., LongBench-v2, LongMemEval, and RULER), FM-DS-V4 compresses the average physical KV cache footprint down to merely 13.5% of the full-context baseline, while consistently preserving or slightly elevating downstream accuracy (+0.6% absolute margin on average). Crucially, at extreme 500K scales, FlashMemory suppresses the physical KV cache overhead by over 90% without destabilizing the backbone's core reasoning capacities.

We ask whether structural properties of intermediate grid states predict whether a symbolic ARC-AGI solver will succeed, framed as a test of conditional mutual information I(X;Y|task) > 0. Across 44,800 runs spanning two architecturally distinct solvers (beam search and Stochastic DFS), 400 ARC tasks, 28 configurations per solver, and both training and evaluation splits, hand-crafted grid descriptors measured at 50% trajectory completion discriminate successful from failed runs within the same task (mean within-task best-feature AUC = 0.885, p < 0.001 under within-task label permutation). Most predictive content lies along a single grid-complexity axis. The result generalizes across solver architectures: a feature selected on one solver predicts success on the other with AUC 0.747-0.762 in all four transfer directions (p < 0.001, leakage controlled). On a pre-registered held-out set of 41 reliable tasks, the frozen feature n_components_final achieves AUC = 0.765 (95% CI [0.717, 0.810], p < 0.001), robust under task-clustered bootstrap resampling and cross-solver task collapsing. The signal is not explained by solver capacity (configuration-residualized AUC = 0.927 and 0.896 for beam search and SDFS, p < 0.001) and is only weakly coupled to score trajectories (R^2 approximately 0). Early stopping at 50% completion reduces beam-search compute by 33.6% while retaining 98.9% of solves; degenerate-trajectory detection reduces SDFS compute by 65.3% with no solve loss. Finally, on 229 of 400 evaluation tasks the DSL primitive library produces no valid transition from the input grid. This 0-step collapse is invariant to search budget and universally failed by beam search, indicating a DSL coverage limitation rather than a search-budget effect.

Scientists have historically relied on mathematical models based on differential equations to relate system inputs -- forces, fluxes, or heat sources -- to outputs, such as displacement, velocity, concentration, and temperature. These models rely on deep domain knowledge to determine the form of the governing differential equation, which is then calibrated with data by solving an inverse problem. In recent years, the field of Scientific Machine Learning has introduced a variety of alternative modeling strategies for physical systems. A method called Sparse Identification of Nonlinear Dynamics learns the governing equation as a sparse linear combination of terms in a user-defined library. Neural Ordinary Differential Equations construct the governing equation by taking in the state and its derivatives at the input layer of a neural network. Entirely foregoing the modeling framework of differential equations, neural operators directly learn a non-linear mapping between the system inputs and outputs. From inverse problems to neural operators, all of these modeling strategies can be conceptualized as data-driven machinery to predict a system's response over a range of inputs. It is then natural to wonder how exactly these various strategies relate to each other, and whether they can be neatly taxonomized. Drawing from the philosophical literature on scientific models, we argue that many model types have a common structure, differing only in the assumed model class of the input-output relation they define. Connecting to philosophical ideas on mechanism, and arguing that data from physical systems arises from solutions to parsimonious differential equations, we propose that only certain models are capable of mechanism discovery, and thus generalization. Our analysis is intended to unite apparently disparate modeling strategies and provide insight into their appropriate use cases.

Vision-Language-Action (VLA) models achieve strong benchmark performance but still struggle in real-world deployment with unseen objects, background shifts, and different robot embodiments. We argue that this stems from the lack of a unified geometry-aware manipulation representation, leaving existing VLAs vulnerable to low-level trajectory supervision, misaligned 3D features, and embodiment differences. To address this, we propose GEAR-VLA, a VLA framework for learning unified geometry-aware action representations for generalizable robotic manipulation. GEAR-VLA adopts coarse-to-fine action learning, where multi-source embodied pretraining equips the VLM with embodied reasoning and discrete action understanding before latent action tokens connect action semantics to a gradient-decoupled DiT continuous action expert. It further performs semantic-aligned 3D integration by aligning a trainable 3D spatial backbone with the VLA representation while freezing the original VLM-aligned visual pathway. To share this representation across robots, GEAR-VLA uses embodiment canonicalization, where embodiment-aware states and embodiment-invariant actions confine robot differences to the low-level interface. Extensive simulation and real-world experiments demonstrate strong generalization: GEAR-VLA achieves state-of-the-art performance on LIBERO, zero-shot LIBERO-Plus, and RoboTwin 2.0, reaches 85.9% success on AgileX and 81.0% on the pretraining-unseen LDT-01 embodiment, and obtains 90.1% success on a 6,360-trial universal grasping benchmark with 212 unseen objects. Code and models will be released at https://github.com/babynabeauty/GEAR-VLA.

Tissue graph counterfactuals ask how a cell's expression would change under altered spatial neighbor contexts. Such queries are central to predicting cell behavior in tissues, but lack a unified definition, with existing methods targeting specific intervention types or treating cells as i.i.d. In this work, we first formalize tissue graph counterfactuals as a class of spatial interventions that either rewire connections between cells (edge perturbation) or modify the expression of their neighbors (node perturbation). We then introduce Cellina (https://cellina.readthedocs.io) - a framework that uses supervised disentanglement to decompose a cell's intrinsic state from its spatial context, using the latter as a conditioning input for counterfactual predictions. Across benchmarks spanning over 2.5 million spatially-resolved cells in colorectal cancer and mouse brain, Cellina outperforms spatially-informed and non-spatial competitors in in-silico graph perturbations, disentanglement, and scalability. Additionally, we show that Cellina reveals biologically distinct cancer subdomains in an unsupervised manner and enables targeted neighbor perturbation simulations.

While recent text-guided video editing models excel at elementary tasks (e.g., style transfer, object insertion), real-world user requests are highly compositional. A single prompt often demands multiple coupled edits, such as modifying subjects, actions, and camera views, while strictly preserving unrelated spatiotemporal content. Existing benchmarks, heavily constrained by isolated edits and coarse global metrics, fail to diagnose how models handle such complex workflows. To address this gap, we introduce CoVEBench, a compositional video editing benchmark comprising 416 curated source videos, 626 multi-point editing instructions, and 9,990 fine-grained checklist items. Covering diverse editing dimensions, CoVEBench evaluates models via MLLM-judged instruction compliance and video fidelity, alongside automated metrics for video quality. Extensive experiments reveal that compositional editing remains a profound challenge: current models frequently omit edits, violate preservation constraints, or introduce artifacts when handling multiple operations simultaneously. CoVEBench provides a challenging, diagnostic testbed to advance video editing toward realistic user workflows.

We introduce GENERIC-FNO, the first neural operator to embed the full GENERIC (metriplectic) structure of nonequilibrium thermodynamics -- reversible, energy-conserving dynamics and irreversible, entropy-producing dynamics coupled through the degeneracy conditions -- directly in function space. Existing structure-preserving neural operators enforce at most a single conservation law or reversible (Hamiltonian) structure, while thermodynamically consistent learning has been confined to finite-dimensional, graph, or particle systems. GENERIC-FNO closes this gap: it learns the energy and entropy functionals as neural operators and parameterizes the Poisson and friction operators as diagonal Fourier multipliers sandwiched between rank-one projections that enforce the degeneracy conditions exactly, by construction, with no penalty term, update projection, or residual. The degeneracy identities hold to machine precision (residuals ~10^-13) for any initialization, dimension, or resolution, so the continuous-time dynamics conserve the learned energy and produce entropy exactly; the explicit time stepping adds only a small O(dt^2) drift (per-step residual ~10^-6). We further note that the (E,S,L,M) decomposition of a given flow is not unique, and introduce a gauge-invariant dissipation diagnostic separating reversible from dissipative dynamics independently of the learned functionals. Across three operator backbones (1D/2D FNOs and DeepONet) and four PDEs spanning reversible, dissipative, and mixed regimes, GENERIC-FNO preserves its exact structural guarantees zero-shot across a 4x super-resolution range (64 to 256), recovers the ground-truth ordering of physical dissipation, and is competitive with strong unconstrained and energy-penalized baselines, outperforming them on several dissipative and mixed problems at comparable or fewer parameters.

Multi-quadruped coordination has attracted increasing attention due to its enhanced payload capacity, broader contact coverage, and improved adaptability to challenging tasks. Existing methods for multi-quadruped manipulation typically focus on predefined or closed task families, often relying on multi-agent reinforcement learning (MARL) to train task-specific coordination policies. However, such methods struggle in open-ended continual learning settings, where tasks arrive sequentially and robots are expected to acquire new coordination skills while reusing previously learned ones without catastrophic forgetting. To address this challenge, we propose Conquer, a semantic skill-library framework that formulates continual multi-quadruped coordination as a retrieve-adapt-update process. First, to accommodate varying team sizes across tasks, we design a team-structured Self-Allies-Goal (SAG) backbone that supports variable-cardinality robot teams by explicitly modeling each robot's own state, teammate context, and task goal. For each incoming task, Conquer constructs a task-level semantic descriptor from pre-execution information and retrieves a relevant skill from the library for adaptation. After successful execution, Conquer updates the skill library by extracting trajectory-level semantic descriptors and organizing them according to semantic distance, thereby enabling continual skill accumulation and cross-task knowledge transfer. Simulation experiments show that Conquer achieves a final average success rate of 95.6%, demonstrating strong forward transfer and negligible catastrophic forgetting. Real-world rollouts on Unitree Go2 teams further validate the deployment feasibility of Conquer for practical multi-quadruped coordination. Simulation and real-robot demonstration videos are available at: https://conquer-project.pages.dev/.

Non-Hermitian systems can evade Anderson localization and exhibit delocalized states even in one dimension. Here, we show that such non-Hermitian delocalized states under periodic boundary conditions (PBC) are intrinsically critical, realizing the universality class of one-dimensional random Dirac fermions. By linking spectral winding to topological Anderson transitions via Hermitization, we demonstrate that the delocalized PBC states exhibit a Dirac-type criticality with universal algebraic correlations. In contrast to Hermitian systems, where this criticality occurs only at fine-tuned transition points, it emerges generically in non-Hermitian systems as a consequence of spectral topology. These results identify a universal mechanism by which non-Hermiticity promotes criticality, providing a unified description of non-Hermitian delocalization in one dimension.

Particulate contaminants decrease the power output of solar panels, the transparency of windows, and are detrimental to microelectronics, where even a single particle can induce a short circuit. Despite significant research on particle adhesion and self-cleaning, it remains unclear when and how a drop can remove a particle from a surface, thus efficiently cleaning the surface. Here, by combining lattice Boltzmann simulations and confocal microscopy experiments, we show that at least six different scenarios arise from the complex interplay between capillary and friction forces when a drop collides with a particle. Notably, the capillary force plays a dual role in particle removal: while its tangential component always drives removal, its normal component can also hinder it. By introducing a dimensionless capillary capture parameter, we can predict particle removal across a wide range of particle and surface properties. These results provide quantitative design principles for easy-to-clean surfaces that minimize water and chemical usage.

We develop a Heisenberg-Langevin theory of an exciton mirror accounting for the retardation and the long-range electron-hole exchange. A particular case of a strong transverse magnetic field is analyzed in detail. The optical bistability due to repulsion between the excitons inside the light cone appears to be prone to a modulational instability towards the non-radiative surface polariton modes. Above the corresponding threshold, the pumped 2D exciton gas acts as an optical parametric generator of twin polariton beams. Conversely, below the threshold, the mirror acquires the phase-conjugating properties.

The orbital angular momentum of light is an intriguing property for developing light driven applications. It emerged as an independent degree of freedom by which to manipulate light and, consequently, the interaction of light with matter. Several methods exist for the generation of light carrying orbital angular momentum, mostly employing transmitting or reflecting optical components, which radially modulate the phase profile of the light. As one of such components, transmissive q-plates established themselves as standard elements due to their usability over a broad wavelength range. Here, we present our approach to build a highly reflective q-plate based on a plasmonic metasurface capable of converting orbital angular momentum from the nanostructure to ultrashort laser pulses without temporal broadening. We highlight its working principle over a wide range of wavelengths for reflection under normal and gracing incidence.

Quantum integrability in a system is characterized by a large number of conserved charges. However, finding the conserved charges in a generic quantum system is highly challenging. The integrable matrix theory provides a unified framework to obtain the conserved charges in a certain class of systems. We demonstrate this framework in case of a single-particle system on a 1D finite-sized lattice with inhomogeneous nearest neighbor hopping to study the statistical properties of the system across its chaotic--integrable crossover from the perspective of the conserved charges. The eigenspectrum of our random matrix model is studied first. We then estimate the conserved charges and find their properties throughout the chaotic to integrable transition of the system. We calculate the number of conserved charges across this crossover and observe that it is nearly equal to the size of the system at its integrable limit. Our result suggests that the number of conserved charges, estimated using the integrable matrix theory, can be a measure of quantum integrability.

Graphene nanoribbons (GNRs) synthesized on metal substrates experience electronic coupling and screening from the underlying surface, which, although often weak, can modify their observed properties and complicate their transfer to device-compatible substrates. Intercalation of GNRs by self-assembled monolayers (SAMs) offers a possible route to reduce this interaction. Here, we investigate the intercalation of 7-atom-wide armchair graphene nanoribbons (7-AGNRs) on Au(111) using N-heterocyclic carbenes (NHCs). Low-temperature scanning tunneling microscopy and spectroscopy, Raman spectroscopy, and density functional theory calculations reveal that the adsorption geometry of the NHCs strongly influences the intercalation yield for GNRs. Methyl-substituted NHCs form flat-lying dimers that partially intercalate the GNRs, producing locally decoupled segments. In contrast, bulkier isopropyl-substituted NHCs form upright monomers that embed the GNRs within the monolayer, preventing intercalation. The low intercalation yield indicates that lifting the nanoribbon from the Au surface is energetically costly. These results establish molecular adsorption geometry and packing as key parameters controlling intercalation at GNR-metal interfaces, with implications for the rational design of decoupling layers for GNR-based device integration pathways.

Deep-layered machines, in which each node computes a Boolean function of all nodes below it, underpin deep learning and digital computation. Yet the statistics of their global output function remain poorly understood. We derive the exact finite-depth distribution of the output of a machine with width $k$ and depth $n$. The distribution depends only on the Hamming weight of the output, and as $n$ increases favors functions with low and high Hamming weights. But this bias peaks at a crossover depth proportional to $2^k$ before collapsing onto the constant functions true and false.

The physics of active matter, wherein constituent particles consume energy to generate autonomous motion, has revolutionized non-equilibrium statistical mechanics. While a large body of work has successfully elucidated the behavior of dilute active systems, the dense regime -- characterized by ``active glasses and active solids'' -- presents profound challenges that defy conventional theoretical frameworks. Recent observations reveal two striking features in these dense systems: an apparent enhancement of Mermin-Wagner-Hohenberg (MWH) fluctuations leading to anomalous long-wavelength density fluctuations, and a remarkable correspondence between activity-induced annealing and annealing via oscillatory shear. In this perspective article, we propose a novel approach toward a deeper understanding of dense active matter: by developing active Hamiltonian models as equilibrium reference frameworks, we map out pathways toward non-equilibrium active systems. This strategy allows us to elucidate both the correspondence between driven and active systems and the enhanced MWH fluctuations, which likely arise from a strong coupling between spatially random active forces and long-wavelength density (phonon) modes. We outline a comprehensive roadmap employing complementary approaches, including the active Hamiltonian formalism, comparative studies of oscillatory shear in active and passive solids, and investigations of chiral active matter. Establishing this activity-oscillatory shear correspondence across diverse systems is essential to demonstrate its universality, reveal the underlying large-scale emergent physics, and place our hypothesis on a firmer theoretical ground.

The dynamics of nanoparticles within nanoscale liquid environments exhibit a range of complex phenomena driven by the interplay of processes at varying length scales. While these dynamics have profound technical implications, such as in nanoscale catalytic kinetics, ion-transport pathways in energy storage, and macromolecular crowding in biological systems, real-space imaging of dense, confined nanoparticle assemblies remains a significant challenge. Here, we present a liquid-phase transmission electron microscopy approach in which dense clouds of gold nanoparticles are formed within microfluidic channels, rendering the particle ensemble visible in bright-field electron imaging. This strategy enables direct imaging of different density-dependent particle ordering phenomena, including a local structuring of the colloidal liquid in nanoscale spaces, disordered dynamic clouds at high nanoparticle densities and the reversible formation of superlattice structures. Our results provide a unique window into the complex processes of colloidal self-organization at the nanoscale.

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 characterize the quasi-stationary distribution (QSD) of the bond directed-percolation line of the Domany--Kinzel automaton using a matrix-product-state representation of the probability distribution, obtained by projecting out the absorbing state and iterating the transfer matrix. Unlike moment- or sampling-based methods, this yields the full conditional distribution and direct access to information-theoretic diagnostics. The spatial structure of the QSD changes sharply across the transition: the active phase is bulk-like with finite density, whereas in the inactive phase the surviving activity collapses into a single flock occupying a vanishing fraction of the chain, with an internal filling that ranges from a single cluster deep in the inactive phase to a loose, partially filled group near criticality. This picture carries a sharp information-theoretic signature: throughout the inactive phase the bipartite mutual information of the QSD equals the entropy of a single binary choice -- whether the flock lies to the left or right of the cut -- so the surviving clusters together encode just one bit of positional information, corresponding to a single effective cluster. The approach extends matrix-product-state techniques to the projected eigenvector defining a QSD, opening information-theoretic diagnostics for absorbing-state systems that bulk-observable methods cannot reach.

Trace norms are fundamental to quantum information theory, yet in many-body systems their evaluation remains a major computational bottleneck, as it generally requires diagonalizing exponentially large operators. Here, we overcome this bottleneck by introducing a controlled tensor-network algorithm for estimating the trace norm of matrix product operators without full diagonalization. The key idea is to combine Zolotarev's rational approximation to the sign function with a variational formulation solved using a density-matrix-renormalization-group-like algorithm. The resulting approximation is systematically improvable, with its accuracy controlled by the rational approximation parameters and the spectral weight near zero. Beyond the reach of exact diagonalization, we demonstrate controlled trace-norm calculations for entanglement negativity, quantum fidelity and quantum Fisher information, achieving substantially improved accuracy over polynomial-based Lanczos approaches. Our results establish trace-norm-based quantities as practical tensor-network observables, opening a route toward tensor-network studies of quantum information in mixed states.

The quantum anomalous Hall effect in altermagnets is difficult to realize because spin-up and spin-down states remain degenerate at the $Γ$ point in the nonrelativistic limit. We start from the Bernevig-Hughes-Zhang model to incorporate nontrivial band topology. We demonstrate that the combined effects of an external magnetic field, spin canting, and ferroelectric orbital hybridization lift the degeneracy at the $Γ$ point, enabling electric-field control of the Chern number. A minimal two-dimensional d-wave altermagnetic model with band inversion then realizes a ferroelectrically tunable Chern insulator with spontaneous spin canting. The ferroelectric polarization controls the topological phase and the orbital angular momentum, enabling a rich phase diagram with C = $\pm 1$ and C = $\pm 2$ through a Berry-curvature reorganization linked to the spin canting response and ferroelectricity. Our results establish a symmetry-consistent route to electrically tunable Chern insulating phases in altermagnetic materials, opening opportunities for low-power topological and orbitronic devices.

Cells dynamically generate, transmit, and dissipate stress. Central to these processes is the actomyosin cortex, an active contractile material that drives cellular mechanical behavior. While prior studies have focused on freely contracting actomyosin systems, the role of mechanical constraints such as adhesion to boundaries remains less explored. To address this, we employ reconstituted actomyosin gels to investigate cellular contractility. We study contraction dynamics under pinned boundary conditions, where the gel is adhered transversely to two opposing surfaces, mimicking supracellular actomyosin networks in tissues and embryos. We find that pinned contraction leads to stress buildup, delaying contraction, producing intermittent dynamics, and generating spatially nonuniform strain fields. Stress is relieved through several pathways, including active-stress-driven symmetric constriction and defect-driven processes such as boundary detachment and internal rupture. We develop a hydrodynamic model incorporating elastic, viscous, and active stress contributions that distinguishes between stress-accumulation and stress-release phases and links variations in active stress to the observed intermittent dynamics. The model predicts distinct energy relaxation rates before and after detachment events, providing insight into stress dissipation. We compare experiments with numerical simulations, which reproduce the observed behavior and reveal how internal energy is generated and dissipated during stress buildup and relaxation. Together, our results demonstrate how boundary conditions and spatial heterogeneity govern the mechanical behavior of contractile active gels. These findings provide insight into stress regulation in cellular and tissue-scale systems and may inform the design of adaptive soft materials and bioinspired robotic systems.

Earthquakes propagating faster than the shear wave-speed are commonly thought to undergo a super-shear transition upon which they discontinuously jump from the sub-Rayleigh regime to the super-shear one. The super-Rayleigh regime, i.e., the range of propagation speeds between the Rayleigh and shear wave-speeds, is regarded as "forbidden" by the two-dimensional classical rupture theory. Here, we revisit the assumptions underlying the classical theory and develop a rupture theory that takes into account the dependence of the fault strength (frictional resistance) on the slip rate. The theory quantitatively agrees with numerical simulations nearly up to the Rayleigh wave-speed. Yet, very close to the latter, two-dimensional rupture solutions change their character due to frictional rate nonlinearity and rupture continuously propagates through the "forbidden" super-Rayleigh range into the super-shear regime, without a sharp super-shear transition. These results demonstrate that frictional rate dependence, generically observed in experiments, can have profound implications for fast earthquake propagation.

The temperature increase in the contact regions between solids in sliding contact can easily reach several hundred Kelvin and thereby dramatically affect friction and wear. Here I extend an earlier multiscale theory for the flash temperature (Ref. \cite{MP}) to the case of accelerated motion, and present numerical results illustrating the theory.

Meta-structures that display axial-twist coupling can be achieved through the emerging kinematics in Kresling origami patterns. A central challenge in these structures is understanding their nonlinear mechanical behaviour, specifically their equilibrium branches and bifurcation diagrams. This involves identifying relationships between desired responses and the geometric variables that define the design space, including the Kresling polygon count, initial twist angle, height, radius, and crease lengths. As the number of constituent units increases in an n-layer chain, we track complex equilibrium branches extending into the post-critical regime under successive instabilities, including branch-point bifurcations and limit-point instabilities. This work begins by establishing the relationship between the geometric design variables and the response curves of the assembled chain by modelling the crease lines as axial-load-carrying elements. Subsequently, equilibrium branches and instabilities are systematically investigated via continuation and bifurcation analysis, beginning with the single-layer system and progressively extending to two- and three-layer configurations. Finally, a generalisation strategy is proposed to extend these findings to an n-layer Kresling chain. This strategy enables the predictive construction of equilibrium paths and the inverse design of multi-layer meta-structures, using prescribed critical points to control post-critical behaviour. It provides a foundation for the inverse design and optimisation of architected mechanical metamaterials with programmable responses.

Nonreciprocal interactions in active solids violate action-reaction symmetry and produce a net response to strain. Assuming invariance under Euclidean symmetries, we derive a shape-space formulation for the elastohydrodynamics of nonreciprocal slender bodies that separates intrinsic deformation from rigid motion. The resulting nonlinear reaction-advection-diffusion system represents a geometric flow whose activity-driven instabilities generate steady, oscillatory, and chaotic patterns. These manifest as rigid, swimming, and chaotic motion, linking nonreciprocal elastohydrodynamics to geometric pattern formation and unifying recent observations in slender active structures.

Topological charges based on Berry phase play the fundamental role in the topological physics. However, such topological charges remain unexplored in quasicrystals, impeding the systematic understanding of topological states in such quasiperiodic systems. In this work, by deriving all the allowed topological charges according to group representation theory and the corresponding low-energy effective Hamiltonians, we establish a universal framework for Berry-phase-based topological charges in two-dimensional quasicrystals. Taking the $C_{8v}$ quasicrystal as an example, we demonstrate and characterize a higher topological charge of $C=4$, which is inaccessible in conventional periodic systems. Applying our framework to photonic quasicrystals, we uncover that the circling of photon momentum around the charge gives a $C$ times winding of the electromagnetic field distribution pattern. Such observable feature provides a direct experimental method to probe the topological charges. Our work paves the way for exploring topological charges in quasiperiodic matter, and fundamentally bridges periodic and quasiperiodic topological band theories.

The stress concentrations caused by inherent roughness of natural and manufactured surfaces often induce plastic deformation at contact interfaces, a challenge compounded by competing influences of the size effect of plastic deformation and self-affine rough surface topography. To address this, we developed a novel methodology based on stochastic theory using compounded Chapman-Kolmogorov equations, for the first time, to solve elastoplastic contact problems involving scale-dependent hardness. Our approach formulates three integral equations describing the evolution of probability density functions of elastic contact pressure, relative plastic contact area, and relative non-contact area across geometrical scales. We thoroughly investigate the effects of scale-dependent hardness on contact pressure distribution, relative elastic and plastic contact areas, and the area-to-load relationship. By adjusting various mechanical and material properties, our model predicts a smooth transition from linear elasticity to elastic-plastic behavior and finally to full plasticity. A key advancement is the derivation of a new topographic yield parameter incorporating a wider range of material and geometrical properties, aiding identification of contact status. Numerical solutions enable highly precise determination of elastic and plastic limits via curve-fitting, and we also provide a new diagram for rapid identification of contact status. This study pioneers a stochastic process framework for applying the compounded Chapman-Kolmogorov equation to rough surface contact analysis, and the integral equations characterizing how interfacial properties evolve with scale could offer valuable insights for other multidisciplinary fields where multiscale roughness is critical, such as earthquakes, electrical contact, and contact electrification.

Current-induced spin-orbit torque has emerged as a powerful technique for manipulating magnetization switching of ferromagnet/nonmagnet (FM/NM) based memory cell. By investigating nonequilibrium spin torque effect in a van der Waals heterobilayer, trigonal $\text{Cr}_{3}\text{Te}_{4}/\text{PtTe}_{2}$, the first-principles quantum transport calculations are applied to determine both local spin induction, resulting from Rashba-Edelstein effect in the FM layer, and spin current injection, flowing from the NM to the FM layer. Our work reveals that local spin induction significantly generates the fieldlike torque, which primarily governs the switching current in systems with strong in-plane magnetic anisotropy. Our work emphasizes the importance of optimizing spin Hall effect in the NM layer for perpendicular magnetic anisotropy (PMA)-based magnetization switching and maximizing the Rashba effect in the FM layer for in-plane magnetic anisotropy (IMA)-based switching.

Resolving the instantaneous, position-dependent isotropic magnetic exchange coupling $J_{ij}$ across a dynamically deforming crystal lattice requires a computational approach that simultaneously handles structural forces and magnetic interactions at length scales inaccessible to first-principles methods. Here we introduce DSpinGNN, a bifurcated machine-learning architecture comprising an $E(3)$-equivariant graph neural network (E-GNN) for classical Langevin structural dynamics and a physics-informed $Δ$-MLP that maps instantaneous local Cr-I-Cr bond geometry to isotropic exchange couplings, with the Goodenough-Kanamori superexchange relationship embedded as an analytical inductive bias. Trained on 345 DFT+U configurations of monolayer CrI$_3$ and evaluated on a strictly withheld 61-configuration test set, DSpinGNN simultaneously achieves an energy MAE of $1.1$ meV/atom, a force MAE of $6.5$ meV/Å, and an exchange coupling MAE of $0.18$ meV ($R^2 = 0.91$). Deployed at 400$\times$ scale in a 3,200-atom supercell under a collinear Ising-constrained adiabatic approximation at $5$ K, the model maps the local exchange response to a propagating biaxial strain wave. Wave reflection at periodic boundaries generates transient constructive interference regions where local compressive strain exceeds the DFT-established FM-to-AFM threshold, producing spatially heterogeneous exchange coupling textures that damp as the wave dissipates. Quantitative analysis yields a domain wall width of $ξ= 1.7 \pm 0.3$~nm and a constructive-interference oscillation period of $τ= 0.27$~ps -- mesoscopic observables inaccessible to direct DFT and constituting testable predictions for cryogenic magnetic force microscopy. DSpinGNN provides a reproducible, transferable framework for mesoscale exchange mapping in strain-driven 2D magnetic materials.

The position operator plays a central role in condensed-matter observables such as velocity, orbital moment, and electric polarization. In solid-state physics, the evaluation of operators incorporating the position operator has not reached a consensus, as observed in the operator-level discrepancy between the local circulation of Wannier functions and the self-rotation of wave packets. Here, to achieve a consistent evaluation of such operators, we propose three rules for evaluating operators involving the position operator in the Bloch representation. The rules are devised to satisfy physical conditions: independence from the choice of unit cell, preservation of Hermitian conjugacy for the product of operators, and recovery of the correct intraband velocity. We further address the gauge dependence of the position operator and introduce a scheme termed gauge filtration, which systematically removes gauge-dependent contributions from the operators containing the position operator. This methodology ensures that the quantities obtained from the operator evaluation correspond to observable physical phenomena. By applying our framework, we reconcile the results concerning the self-rotation of the wave packet and the local circulation of the Wannier function. We expect our proposal to establish a consistent framework for evaluating operators involving the position operator.

Soft glassy materials display rich and complex flow behaviors across both macroscopic and molecular scales, and a fundamental understanding of these phenomena remains an outstanding challenge. Here, we propose a theoretical model for the flow of supercooled liquids -- a typical class of glassy fluids -- based on a two-state paradigm that conceptualizes the flow as a dynamic coexistence of transient solid-like and liquid-like regions. The model rests on two essential physical ingredients: a correlation length that captures medium-range structural order, and a localized elasticity-mediated interaction that restricts stress propagation within solid-like regions. Remarkably, with all parameters determined solely from equilibrium state, the model quantitatively reproduces rheological responses -- including both steady-state and start-up shear -- for a broad range of shear rates. Furthermore, it simultaneously captures the evolution of molecular dynamic heterogeneity. This dual success -- spanning macroscopic rheology and microscopic spatiotemporal fluctuations -- underscores the pivotal role of structural and dynamic heterogeneities in governing the rheological response. Moreover, it provides a direct understanding of how the flow behaviors of a supercooled liquid are embedded in its equilibrium properties.

Participation entropy (PE) quantifies the spread of a many-body wavefunction across configuration space. While PE relaxes rapidly in generic chaotic systems, we show that $\mathrm{U}(1)$ conservation laws slow it down by imprinting with the slow hydrodynamic modes. Using a cluster expansion around equilibrium, we show that, after local density inhomogeneities decay, the leading PE deficit is dominated by squared connected density correlations. The long time relaxation is therefore controlled by diffusive correlation spreading, giving $ΔS(t)\sim t^{-1/2}$ in the hydrodynamic regime and crossing over to $\sim \exp[-O(t/L^2)]$ when $t\geq L^2$. We confirm this entropy correlation relation using exact computation and infinite system tensor network simulations in various quantum $\mathrm{U}(1)$ conserving circuits. Our results establish PE as a sensitive probe of hydrodynamic memory and suggest that slow relaxation is a generic consequence of conservation laws.

By forming hybrid light-matter states, optical cavities offer a route for engineering material properties, however, unambiguously probing the effects of light-matter coupling remains difficult. Here, we show that the polarization-resolved statistics of photons transmitted through a cavity, measurable via $g^{(2)}$, provide one such diagnostic. By relating $g^{(2)}$ to matter correlation functions such as the Raman structure factor, we link photon bunching and antibunching to material properties. By applying this method to the stripy-to-antiferromagnetic transition in the Kitaev-Heisenberg spin model, we find that polarization-dependent patterns of bunching and antibunching encode the magnetic point-group symmetries of each phase and characterize the behavior at the phase boundary. Finally, we predict measuring $g^{(2)}$ for output photon pairs polarized orthogonal to the input field will isolate higher-order light-matter scattering processes that probe higher-order material correlations.

The bulk photovoltaic effect has become a valuable probe of the quantum geometry of Bloch bands. While it is restricted to inversion-broken systems within the dipole approximation, the finite momentum of light is appreciated to give rise to this effect even in centrosymmetric crystals, an effect referred to as "photon drag." In this work, we recast the photon drag effect at leading order in the optical wavevector, highlighting a previously neglected contribution arising intuitively from the electric quadrupole correction to light-matter coupling. In the language of band geometry, we identify this interband quadrupole as a multiband metric tensor that quantifies the variation of two resonantly driven states extrinsic to the subspace they span. We predict that systems in which three or more bands strongly admix in momentum space, such as twisted MoTe$_2$ bilayers, will have anomalously large photon drag due to this quadrupolar effect. Our work provides a conceptual bridge between band-geometric organizing principles and electromagnetic multipole corrections in nonlinear optics.

Unitary time evolution of initially simple quantum many-body states rapidly generates entanglement and complex correlations, which limits direct numerical simulations. The late-time dynamics of physical observables, however, typically exhibits an effective simplicity in the form of hydrodynamics or kinetic theory. This leads to the question whether microscopic equations of motion can remain accurate and tractable up to long time scales by discarding irrelevant information in a controlled manner. Here, we introduce compressed minimum-purity time evolution (CoMPuTE) as an approach to keep track of a consistent set of reduced local density matrices, closing the hierarchical equations of motion using a minimum-purity principle. In benchmark applications we demonstrate (i) accurate description of energy diffusion in the one-dimensional mixed-field Ising model, (ii) the applicability to genuinely out-of-equilibrium Floquet dynamics starting from a pure state, and (iii) the limitations of the local reduced density matrix approximation when describing transport in the XXZ chain at $Δ=1$ that is governed by increasingly non-local integrals of motion. The CoMPuTE method enhances computational efficiency in comparison to the closely related local-information time evolution algorithm, opening a possible route towards an extension to systems in higher spatial dimensions.

We study stochastic interacting particle systems whose interaction structure is described by dense weighted directed graphs converging to a graphon. In the thermodynamic limit, we prove a law of large numbers for the empirical measure process and derive a deterministic nonlinear master equation describing the macroscopic evolution. The limiting equation retains the heterogeneous interaction structure of the microscopic system through the limiting graphon, allowing for spatially non-homogeneous behaviors such as localized or community-type interactions.

Positive longitudinal magnetoconductance (LMC) and planar Hall conductance (PHC) are hallmark transport signatures of the chiral anomaly in Weyl semimetals. Recent theoretical developments have extended Weyl fermions to multifold Fermionic systems with higher-pseudospin quasiparticle excitations, motivating the study of their magnetotransport properties. Here, we employ semiclassical Boltzmann transport theory within the relaxation-time approximation to investigate magnetotransport in pseudospin-1 Weyl semimetals, incorporating momentum-dependent scattering, orbital magnetic moment corrections, and charge-conservation constraints. To obtain a finite PHC, we break azimuthal symmetry through either a generic tilt of the quasiparticle dispersion or a finite misalignment between the electric and magnetic fields. In the untilted case, the PHC is positive and scales quadratically with magnetic field strength. Increasing the scattering strength induces a sign reversal of the PHC, producing a transition from positive to negative values. The PHC further exhibits the characteristic angular dependence $\sin 2γ$, where $γ$ is the angle between the magnetic field and the $x$-axis. Tilt qualitatively alters this behavior: tilt along the $x$- and $z$-directions changes the angular response to $\sinγ$ and $\cosγ$, respectively, generating strong anisotropy in the planar Hall signal. Moreover, the PHC shows a nonmonotonic dependence on tilt magnitude, revealing the interplay between tilt-induced symmetry breaking and chiral-anomaly-driven transport. Our results provide experimentally accessible signatures of multifold fermions and a framework for interpreting magnetotransport measurements in candidate materials of space groups 199, 214, and 220.

Ideal Bose-Einstein condensation (BEC) remains a paradigmatic example of a continuous phase transition and a cornerstone for understanding quantum degenerate bosonic matter. We demonstrate that critical behavior of the ideal Bose gas near the BEC phase transition falls into three distinct classes, determined exclusively by the low-energy scaling of the density of states. Depending on its scaling exponent, which is controlled by dimensionality and confinement, the transition displays either the usual algebraic divergences of thermodynamic susceptibilities, divergent behavior with marginal logarithmic corrections, or a more subtle form of criticality, where only the correlation length diverges. Our work provides a unified framework for criticality in noninteracting bosonic systems. This classification applies broadly to atomic, photonic, polaritonic, and magnonic condensates, where dimensionality, confinement, and spectral engineering can strongly reshape the density of states.

Exact results using dual-unitarity largely rely on nearest-neighbor structures, while finite-range interactions typically lead to complications. Going beyond the usual nearest-neighbor setting, we introduce an analytically tractable family of finite-range kicked Ising models that admit exact closed-form entanglement dynamics. The construction is based on a staggered structure in which dual-unitarity is present on sublattices that are then coupled to each other. The central observation is that these inter-sublattice couplings do not obstruct the dual-unitarity of the resulting model. For the minimal interaction range of $r= 2$, we derive exact expressions for all the $n-$Rényi entanglement entropies at all times and show that the result is the sum of the two coupled sublattice contributions. Our framework extends naturally to larger finite interaction ranges and to systems with heterogeneous local Hilbert spaces, without additional assumptions. It thus provides a controlled setting for studying exact entanglement growth beyond strictly nearest-neighbor dual-unitary models.

We investigate the nonequilibrium dynamics of a one-dimensional interacting system across a transition from a charge-density-wave (CDW) phase to a symmetry-protected topological (SPT) phase. Starting from a CDW initial state, we study both sudden quenches and slow ramps into the SPT regime. While the CDW order melts under both protocols, the fate of topological order is sharply different. Following a sudden quench, long-range SPT order does not emerge because the post-quench state contains a finite density of excitations above the topological ground state. In contrast, slow ramps allow the system to follow the instantaneous ground state away from the critical region, enabling the buildup of SPT order with deviations governed by Kibble-Zurek defect production. The dynamics is solvable via a unitary mapping to a quadratic fermionic Hamiltonian, allowing us to compute the Loschmidt echo, correlation functions, and string correlator. The Loschmidt rate function exhibits cusps signaling dynamical quantum phase transitions, while the correlation dynamics reveal the contrasting mechanisms governing quenches and ramps across the transition. These results demonstrate that entering the topological regime is not sufficient for the emergence of topological order; the decisive factor is the suppression of excitation production during the evolution.

We present a mechanism for ferromagnetism in narrow bands consisting of Anderson-localized states. We exploit single-particle localization to derive a controlled theory of exchange interactions within the narrow band. For quasiperiodic systems with a half-filled moiré band, we show that the critical interaction strength for ferromagnetism is highly sensitive to the geometry of real-space overlaps between localized orbitals: we find well-defined resonances at which ferromagnetism sets in for interaction energies that are far lower than the gap to other bands. Near these resonances, all the approximations in our theory are controlled, so our critical point predictions are quantitative. We show examples both in one and two dimensions. Our work identifies a route to ferromagnetism based on the geometry of real-space wavefunctions, distinct from previously found mechanisms based on the quantum geometry of Bloch bands.

Surface altermagnetism extends spin splitting beyond bulk altermagnets through symmetry reduction at surfaces and interfaces. An existing classification applies to the local symmetry of atomically flat surface terraces. The present paper addresses the symmetry of macroscopic spin-momentum correlations that survive averaging over compensated rough surfaces. These correlations are governed by the surface antisymmetry Laue point group. Rough-surface altermagnetism is forbidden at any surface of a magnet whose antisymmetry space group contains antitranslations, and the classification therefore reduces to $PT$-symmetric antiferromagnets. By restoring all symmetries leaving the surface normal invariant, roughness can generate compensated surface altermagnets from uncompensated flat terminations, increase the surface symmetry, or suppress spin splitting. By combining bulk switchability with altermagnetic surface transport properties, roughness-robust surface altermagnetism in $PT$-symmetric antiferromagnets provides a route toward spintronic functionality.

Electrochemical H2O2 synthesis was investigated using a 3 percent WO3 C gas diffusion electrode GDE. The catalyst outperformed bare Vulcan carbon, generating H2O2 concentrations of 423, 586, and 916 mg L at 50, 75, and 100 mA cm2, respectively, maintaining around 70 percent current efficiency. This performance and lower energy consumption are attributed to the WO3 carbon synergistic effect. In electro Fenton EF applications, the WO3 C GDE achieved rapid initial ciprofloxacin CIP degradation of around 70 percent in 30 min, though limited later by slow Fe2 regeneration. Incorporating UV light photoelectro Fenton overcame this constraint, yielding complete CIP removal within 90 min, while a boron doped diamond BDD anode enhanced total organic carbon TOC mineralization to 66 percent. The proposed degradation mechanism proceeds via hydroxyl radical attack on the piperazine ring with or without defluorination and quinolone ring oxidation, with theoretical analysis confirming reduced environmental toxicity of the transformation products. Overall, WO3 C GDEs represent a highly efficient strategy for H2O2 generation and wastewater remediation.

How can a growing biological filament remain straight despite stochastic fluctuations in growth? Motivated by filamentary structures that develop reproducibly across biological systems, we study the stability of a noisy, growing elastic filament regulated by feedback. We formulate a minimal model in which growth responds to the filament's strain, curvature, and orientation through local or nonlocal spatiotemporal feedback laws. Linear stability analysis identifies the conditions under which these feedback mechanisms stabilize a straight configuration. In the presence of noise, we show that purely local feedback requires orientation sensing to suppress long-wavelength instabilities, whereas nonlocal feedback allows stabilization through proprioceptive (curvature) sensing alone. Coupling to an elastic substrate further suppresses large-scale fluctuations. Our results establish minimal control strategies that ensure robust straight growth and suggest experimental signatures for identifying the feedback mechanisms underlying morphogenesis.

Reliable density-functional simulations require numerical settings whose residual errors are smaller than the chemical and materials trends being interpreted. In CP2K/Quickstep, this requirement is complicated by the joint use of atom-centered Gaussian basis sets and norm-conserving pseudopotentials: a code-to-code discrepancy usually contains both contributions. We present the UZH protocol, a closed-loop CP2K workflow that calibrates molecularly optimized Gaussian basis sets on small molecules, validates the resulting settings in unary-crystal equation-of-state benchmarks, identifies whether the limiting approximation is the Gaussian basis or the pseudopotential. The diagnosis is then used to revise the parameter files. The central diagnostic is a three-way comparison between production CP2K-GTH-UZH calculations, SIRIUS calculations using the same Goedecker--Teter--Hutter pseudopotential in a systematic plane-wave representation, and all-electron full-potential linearized augmented-plane-wave SIRIUS references. This construction decomposes the practical CP2K error into a Gaussian-basis component and a pseudopotential component. The protocol distinguishes basis-limited noble-gas and heavy-element cases from pseudopotential-limited transition-metal cases, guides targeted revisions with the CP2K basis and pseudopotential optimizers, and produces improved MOLOPT basis sets and GTH pseudopotentials as explicit outputs of the workflow. The UZH protocol is therefore constructive: it does not merely measure or reduce errors a posteriori, but allows turning verification outliers into validated CP2K parameter files for simulations across molecules and condensed phases.

The physics of polymer chains is often probed using molecular stretching experiments and various idealized single-chain models. The majority of these models consist of a discrete sequence of links, which may be treated as rigid or extensible. Although such models are well established and many specific extensible variants have been proposed, no generally applicable theory has been presented. Moreover, most existing treatments are heuristic rather than systematically and rigorously derived. This critical gap is closed here through the development of a generally applicable asymptotic theory for including link extensibility in a broad class of discrete models for single-chain thermodynamics. The theory is verified analytically using the freely jointed chain and validated numerically using the freely rotating chain. The resulting approximation is first-order accurate in inverse link stiffness, with quadratically decreasing error, and recovers extensible behavior across all link stiffnesses from a single rigid-link reference calculation.

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.

Single-file transport means no overtaking: particles move in a narrow channel while preserving their longitudinal order. This simple constraint has profound dynamical consequences, most notably tracer subdiffusion, and has made single-file transport a paradigmatic form of confined many-body motion, observed from molecular transport in zeolites to single-file diffusion of colloids in narrow channels. The standard view is that, once overtaking is suppressed, collective transport reduces to that of a strictly one-dimensional file. Here we show that this reduction fails in experimentally relevant finite-width channels: even when exchange is forbidden, the transverse equilibrium structure controls the transport laws and can qualitatively reshape them. Starting from the Brownian dynamics in the full confined geometry, we derive an exact large-scale one-dimensional fluctuating hydrodynamics for the longitudinal density, whose coefficients are fixed by the confined equilibrium equation of state. In the minimal hard-core setting, this yields a collective diffusivity that can become non-monotonic in density. This geometric anomaly propagates to exact large-scale predictions for integrated-current fluctuations, tracer displacement fluctuations and the associated density profiles. The effect is robust to the interaction potential, channel geometry, initial preparation and microscopic dynamics. Quasi-one-dimensional single-file transport therefore defines a distinct regime in which forbidding overtaking does not erase geometry from collective transport.

Force-induced dissociation of short double-stranded DNA (dsDNA) is central to single-molecule biophysics and DNA nanotechnology, yet a physically grounded kinetic description of shear-induced rupture for finite-length constructs remains lacking. Here we develop a master equation framework built on a force-dependent nucleation-zipper pathway with single-base transitions, enabling direct calculation of dissociation rates and transition state distances over a broad force range. Applied to a DNA-gold nanoparticle-DNA construct under constant shear force, the model accurately reproduces the experimental room-temperature data in the covered force regime and provides a unified interpretation of prior measurements on similarly sheared duplexes across all force regimes. A central result is that the three-dimensional helical geometry of dsDNA is essential for correctly defining the end to end distance under shear in the rod-like polymer model of short dsDNA. We further show that the extracted transition state distances are robust to variations in ssDNA polymer parameters within the experimentally relevant regime. Finally, we analyze the temperature dependence of the transition state distance and discuss how our framework captures globally-heated rupture while identifying the additional complications introduced by localized plasmonic heating in gold nanoparticle-coupled constructs. These results provide a predictive kinetic foundation for interpreting force-rupture experiments and for designing force- and temperature-actuated DNA nanostructures.

For groups without strict hierarchy, collective decisions often emerge through compromise. We develop a second-order network model of collective decision-making using a dissipative Hamiltonian formulation, in which informed agents introduce preferred directions while uninformed participants contribute only direction-free dissipation. We show that under low conflict, the model admits a locally unique, exponentially stable compromise state. Using a structured modular network we further show that as conflict increases the local compromise branch terminates through a saddle-node fold rather than through a smooth mean-field symmetry-breaking transition. Modular polarized states persist on branches that are locally separated from the compromise branch. Direction-free dissipation does not shift the static structural threshold, but it delays escape from the saddle-node ghost and pushes the observable onset of polarization to larger conflicts. Our work identifies a dissipation-mediated mechanism, complementary to connectivity-based accounts, through which uninformed participants stabilize collective behavior in biological and engineered swarms.

The Schramm-Loewner Evolution (SLE) describes a family of fractal curves that arise in the study of the scaling limits of many planar Statistical Physics models. These curves are modeled using the Loewner Differential Equation for the conformal maps $g_t(z)$ with a Brownian motion driver. Using Euler's Method, in the current work we performed numerical experiments to study at a fixed time the quantities $|g_t(z) - \overline{g_t(z)}|$ and $Re(g_t(z)) - Re(\overline{g_t(z)})$, where $Re$ denotes the real part and $\overline{g_t(z)}$ refers to the sample average. These random variables measure the 'spread' of the dynamics from the average behavior at fixed time. One of the scopes of this work is to give numerical predictions for future theoretical investigations on these quantities. When investigating these quantities in the SLE case our experiments predict that the distribution is bimodal when the dynamics started close to the origin, and it can become bell-shaped if the dynamics is started further from the origin. In the second part, we performed experiments for a Multiple SLE model whose driver is Dyson Brownian Motion. Due to singularity in the dynamics of the drivers and the many data points needed, this part is challenging from a computational perspective. In the multiple SLE case, our experiments predict that the distribution is bell-shaped in all cases. In addition, we check the changes in the distributions as we vary the parameter $κ$ in the SLE case and $β$ in the Multiple SLE case.

We investigate energy supratransmission in a boundary-driven two-dimensional nonlinear Schrödinger equation with a central hole. Harmonic forcing with azimuthal modulation generates standing-wave states whose existence and stability depend on the driving amplitude, the inner radius, and the imposed azimuthal charge. Bifurcation analysis shows that small inner radii produce strongly confined states with higher destabilization thresholds, whereas larger radii yield broader profiles and smoother transitions between stable and unstable branches. The cubic--quintic and saturable models exhibit similar qualitative behaviour but differ quantitatively in their critical amplitudes and parameter dependence. A variational approximation captures the dependence of the critical drive on the azimuthal charge and nonlinear parameters, and clarifies how the nonlinear response shapes the stationary states near the turning point. Time-dependent simulations show that supratransmission occurs through the emission of localized pulses, with nonzero azimuthal charge triggering symmetry breaking and producing two-dimensional localized excitations (pearls). Isosurface plots provide a complementary view of the resulting radial and angular excursions. These results establish a quantitative framework for supratransmission in two-dimensional geometries and are relevant to driven nonlinear systems in optics, Bose--Einstein condensates, and structured media.

We analyze five-neighbor particle cellular automata whose conventional two-dimensional fundamental diagrams are multivalued, but whose mean flow is uniquely determined by introducing a second density. We first consider binary rules for which the second density is conserved, and then examine rules for which the second density is not conserved but converges asymptotically. These examples give three-dimensional fundamental diagrams in which the mean flow is determined by the particle density and the second density. We then investigate whether this single-valued structure is preserved under real-valued max-plus extensions. There are some rules where two different max-plus extensions are introduced, and numerical simulations show that both extensions preserve the same single-valued three-dimensional fundamental diagram. These observations imply that, in constructing real-valued max-plus extensions, it is important to choose the flux function and the second density consistently.

We present a one-month-ahead conditional probabilistic framework for wind-power forecasting at ten-minute resolution. Monthly Weibull shape and scale parameters are estimated from serially dependent SCADA wind-speed data, corrected through a Godambe covariance, and forecast by a heteroskedastic Kalman filter on a bivariate VAR(1) state-space model. Conditional on the MMSE forecasted Weibull invariant law, we construct and compare three positive wind-speed SDE models: an Ornstein-Uhlenbeck-Weibull transform, a Fokker-Planck drift-first diffusion, and a Fokker-Planck diffusion-first model. The simulated wind-speed ensembles are mapped to power through a calibrated XGBoost power curve. Applied to January 2021 data from a Senvion MM92 turbine at Kelmarsh Wind Farm, the three SDE formulations are statistically indistinguishable in probabilistic accuracy, with mean CRPS values between 1.569 and 1.575 m/s. The diffusion-first model is therefore preferred on computational grounds, reducing runtime by about a factor of seven relative to the OU-Weibull model. In the power domain, the Wasserstein distance between simulated and observed distributions is 26.1-27.6 kW, below $1.4\%$ of rated capacity, while the monthly energy-yield bias is about $-7.3\%$ for the examined month. Exceedance-probability errors remain below 1.6 percentage points over the 0-1500 kW range and about 2.2 percentage points near rated power. These quantities provide decision-relevant probabilistic inputs for downstream operational problems, rather than completed reserve, storage, market, or fatigue-optimization decisions. Full marginalisation over the Kalman predictive law of the Weibull parameters is left as a natural extension.

Elastically accommodated grain-boundary sliding (EAGBS) is a plausible source of upper-mantle seismic attenuation and dispersion, yet classical theory predicts a localized Debye-like peak that is absent or only weakly expressed in dry olivine experiments. Here we test whether microstructural heterogeneity can explain this discrepancy using 2-D finite-element simulations on periodic Voronoi tessellations. We find that irregular grain geometry changes the baseline EAGBS response relative to the regular hexagonal benchmark, but increasing grain-size variance alone produces only modest changes in modulus and peak height, with little spectral broadening. In contrast, a broad distribution of grain-boundary viscosities progressively suppresses and broadens the Debye-like loss peak into a weak background spanning a wide frequency interval. This broadening arises from the superposition of many localized relaxation processes with distinct characteristic timescales and motivates a reduced-order 0-D description of the aggregate response. These results suggest that the absence of a pronounced EAGBS peak in dry olivine does not necessarily imply the absence of EAGBS mechanism itself. If grain boundaries sample a sufficiently broad viscosity distribution, the macroscopic EAGBS contribution may appear experimentally only as part of a broad attenuation background, while still remaining relevant for upper-mantle seismic attenuation and velocity dispersion.

The combinatorial design space of multilayer perovskite solar cells is vast, yet exhaustive experimental or computational screening of all possible material combinations remains impractical. Here, we integrate SCAPS-1D device simulations with machine learning to systematically explore 125 device architectures constructed from five electron transport layers (ETL), five absorbers (including lead-free double perovskites), and five hole transport layers (HTL). A representative subset of configurations is used to train a machine learning (ML) model, which predicts the power conversion efficiency (PCE) of the remaining unexplored structures. Leave-One-Group-Out cross-validation yields a Spearman rank correlation, demonstrating reliable ranking capability. SHAP (SHapley Additive exPlanations) analysis reveals that the HTL band gap, absorber band gap, and ETL electron affinity are the most influential descriptors, providing physical insights into interfacial recombination and charge extraction. The machine learning model identifies several high-performance configurations that are subsequently verified by full SCAPS-1D simulations. Among them, the device FTO/TiO$_2$/Cs$_2$AgBiBr$_6$/NiO/Ag achieves a PCE of 28.85%, and the ML-suggested structure FTO/SnO$_2$/Cs$_2$AgInBr$_6$/NiO/Ag exhibits 28.62%, outperforming a closely related literature architecture by approximately 4% absolute. Notably, eight of the top-11 structures employ the lead-free double perovskite Cs$_2$AgInBr$_6$. This work demonstrates that a physics-based, data-driven workflow combining SCAPS-1D, ML, and SHAP can accelerate the discovery of high-efficiency, environmentally friendly perovskite solar cells while providing transparent design rules. The approach is generalizable to other multilayer optoelectronic systems.

Mirrors that preserve the handedness of optical radiation upon reflection are an essential building block for the design of numerous resonant nanophotonic structures with capabilities for enantiomeric discrimination. Ordinary metallic and Bragg dielectric mirrors are not suitable in these context since they flip handedness of electromagnetic field upon reflection around normal incidence. While there has been considerable progress in the development of such reflecting structures, this research area remains largely unexplored. Here, we present a detailed numerical and theoretical analysis of the potential of low-symmetry periodic lattices composed of high-symmetry non-chiral meta-atoms for resonant reflection with handedness preservation (HP). Using full-wave numerical simulations, we analyze a family of rhombic and monoclinic (oblique) lattices of circular dielectric disks and/or holes, and in each identify the regime of near-perfect HP reflection. We study the robustness of these structures to geometric deviations, material losses, and incidence angle. Finally, we describe the resonant HP response of these structures using the coupled-mode theory formalism.

This Letter presents a novel analytical framework showing that the dark count probability (PDC) of perimeter-gated single-photon avalanche diodes (pg-SPADs) follows a complementary Gompertz function. Specifically, we show that PDC follows a complementary Gompertz form from which we derive a pixel-specific descriptor, the midpoint perimeter gate voltage, which characterizes a pixel's equiprobable operating point. We further show that a perimeter gate voltage compensation rate may be obtained from this descriptor to offset temperature-induced changes in the pixel's activation function. The proposed framework is experimentally validated using 4,096 pg-SPADs arranged in a 64 x 64 array and manufactured in a 0.35 $μ$m CMOS process. The devices were characterized at temperatures ranging from -5 $^o$C to 55 $^o$C and perimeter gate voltage magnitudes of 0 to 5 V. The measured results demonstrate deterministic bias control of dark count probability across process and temperature variations.

A fundamental concept in linear algebra and its applications to network analysis is the Perron--Frobenius (PF) theorem, which underpins eigenvector-based centrality measures such as eigenvector centrality, PageRank, and hubs and authorities. By invoking the PF theorem, we know for strongly connected networks with positive edge weights that the eigenvector corresponding to the largest eigenvalue of the weight matrix yields a well-defined centrality measure (namely, eigenvector centrality). Traditional formulations of the PF theorem and associated centrality measures assume that networks have real-valued weights. However, many networks in areas such as quantum information, quantum chemistry, electrodynamics, and machine learning have complex-valued edge weights. In this paper, we study generalizations of the PF theorem to complex-valued matrices, establish connections between these generalizations, and propose generalized eigenvector-based centrality measures to analyzing node importances in networks with complex edge weights. We also prove results about the existence of complex-weighted networks that satisfy generalized PF properties and calculate associated centrality measures for several examples, which we draw from application areas such as electron transport, circuit analysis, mathematical chemistry, and communication networks.

Human behaviors, particularly non-pharmaceutical interventions (NPIs), are dynamically coupled with epidemic spreading. While prior studies mainly assume a linear interplay, real-world behavioral evolution is driven by nonlinear responses and social influence. Here, we incorporate these multifaceted mechanisms into a co-evolutionary model and analytically derive the critical thresholds. Notably, as the infection rate grows, NPI compliance initially rises but then abruptly drops to zero. This paradoxical decline indicates an emergent social dilemma: at high infection rates, abandoning NPIs is individually optimal but detrimentally triggers an explosive surge in epidemic prevalence. We further show that socially induced overestimation of the infection rate can counterintuitively prompt individuals to abandon NPIs. Moreover, the interplay with social influence induces periodic oscillations, reflecting a tragic cycle of recurrent epidemic waves. Furthermore, we validate the robustness of this NPI-abandonment dilemma in networked population. Our work illustrates rich emergent phenomena in the co-evolution of epidemic and behavior, challenging traditional views on this coupled dynamics.

AlScN wurtzite ferroelectrics are promising candidates for energy-efficient non-volatile memory. However, AlScN suffers from a high coercive field and reduced cycling endurance, and the limited tunability of its properties constrains further optimization. Co-doping AlScN with boron offers the promise of independently tailoring the chemical and structural properties, making AlScBN an attractive quaternary system. This material has already been explored for a few selected compositions, however, no systematic study of the full AlScBN compositional space exists. A combinatorial approach consisting of gradient deposition with HiPIMS at low temperatures of 250°C and automatic analysis of film properties allowed us to analyze a total of 850 unique samples within the AlScBN phase space. In addition to a full screening of the materials' chemical and structural properties, we fabricate and characterize combinatorial device libraries. XPS charge transfer analysis experimentally confirms that bond ionicity correlates with a reduction in the coercive field for AlScN and AlScBN systems, opposite trends are instead observed for AlBN. While the films maintain a high remanent polarization of 130-150 μC/cm2, Sc and B co-doping reduces the coercive field from 7 MV/cm to 3 MV/cm. Notably, B co-alloying lowers the amount of Sc needed to lower the coercive field, reducing reliance on this scarce element. In addition, we find that co-alloying with B, notably improves cycling endurance, which is related to a reduction in defect density. These results establish AlScBN as a scalable, CMOS-compatible ferroelectric, positioning it as an interesting alternative to AlScN.

Molecular rare-earth-ion complexes offer unique opportunities for quantum technologies by combining the intrinsic coherence properties of rare-earth ions with chemically tunable molecular environments. A crucial capability is the realization of multi-qubit architectures with defined qubit couplings to enable two-qubit quantum gates. Here, we investigate the optical coherence properties and excitation-induced interactions of two Eu$^{3+}$-based molecular complexes, comparing a mononuclear reference system with a dinuclear analogue in which two Eu$^{3+}$ ions are positioned at a well-defined intramolecular distance of about 7 Angstrom. Using cryogenic ensemble spectroscopy, including spectral hole burning, free-induction decay, and photon echo measurements at temperatures down to 100 mK, we demonstrate long optical coherence times $T_{2,\text{o}}$ of up to 9 $μ$s. As a key step toward scalable multi-qubit architectures, a control-target sequence was implemented to probe conditional ion-ion interactions, revealing a stronger interaction-induced dephasing in the dinuclear complex. Finally, we show the integration of the dinuclear complex into a fiber-based optical microcavity, and observe an 380-fold emission enhancement of the $\mathrm{}^5\mathrm{D}_0\rightarrow\mathrm{}^7\mathrm{F}_0$ transition. Together, these results position molecular rare-earth complexes as versatile and chemically tunable building blocks for scalable quantum technologies.

The effect of using a graphene layer as a semitransparent contact layer is studied in solar cells based on AlInN on Si (100) substrates. The devices consist of AlxIn1-xN layers deposited on p-type Si (100) substrates incorporating a thin amorphous silicon (a-Si) buffer layer to improve the heterointerface quality. Three aluminum contents are studied, namely: x=0.22, 0.35 and 0.43. Subsequently, a monolayer graphene film was transferred onto the front surface of the devices using a simple and low-temperature transfer process, acting as a semitransparent conductive contact. The photovoltaic characteristics were then evaluated under illumination and dark conditions in devices with and without the graphene layer. The results show that the incorporation of graphene leads to a clear improvement in the short-circuit current density, fill factor, and overall power conversion efficiency for all studied compositions, while the open-circuit voltage remains largely unaffected. These findings demonstrate the potential of graphene as an effective transparent conductive contact for nitride-based solar cells.

The native processing of time-dependent signals from optical sensors by integrated photonic circuits can potentially bring significant advantages in terms of energy consumption, latency and processing power, as it allows skipping or reducing the use of fast digital electronics and directly exploiting optical degrees of freedom and parallelism. However, due to a short memory, optical operations usually struggle to directly process optical signals with relatively slow (<MHz) dynamics from optical sensors. In this work, we experimentally show that these limitations can be overcome by exploiting the self-pulsing dynamics in a microring resonator (MRR) network. In particular, we demonstrate that such dynamics can expand and retain information about perturbations sensed by a fiber sensor. This reduces the minimum sampling rate for the digitization of the sensor signal by at least one order of magnitude. The reduction is achieved by combining fiber sensing measurements at two different perturbation locations and frequencies with MRR network measurements at multiple output ports, input power levels and laser wavelengths. This work represents a first step in bridging time-dependent optical processing and optical sensing at sub-μs time scales.

Frequency conversion of ultrafast lasers is fundamentally constrained by the trade-off between pulse energy and repetition rate, limiting access to regimes critical for fundamental science and industry. Here, we demonstrate a frequency-conversion mechanism in which molecular coherence accumulates across ultrafast pulse trains in gas-filled hollow-core fibers. Unlike conventional nonlinear interactions initiated by individual high-energy pulses, this mechanism relies on the collective buildup of coherent molecular oscillations driven by successive pulses. Using this mechanism, we achieve Raman frequency conversion at repetition rates up to 3 GHz with nanojoule pulse energies. The results establish a regime of nonlinear optical interaction governed by coherence accumulation of gas molecular oscillations with broad implications for ultrafast laser science and frequency conversion technologies.

High-frequency focused acoustic beams are promising for selective trapping of cells in fluids, but the related acoustic absorption may generate large acoustothermal effect which could cause thermal heating on cells or microparticles and bring extra acoustic body force due to the thermal gradient. The theory of the bulk acoustic streaming and acoustic radiation force in a focused-beam for the three-dimensional selective trapping of a cell has been developed [Li and Gong, Phys. Rev. Fluids, 11, 054201 (2026)], however, the acoustothermal effect and its feedback on the acoustic field at high frequency with strong absorption remain weakly understood. To solve this issue, we develop a theoretical and numerical model that couples acoustic propagation, bulk acoustic streaming, and acoustothermal effect in water. The acoustic body force is decomposed into a viscous-attenuation-induced acoustic body force $\mathbf{f}_{\mathrm{E}}$ and a temperature-gradient-induced acoustic body force $\mathbf{f}_{\mathrm{T}}$, while the temperature field is fed back to the frequency-domain acoustic calculation through the temperature-dependent material properties. Taking the single focused beam for example, within the pressure range constrained by the mechanical index, $\mathbf{f}_{\mathrm{T}}$ remains weaker than $\mathbf{f}_{\mathrm{E}}$, whereas streaming-induced convection can markedly reduce the temperature rise when the thermal Peclet number ($Pe_T$) exceeds unity. This work establishes a theoretical basis for predicting and controlling the intercoupling of bulk acoustic streaming and acoustothermal effec of high-frequency focused beams which will be helpful for the design of single-beam acoustical tweezers.

We investigate polarization-selective absorption in a visible-wavelength heterostructure consisting of a ReS$_2$ stripe grating, an $α$-MoO$_3$ spacer, and an Au back-reflector using finite-difference time-domain simulations. For an optimized geometry with a grating period of 500 nm, stripe width of 250 nm, and ReS$_2$ thickness of 80 nm, the structure exhibits near-unity absorption of 99.99\% at 650.5 nm under TE-polarized illumination. The resonant field is concentrated near the outer edges of the ReS$_2$ stripe, while absorption power density is localized in the same region, consistent with a localized edge mode. The absorption response depends strongly on polarization, producing a TE--TM resonance separation of 16.2 nm. Replacing either the biaxial ReS$_2$ layer or the anisotropic $α$-MoO$_3$ spacer with isotropic equivalents substantially modifies the spectral response and reduces the polarization-dependent wavelength separation. In addition, rotating the crystal orientation of the ReS$_2$/$α$-MoO$_3$ stack shifts both the resonance wavelength and peak absorption without changing the device geometry. The results show that the combination of anisotropic resonator and spacer layers provides an effective means of controlling resonant absorption and polarization selectivity in van der Waals photonic structures.

How should one design efficient chemically open optical cavities for molecular strong coupling? Addressing this question is important for the development of soft-cavity platforms for dynamically tunable light--matter interactions, where direct access to confined electromagnetic modes is essential. Conventional cavity figures of merit such as $Q/\sqrt{V}$ and cooperativity successfully describe spectral confinement and dissipation but do not fully capture the role of linewidth asymmetry between cavity and molecular degrees of freedom. Here, we systematically investigate strong coupling between TDBC dye molecules and whispering gallery modes of polystyrene microspheres by varying the microsphere radius over a broad range. To quantify the robustness of strong coupling, we define the parameter $χ= \frac{g}{\max(κ,γ)}$, where $g$ is the coupling strength, while $κ$ and $γ$ denote the cavity and molecular linewidths, respectively. Although the coupling strength decreases monotonically with increasing cavity size due to mode-volume scaling, we find that $χ$ exhibits a pronounced maximum near the condition $κ\approx γ$. This observation suggests that linewidth matching is not merely a criterion for improved spectral visibility, but reflects a dissipation-matching condition that optimizes the robustness of coherent light--matter exchange in soft-cavities. Our results provide an alternative framework for designing morphology-dependent cavities for molecular strong coupling.

Artificial intelligence is driving intense interest in alternative computing hardware capable of neural information processing beyond conventional charge-based electronics. Among emerging approaches, wave-based computing promises highly parallel and energy-efficient operation, but scalable physical neural hardware has remained elusive because wave systems generally lack cascadable nonlinear neurons with signal regeneration and phase-robust operation. Here we demonstrate integrated magnonic neural circuits based on nonlinear threshold neurons realized in nanoscale yttrium iron garnet waveguides. The neurons perform weighted summation of multiple spin-wave inputs, while a pump-controlled nonlinear activation defines continuously tunable firing thresholds. Owing to deeply nonlinear spin-wave dynamics, the activated neurons emit self-normalized outputs whose intensities are largely independent of the input amplitudes, while nonlinear phase self-adjustment suppresses sensitivity to the relative input phases, enabling deterministic neuron-to-neuron cascading without external signal restoration. We experimentally realize programmable threshold neurons, reconfigurable weighted classification and deterministic cascading between sequential neuronal stages, and further demonstrate reconfigurable physical pattern recognition in a seven-neuron integrated magnonic circuit through experimental classification of the binary letter patterns 'HUST'. These results establish nonlinear magnons as a scalable platform for integrated neural hardware and position nonlinear wave dynamics as a general paradigm for physical neuromorphic computing.

Similar to the behavior of elementary particles, such as photons and electrons, the interference of phonon waves in artificial periodic nanostructures coherently modulates phonon band structures, serving as the foundation for phonon band engineering. However, direct observation of such coherently modulated phonon band structures remains challenging despite substantial insights from existing literature. Here, utilizing high-resolution inelastic X-ray scattering, we observed coherently modulated phonon band structures with phononic band gaps in a short-period GaAs/AlAs superlattice at 300 K and 500 K. Our findings provide the first direct evidence of phonon coherence at and above room temperatures, signifying a major advancement in the artificial engineering of phonon band structures. Furthermore, our experimental observations and ab initio lattice dynamics revealed that the coherently modulated phonon band structure enhances three-phonon scattering channels, strengthening high-order anharmonic effects such as three-phonon scattering and optical phonon softening. Our observations demonstrate the robustness of phonon coherence at high temperatures, and opens new routes for engineering phonon band structure and high-order phonon-phonon scattering by employing a flexible, bottom-up nanostructuring approach, with extensive applications in phononic metamaterials, microelectronics, and thermoelectrics.

On a charged interface, broken inversion symmetry permits a large field-linear Pockels response through $χ^{(2)}(ω;ω,0)$; at the water/ITO interface $|r_{13}|$ has been reported to reach the $10^{2}\,\mathrm{pm/V}$ order. The coexisting third-order DC Kerr term $χ^{(3)}(ω;ω,0,0)$ -- small in bulk water ($|χ^{(3)}_{\mathrm{bulk}}|\sim5.5\times10^{-21}\,\mathrm{m^2/V^2}$) -- had not been jointly parameterized with the Pockels term along the fundamental-frequency ($ω$) electro-optic path. Superimposing an AC modulation and a DC bias in 0.1\,M NaCl mixes the $χ^{(3)}$ contribution with the 1f response through the cross-term $2 s_{1133}\,E_{\mathrm{DC}}$, so that the water refractive-index modulation $Δn_{\mathrm{water}}$ varies linearly with $V_{\mathrm{WE}}$; a model-assisted linear fit then determines both terms from a single AC$+$DC sweep. At $V_{\mathrm{WE}}=0\,\mathrm{V}$ vs Ag/AgCl, $|r_{13}|=(1.18 \pm 0.06_{\mathrm{PZC}})\times10^{2}\,\mathrm{pm/V}$ and, under $η_{\mathrm{DC}}=1$, the thickness-normalized DC Kerr coefficient $|s_{1133}/d_{\mathrm{EDL}}|=33.0 \pm 5.6\,\mathrm{pm/V^{2}}$. Across physically reasonable $d_{\mathrm{EDL}}$ ($0.6$--$1.6\,\mathrm{nm}$), the interfacial DC Kerr susceptibility reaches $|χ^{(3),\mathrm{int}}_{1133}| \approx (2\text{--}5.5)\times10^{-20}\,\mathrm{m^2/V^2}$, several-fold above the visible-range bulk-water value. This response is a property of the specific interface, tunable through the choice of electrode, electrolyte, and solvent rather than intrinsic to bulk water. Amid renewed interest in the Kerr response of water (including recent THz-band optical Kerr studies), the method directly probes this DC Kerr term along the $ω$ path and complements SHG/SFG (the $2ω$ path).

The coupling theory of quasinormal modes (QNMs) for a coupled system of generally lossy and dispersive optical nanoresonators has been established in a rigorous manner based on the first principle of Maxwell's equations [Phys. Rev. B 102, 045430 (2020)], and can achieve superior computational efficiency and physical intuitiveness compared with full-wave numerical methods if a small set of modes can achieve a high accuracy. The QNMs suffer from an exponential divergence of far field and can form a complete basis inside but not outside the resonator. In the QNM coupling theory (QCT), it is required that the QNMs of each resonator form a complete basis in expanding the scattered field both inside and outside the resonator, which can be achieved by using regularized QNMs (RQNMs). However, a strict numerical demonstration of the impact of the mode completeness of RQNMs on the accuracy of QCT by using a virtually complete basis of RQNMs is still absent. In this paper, we will provide such a numerical demonstration along with an improvement of the QCT and some theoretical demonstrations on a rigorous incorporation of RQNMs into the QCT. The RQNMs are obtained by introducing an equivalent surface current (ESC) encircling the resonator (called ESC-RQNMs) or the perfectly matched layer (PML) surrounding the computational domain (called PML-RQNMs). The numerical example is selected as two one-dimensional resonators of slabs in the extreme coupling case of direct contact, for which a virtually complete basis of RQNMs can be solved either analytically (for ESC-RQNMs) or numerically (for PML-RQNMs). The results show that by using a virtually complete basis of RQNMs, the QCT can achieve a high accuracy in predicting both the source-free eigenmodes and the source-excited scattered field of the coupled system, which is not true if using the incomplete basis of not-regularized QNMs (i.e., physical QNMs).

We present the exact analytical paraxial propagation of structured light beams that transition from Ring annular profiles to top-hat intensity distributions. The initial field is defined as a superposition of a Gaussian-weighted power-law core and a singular inverse-quadratic modulation term, both carrying an azimuthal phase factor. By solving the Fresnel diffraction integral in cylindrical coordinates, we obtain exact closed-form expressions for the propagated field at arbitrary planes. The paraxial evolution is shown to be governed by a Cauchy-Riemann beam term and an infinite series of modified Bessel functions of the second kind weighted by an azimuthal phase factor. This analytical framework demonstrates how tuning the source parameters enables a continuous transition from ring-dominated annular profiles to uniform top-hat beams. For the fundamental mode ($l=0$), the singular component fills the central intensity null, producing a flat transverse plateau.

A time-reversal characteristic-mode decomposition is developed for reciprocal lossy electromagnetic structures. The formulation is built on a transmit--receive interpretation of reciprocity: the far-field pattern radiated by a mode determines the time-reversed incident field that is optimally matched to couple energy back into that same mode. This physical picture leads to an antilinear characteristic-mode equation whose solutions remain radiation-power orthogonal even in the presence of material loss, lossy loading, or matched absorption. As a result, the modal expansion coefficients directly represent the radiated-power contributions of the corresponding modes and avoid the singular biorthogonal normalization that may arise in nonnormal classical characteristic-mode expansions. Equivalent formulations are derived in the scattering-operator, T-matrix, and method-of-moments (MoM) frameworks, thereby connecting external wave-channel descriptions with current-space and port-excitation descriptions. The proposed modes reduce to classical characteristic modes in the lossless limit. Numerical examples involving a lossy two-sphere system and a loaded folded antenna demonstrate the radiation-power orthogonality, modal-expansion stability, and power interpretability of the proposed decomposition near exceptional points, where classical characteristic-mode expansions become singular or lose their radiated-power meaning.

Recent breakthrough studies of nonlinearities at extreme pump intensities ($\sim$1 $\text{TW/cm}^2$) in transparent conducting oxides (TCOs) have rewritten our understanding of the dynamics in these materials. However, exploring TCO dynamics beyond these intensities is prohibited by the damage threshold of the material. In this work, we overcome this problem by using a few-cycle pump laser pulse (sub-8\,fs) to maximize the intensity while keeping the optical fluence below the damage threshold. We observe a reversal in the optical response trend starting at optical pump laser intensities of $\sim$5 $\text{TW/cm}^2$ similar to Segal et al. At the highest pump pulse intensities, we obtain a complete change in the sign of the modulation for both transmission and reflection, producing a full-cycle oscillation of the refractive index modulation within 300\,fs. The amplitude of the sign reversal scales quadratically with the intensity. We therefore propose a simple two-photon absorption (TPA) model to explain the observed behaviour. The TPA, which is normally forbidden by the Pauli blocking, is enabled here by intraband excitations from the lower to the upper non-equilibrium states of the conduction band (CB). Such excitations vacate the states at the bottom of the CB, lifting up the blocking and thus making interband TPA possible. The model is in good agreement with experimental results, capturing the essential trends in the observed data and revealing the dynamics of competing channels caused by the interplay between interband and intraband transitions. This intensity-controlled mechanism could be the key to unlocking new applications of TCOs for time-varying photonics such as photonic time crystals.

Topological materials provide a new tool to direct wave energy with unprecedented precision and robustness. Three elastic topological phases, the valley Hall, Chern and spin Hall insulators, are currently studied, and they are achieved separately in rather distinct configurations. Here, we explore analytically various topological phase transitions for in-plane elastic wave in a unified mass-spring honeycomb lattice. It is demonstrated that the three elastic topological phases can be realized in this single lattice by designing mass, stiffness or introducing Coriolis' effect. In particular, the interface between valley Hall and Chern insulators is found to support topological interface mode for the first time. Perturbation method is used to derive the analytic effective continuum model in the neighbor of band degeneracy, and the physics in topological phase transitions are revealed through evaluation of topological invariants. The topologically protected interface states, their decaying profile as well as the pseudo-spin-indicating polarization specific for elastic wave are systematically analyzed, and these results are further confirmed numerically by Bloch wave analysis of domain wall strip and transient simulation of finite sized sample. This study offers a concise and unified analytical model to explore topology nature of elastic wave, and can provide intuitive guidance to design of continuum mechanical topological materials.

Zero modes, which are deformations that cost zero energy, underlie many exotic behaviors in elastic metamaterials. While classical linear Cauchy elasticity explains many of these modes, those linked to the rotations of metamaterial inner components often lie beyond its scope. Micropolar elasticity, which incorporates translation and rotation degrees of freedom, provides a framework for capturing these rotational modes. Herein, we present the first complete symmetry-based classification of zero modes in two-dimensional micropolar solids, with an emphasis on rotation-related modes. Guided by this classification, we construct threefold rotationally symmetric micropolar metamaterials and realize typical rotational micropolar zero modes. We further show that these metamaterials exhibit wave phenomena forbidden in Cauchy continua, including the emergence of three bulk waves in the long-wavelength limit and associated triple refraction, chiral acoustic modes, as well as strong wave anisotropy. All intriguing properties are quantitatively captured by micropolar continuum descriptions, whereas the classical Cauchy continuum theory fails to predict these behaviors, even at a qualitative level. Our results establish a general framework for engineering rotation-based zero modes, opening avenues for designing metamaterials with novel wave properties.

Charge-exchange (CEX) collisions can affect measurements of plasma plumes and neutralized ion flows in vacuum facilities, particularly when the background gas pressure increases and the CEX mean free path becomes comparable to the characteristic plume or facility dimension. Here, we investigate that regime in the plasma plume of a gridded ion source operating with a 400 eV argon ion beam. The fast-ion flux and low-energy ion flux were measured using a retarding potential analyzer (RPA) and planar probes, while the fast-neutral flux was inferred from deposited-power measurements with a thermal flux probe using a power-balance analysis. The low-energy ion flux increases with both background gas pressure and axial distance and its detection also depends on probe geometry. After the fast-ion component is isolated, its attenuation is described more accurately by an analytical reduced semi-empirical quasi-2D model that includes charge exchange and the experimentally observed plume divergence than by a one-dimensional attenuation law. The inferred fast-neutral flux also increases with pressure; however, the model underpredicts it at small axial distance and overpredicts it at elevated pressure and larger axial distance. This discrepancy suggests additional angular and collisional effects, as well as possible fast-neutral production near or inside the ion source, that are not captured by the present model. These results show that background gas pressure affects both the plasma plume and the diagnostic response, and that complementary electrostatic, thermal, and energy-selective diagnostics are required to distinguish source behavior from facility-induced effects.

High damage threshold gratings are in demand worldwide as critical components for next generation ultrahigh intensity lasers. Here we investigate the pulse-duration dependence of ultrafast laser-induced damage thresholds (LIDT) in hybrid multilayer dielectric gratings, touted to combine superior performance properties of both metallic and multilayer dielectric (MLD) gratings, using a dynamic finite-difference time-domain model incorporated with linear and non-linear absorption models. Simulations agree with reported experimental LIDT values for three representative designs and predict scaling exponents which vary with pulse durations ranging from 10 to 500 fs. The results reveal strong dependence on both material bandgap and grating field distribution, providing guidance for designing high LIDT gratings.

Quantifying the coherence properties of complex optical fields is essential for applications ranging from high power laser arrays to quantum coherent systems. Here, we present a new paradigm for coherence retrieval inspired by the Gerchberg Saxton framework, enabling reconstruction of the first order spatial coherence function (mutual intensity) of partially coherent light from intensity measurements in the near and far fields. We introduce two complementary approaches: a four dimensional Tensor GS algorithm that directly reconstructs the mutual intensity with high accuracy, and a Monte Carlo GS variant that significantly reduces computational cost at the expense of controlled approximation. We validated both methods by reconstructing partially coherent fields in simulated linear and ring arrays of up to 600 beams with prescribed Gaussian decaying coherence. Experimentally, we applied the Tensor GS method to a triangular array of 130 coupled lasers with inhomogeneous spatial coherence, achieving good agreement with theory and a root mean square phase error as low as 2pi over 250.

A substantial body of recent work has debated whether large language models (LLMs) can serve as substitutes for human participants in behavioural research. This debate, however, captures only one direction of a rapidly changing relationship. The more consequential question is not simply whether researchers should use LLMs as human surrogates, but whether - and under what conditions - humans are beginning to use LLMs as surrogates for their own deliberation. Across domains including health, law, finance, education, and personal guidance, increasing numbers of people consult generative AI systems before, alongside, or instead of human experts, peers, or independent judgment. Although evidence for actual delegation remains uneven, this uncertainty makes the phenomenon an urgent social-scientific object of study. We argue for a research programme that treats LLMs as consequential social actors in a functional sense: systems whose outputs shape human decisions, social norms, and collective dynamics.

Nuclear magnetic resonance (NMR) is a powerful probe of atomic structure, but accurate quantum-mechanical predictions of tensorial NMR parameters are computationally demanding. This creates a bottleneck both for direct quantum-mechanical studies and for collecting high-quality training data for machine-learning (ML) models. Here, we introduce a synthetic pre-training and fine-tuning protocol for graph-based ML models of solid-state NMR parameters. We first pre-train models on synthetic tensorial data, as obtained using an existing ML model, and subsequently fine-tune those models on new ground-truth data. We observe a pronounced improvement in data efficiency when pre-training and fine-tuning span the same compositional and configurational space, and we carry out initial experiments regarding chemical transferability. Our work outlines a route toward future data-efficient training workflows for tensorial ML models for solid-state NMR, combining inexpensive synthetic supervision with targeted first-principles refinement.

Frame-based ranging systems are constrained by frame rate and provide no intrinsic mechanism for background rejection, limiting utility in high-throughput or cluttered environments. We present eSCHORTY, a Scheimpflug LiDAR integrating an event-based sensor with a modulated continuous-wave line laser to enable dense 3D point clouds, generated from over one million megaevents per second. We demonstrate that laser modulation provides a trade-off between event-space feature detection and localization, and that logarithmic event encoding suppresses the reflectance-induced centroid artifact demonstrated in intensity-based ranging. Reconstructions of natural scenes confirm spatially coherent depth recovery, with the Scheimpflug geometry supporting adaptation from millimeter- to kilometer-scale applications.

In this work, we investigate the use of a quantum circuit as a binary classification model in the context of quantum machine learning. We call this model, binary quantum classifier. First, we describe fundamental concepts of quantum computing and introduce the computational tool used: Qibo, an open-source framework for efficient quantum simulations and quantum hardware control. Then, we describe how to design a binary quantum classifier for the classification of images and small arrays of variables by showing how to input data in the circuit, defining a quantum circuit model Ansatz with trainable parameters and a loss function, and implementing multiple minimizers. We test our quantum classifier with two data sets. The first one is the MNIST data set which is composed of handwritten digits (reduced to only handwritten zeros and handwritten ones for binary classification). We study the behavior of different minimizers by increasing the number of layers of the Ansatz. The second data set represents two different high energy collisions that can occur at colliders such as LHC (CERN). Due to in-time proton-proton interactions known as pile-up, we distinguish two different data sets: "without pile-up" and "with pile-up". These collisions can be represented by images of size 32x32 or by six high-level variables that we call features. By increasing the size of the training data set and the number of layers of the Ansatz, we search for the best minimizer. Splitting the data set in training set and test set, we compute: ROC curve, AUC score, confusion matrices and test set accuracy. For "with pile-up" images, we compare the results obtained with the quantum classifier with a small convolutional neural network. We conclude that is possible to build a binary quantum classifier with a quantum circuit and we highlight its performances and limitations in comparison with classical technologies.

N6-methyladenosine (m6A) is the most abundant internal modification in eukaryotic mRNA. However, most existing predictors use adenosine-centered formulations that are computationally inefficient and prone to false positives. Here we present m6A-FORM, a transformer-based foundation model for RNA methylation that uses MeRIP-seq peaks as methylation-enriched priors and is pretrained on approximately 22 million peak-derived sequences from 143 human MeRIP-seq studies. After fine-tuning with high-confidence single-nucleotide m6A annotations from m6A-Atlas v2.0 and GLORI, m6A-FORM-sites achieves state-of-the-art m6A site prediction performance, with a PR-AUC of 0.635 and ROC-AUC of 0.988, improving PR-AUC by at least 0.14 over existing methods while enabling substantially faster inference. Task-specific adaptation further supports prediction of binding sites for 19 m6A-associated regulators and identification of YTHDF2-bound m6A sites associated with mRNA degradation. Applying m6A-FORM across 67 datasets from 24 human tissues identifies 19,631 tissue-conserved sites with distinct localization, clustering, methylation, expression, RBP-interaction, and decay-associated signatures.

Microbial genomes and metagenomes contain millions of proteins whose enzymatic functions remain unknown, the enzyme dark matter. While deep learning has improved protein function prediction, most methods are black boxes relying on sequence or structural similarity, limiting discovery of novel catalytic activities. The ESMC-6B protein language model and its sparse autoencoder with a 16,384-dimensional codebook of interpretable biological concepts, each annotated by GPT-5, creates a new opportunity: using these features directly as semantic signatures for enzyme function. Here, we show that ESMC-SAE features enable accurate and interpretable enzyme commission (EC) number prediction without task-specific training or GPU-intensive computation. On a balanced benchmark of 4,868 microbial SwissProt enzymes across 161 EC3 subclasses, ESMC-SAE binary features achieve 78.9% top-1 and 88.5% top-5 accuracy, 37.6% higher than 3-mer baselines (57.3%). In leave-one-EC3-class-out evaluation simulating discovery of novel enzyme classes, SAE features recover the EC1 superclass in 47.7% of cases (3.3x random, 14.3%), versus 26.6% for sequence methods. Discriminative features correspond to mechanistically interpretable concepts: catalytic triad geometry for hydrolases, NAD(P)H-binding Rossmann folds for oxidoreductases, phosphate-binding P-loops for transferases. We also survey the ESM Atlas of 7.7 million clusters and identify 169,859 dark enzyme-like candidates across all major microbial phyla. Our results establish a paradigm for enzyme function discovery in microbial dark matter: interpretable by design, scalable without GPU clusters, and applicable to the billions of proteins in the ESM Atlas.

Multispecies spatial data arise in many applications where interactions between different entities are central to system behaviour, including biomedical imaging, geospatial analysis, and species ecology. Despite their importance, relatively few quantitative tools exist to capture such interactions. In this work, we propose magnitude-based features for the analysis of multispecies spatial data. Magnitude is a real-valued invariant of finite metric spaces that can be interpreted as an effective number of points, incorporating both spatial configuration and scale. We develop global and local magnitude feature vectors and demonstrate their utility on synthetic tumour microenvironment data, and in tissue microarray data from human colorectal cancer samples. Locally, the method identifies distinct neighbourhood types and reveals spatial heterogeneity; in the model, this includes radial patterns associated with different qualitative outcomes of the simulations, while in the real-world data it reflects the importance of tertiary lymphoid structure-like interactions between B and T cell populations. Globally, the approach recovers known classifications of long-term simulation outcomes across parameter regimes in synthetic data, and suggests important roles for CD4+ T cells and CD163+ macrophages in distinguishing patients with favourable Crohn's like reactions from unfavourable diffuse immune infiltration. Together, these results suggest that magnitude-based features provide a powerful and flexible tool for the analysis of multispecies spatial data.

Interpersonal communication requires building shared semantics that enable listeners to understand speakers' meanings from their unfolding language, but the dimensional structure of this shared neural representation remains unclear. LLMs increasingly approximate human language capability and neural responses, raising the question of whether they capture the same semantic structure shared between human brains. Here, we combined storytelling-listening pseudo-hyperscanning MEG with dimension-resolved interbrain encoding modeling to compare human- and LLM-derived accounts of shared neural semantic representations. Content words from the speaker's narratives were rated by humans and five recent LLMs along ten semantic dimensions (i.e., perception, motor, space, time, socialness, animacy, emotion, attention, causality, and drive). We tested whether these dimensions explained speaker-listener neural synchronization (NS) beyond acoustic and phonological features. Both human- and LLM-derived semantic spaces explained NS, but these shared semantics are better characterized as a multidimensional neural structure rather than a single global signal. These patterns also predicted individual differences in listeners' story comprehension, linking neural alignment to cognition. However, comparable overall prediction concealed systematic differences in representational geometry. Larger LLMs aligned more closely and showed greater overlap with humans in semantic structure and NS, but this was incomplete and dimension-dependent. The largest divergences emerged for dimensions closely tied to agency, affect, and social experience. These findings show that LLMs capture substantial components of human shared neural semantics, but their alignment is selective. Larger or more capable models improve the approximation, whereas socially and affectively grounded dimensions are captured only partially.

Biotic communities vary continuously across space, yet biome maps impose categorical boundaries that compress this variation, particularly at ecotones where transitional communities are ecologically distinct. Could Earth observation (EO) foundation models, which encode spectral, spatial, and temporal information with dense embeddings, convert discrete biome maps into continuous representations that better capture ecological variation? Here, we fit a linear classifier on Clay v1.5 satellite image embeddings to predict biome labels from a categorical map. The softmax output yields a continuous probability vector whose dimensions correspond to named biome classes. We evaluate this approach using six Brazilian biomes, 1.3 million embeddings, and 10,015 withheld forest inventory plots spanning 4,672 plant species. The continuous biome representation outperforms discrete biome labels for predicting species occurrence (mean per-species AUC 0.618 vs. 0.570 across 10 spatial cross-validation folds). Decomposing this gain shows that continuity in the graded probability output, rather than label reassignment, accounts for the improvement; the pattern holds across all distances from biome boundaries. The raw 1024-dimensional embedding remains the strongest predictor we tested (mean AUC 0.646 vs. 0.618), but the continuous representation recovers most of the embedding's gain over discrete labels. This simple approach provides a probabilistic replacement for categorical map labels, preserving their meaning while encoding graded variation that discrete maps suppress.

Chemical space is unimaginably vast with common heuristic estimates suggesting that there are ca. 10^60 'drug-like' molecules possible below a molecular mass of 500 Da. However, these estimates largely ignore the structural and synthetic complexity of the molecules enumerated. Here we present a first-principles estimate of the size of chemical space using the Assembly Theory, which quantifies the amount of causation required to form a molecule, captured in the assembly Index. This is a measurable molecular complexity measure derived from the minimum number of recursive bond-joining operations required to construct a molecular graph. Assembly Theory partitions chemical space into levels defined by Assembly Index, allowing bounds to be placed on its growth as molecular complexity increases. We show that chemical space (the accumulated Assembly Index level sets) grows at least super-exponentially, and at most, double-exponentially with respect to the Assembly Index. Using the GDB-13 database as a reference for growth-rate estimation, we model how chemical space expands under increasing complexity and contracts under structural constraints, including atom and bond types, number of rings, ring size, and chemical motifs. Under constraints comparable to standard drug-like estimates, including molecular mass below 500 Da, our analysis yields a chemical space of approximately 10117 molecules at Assembly Index 25. Finally, we constrain chemical space by biologically relevant motifs and identify structurally relevant molecules near the accessible boundaries of these assembly-defined spaces.

While long treated as empirical fits, Hill functions have been postulated to be the universal Hopfield barrier for sharpness of input-output responses by Martinez-Corral, Nam, DePace, and Gunawardena. A Hopfield barrier is a fundamental limit on how well biological systems can process information without expending energy. Their case rested on numerical findings for Hill coefficients $4$ and $6$. We give a precise formulation and proof of this: measuring sharpness by the supremum of the derivative in semi-log scale, any rational function $r(x)=(α_0+α_1 x+ \cdots +α_n x^n)/(β_0 + β_1 x+ \cdots + β_n x^n)$ with real coefficients $0\leq α_i\leq β_i$ has sharpness at most $n/4$, with equality if and only if $r$ is a Hill function with Hill coefficient $n$.

Network control theory can be used to model intrinsic and extrinsic strategies to steer neural dynamics. Standard approaches are node-centric, structural, and focused on achieving desired instantaneous states. Here, we develop an edge-centric approach which incorporates both structure and function to achieve extended patterns of neural dynamics characterized by desired synchronization states. Our method, Quantifying Underutilized Influential Edges for Targeted Synchronization (QUIET), is an edge-centric framework that integrates structural controllability of individual white matter connections and mutual information between pairwise functional timeseries to identify energy-efficient synchronization pathways. QUIET identifies quiet highways, edges that are structurally influential but functionally underutilized, to optimize regional synchronization. We validated QUIET across 75 synthetic configurations, where QUIET-ranked edge sets significantly outperformed random selection in 93% of cases (p<0.01). The framework, tested on Human Connectome Project participants, revealed that the control energy required for synchronization of the salience network correlates with fluid intelligence. QUIET, applied to healthy adults undergoing dexmedetomidine-induced unresponsiveness, showed that the frontoparietal and default-mode networks exhibited the largest control energy required for synchronization in both awake and sedated states. QUIET is released as a stand-alone software to be used to study theoretically-defined synchronization pathways, which in turn could inform testable hypotheses in perturbative studies.

Conventional beamforming techniques primarily steer energy along desired directions or focus it at specific locations. These techniques become fragile when facing frequent blockage and highly dynamic propagation environments. In this article, we present caustic beamforming as a new paradigm for wireless beam control. First, we classify representative caustic beams according to their underlying mathematical origins and present three unique properties, namely self-bending, self-healing, and near-field non-diffracting. Building on these propagation properties, we then propose several application scenarios in sixth-generation (6G) networks. We undertake two case studies focused on physical layer security and service stability that highlight the capability of caustic beams to bypass potential eavesdroppers, deliver more uniform coverage, and sustain blockage-resilient links. We further discuss the enabling hardware architectures that facilitate practical deployments, and finally outline key open challenges regarding caustic beams that require further research.

This paper presents a theoretical framework for multi-band localization for a single-path single-input multiple-output (SIMO) system. We derive closed-form Cramer-Rao bounds (CRBs) for angle-of-arrival (AoA) and distance for uniform linear arrays (ULAs), and an intermediate matrix-form formulation for arbitrary array shapes. We also develop benchmark single- and multi-band maximum-likelihood (ML) estimators for AoA-Distance, leveraging a structured Levenberg-Marquardt (LM) refinement procedure. A key contribution is an analytical characterization of the threshold SNR (TSNR) for the proposed estimators. This is the SNR threshold at which the estimator transitions from "off the chart" to CRB-approaching performance, for both TDoA and distance estimation. Numerical simulations confirm that the proposed single- and multi-band estimators achieve the CRB at SNRs above the predicted TSNR, and that multi-band processing simultaneously improves estimation accuracy and reduces SNR requirements. The resulting framework provides a rigorous foundation for next-generation multi-band localization and can be readily extended to elevation estimation, distributed arrays, and multi-path environments.

Future 6G heterogeneous wireless networks (HWNs) are expected to support multiple radio access technologies (RATs), dynamic wireless environments, and applications with diverse quality-of-service (QoS) requirements. In such environments, network selection (NS) cannot rely only on instantaneous radio measurements or static ranking rules. Instead, access decisions must account for the evolving wireless state, service intent, packet-level QoS behavior, and candidate-RAT dynamics. This paper proposes a large language model (LLM)-based digital twin (DT) framework for stable, application-aware RAT selection under candidate-set evolution. The main idea is to shift NS from an instantaneous decision-matrix operation to a decision process over an evolving wireless DT state. The constructed DT combines site-specific geometry, Sionna RT-based propagation descriptors, ns-3 packet-level QoS emulation, service context, candidate-RAT information, and decision memory. Rather than acting as a general-purpose controller for 6G networks, the LLM is used for DT-grounded decision intelligence in this specific NS task. On top of this DT, a unified intent agent translates user and service requirements into structured decision priorities for two complementary NS branches: an LLM-assisted multi-attribute decision-making branch (MADM--LLM--NS) and a direct LLM-based ranking branch (LLM--NS). To improve decision stability, the framework further introduces history-aware adaptive normalization (HAAN) and DT-memory-driven retrieval-augmented in-context learning (RA--ICL). Numerical results show that the proposed framework reduces rank-reversal problem and unnecessary handover events, while improving service-aware QoS satisfaction compared with representative MADM-based NS baselines.

SSpeech imagery is attractive as a brain-computer interface paradigm for communication because it is endogenous and intrinsically linguistic. Yet despite growing interest, its dominant scalp-EEG spatiotemporal characteristics remain poorly characterized. Here, we asked how speech imagery appears in scalp EEG and compared it against finger motor imagery. Using a within-subject dataset containing speech imagery, finger motor imagery, and no-task trials recorded under the same trial structure, we analyzed band-power dynamics across channels and time. Finger motor imagery showed the expected contralateral mu/alpha and low-beta desynchronization over sensorimotor areas, whereas speech imagery showed a weaker, more distributed alpha-dominant increase. After normalization to each condition's own post-trial interval, the speech-related alpha increase changed only modestly after cue onset, indicating that much of the speech-versus-no-task difference was already present during the instruction period. A classifier discriminating imagery from no-task reached mean balanced accuracies of 0.563 $\pm$ 0.072 for speech imagery and 0.718 $\pm$ 0.127 for motor imagery, with a stronger alpha/beta dependence for motor imagery than for speech imagery. Together, these results provide a clearer group-level characterization of speech imagery in scalp EEG and indicate that its dominant spatiotemporal pattern differs from that of finger motor imagery and is more consistent with substantial non-articulatory task-related contributions than with a clear articulatory-motor analogue.

In this paper, we study the power allocation for an integrated sensing and communication (ISAC) system which tracks a mobile target. We first model the problem as a Markov decision process, and then tackle it with a soft actor-critic (SAC) based deep reinforcement learning (DRL) approach. We also combine a Dirichlet policy, which naturally produces normalized continuous actions under random target motion. To exploit different features of sensing and communication operations, we carefully design a reward function such that the system can dynamically control power allocation to conserve resources. The simulation results demonstrate that the proposed scheme enhances tracking performance compared to other baselines while sustaining communication performance.

Recently, movable antenna (MA) has attracted wide attention in wireless communications due to its potential in enhancing wireless communication performance via local movement within a confined region. However, antenna position optimization (APO) has emerged as a major challenge for MAs, due to the lack of a tractable, analytical, and accurate channel model in terms of antenna positions. Although existing works have developed various algorithms for APO, most of them are based on simplified theoretical channel models, which limit their generality. To address this challenge, in this article, we present more general and effective APO algorithms for different purposes, categorized as continuous APO and discrete APO, respectively. Continuous APO is mainly applied for flexible array signal processing to boost large-scale communication performance, while discrete APO is applied for small-scale multi-path channel reshaping. Specifically, the discrete APO discretizes the antenna movement region into multiple sampling points and employs discrete algorithms to determine the optimal MA positions based on the point-wise channel state information (CSI), without the need for an analytical channel model. To reduce the overhead for CSI acquisition, we also present more efficient learning-based APO algorithms that operate without requiring full point-wise CSI. Finally, we compare the application scenarios of the proposed algorithms and validate their effectiveness with numerical results.

Epilepsy affects over 50 million individuals globally, underscoring the need for automated seizure detection systems that can alleviate clinicians workload and enhance the accuracy of patient seizure diaries. In wearable EEG applications, however, reliable detection remains challenging due to the limited spatial resolution of low-density electrode configurations, reduced signal-to-noise ratios, and the scarcity of diverse, publicly available training datasets. This study investigates the efficacy of hybrid deep learning architectures for automated seizure detection using a simulated behind-the-ear montage derived from the Temple University Seizure Corpus (TUSZ, v2.0.3). We conduct a systematic comparison of several CNN-RNN models, including LSTM- and GRU-based variants, across multiple EEG montages to evaluate their capacity to compensate for the loss of spatial information inherent to reduced electrode configurations. The proposed CNN-Merged model, which integrates temporal and spectral feature representations, demonstrates superior performance, achieving a ROC AUC of 85.89% and a balanced accuracy of 79.11% on the held-out test set. Furthermore, the model exhibits strong robustness across different reference montages, effectively bridging the performance gap between conventional full-scalp recordings and resource-constrained wearable systems. These findings substantiate the potential of hybrid deep learning models as a promising avenue toward robust, patient-independent seizure detection in low-density EEG applications.

To meet the demands of 6G wireless systems operating in high-mobility scenarios, this paper presents a design of a random multiplexing (RM) communication system that is both storage-efficient and highly reliable. In principle, RM with cross-domain memory approximate message passing (CD-MAMP) can achieve replica maximum a posteriori (MAP)-optimal performance by constructing a fully dense equivalent channel matrix. However, its practical implementation is hindered by the large storage overhead of conventional interleavers and by performance degradation in severely ill-conditioned channels, which existing related work (focusing on interleaving and transform designs) fails to address simultaneously. To overcome these issues, we develop a storage-efficient and highly reliable system that integrates RM with CD-MAMP, referred to as RM-MAMP. Specifically, we propose a Logistic chaotic mapping interleaver with a quantitative parameter-selection criterion, and a dual-stage high-order permutation polynomial interleaver, both of which achieve nearly identical bit-error-rate (BER) as fully random interleavers while reducing the interleaver storage from O(N) to O(1) and significantly lowering interleaver signaling overhead. We further propose a highly reliable interleaved transform framework, comprising an interleaved phase perturbation transform and a multi-layer interleaved coupled transform, to enhance the incoherence and diversity of the equivalent channel matrix. Simulation results show that the proposed storage-efficient interleavers maintain BER performance comparable to fully random interleavers, while the highly reliable transforms provide over 4 dB gain in severely time-varying channels, confirming the dual benefits of reduced storage overhead and improved robustness for the enhanced RM-MAMP system.

We investigate the impact of power amplifier (PA) nonlinearities on the sensing performance of affine filter bank modulation (AFBM). While AFBM offers several advantageous properties for integrated sensing and communications (ISAC) - including reduced out-of-band emission (OOBE), low peak-to-average power ratio (PAPR), and natural robustness to doubly-dispersive (DD) channel effects - mitigating waveform distortion typically requires highly linear PAs. This creates a fundamental contradiction with ISAC applications, which demand high transmit power for reliable sensing. Our analytical results reveal that the structure of the effective AFBM modulation matrix dictates how distortion propagates within the ambiguity function (AF). Furthermore, simulations demonstrate that both the AF and the overall sensing performance of AFBM remain remarkably insensitive to such nonlinearities. These findings highlight the robustness of AFBM, making it a highly viable candidate for practical ISAC deployments constrained by hardware impairments.

Due to high power consumption and hardware costs of fully digital arrays, hybrid beamformers are often considered as a more economic alternative. Furthermore, using high resolution analog to digital converters (ADCs) can also have prohibitive power consumption, which leads to lower resolution converters being considered for radio frequency (RF) front end design. The finite quantization resolution as well as the nonlinearities caused by the power amplifiers (PAs) and low noise amplifiers (LNAs) can have a substantial impact on system performance. While widely studied for communications, the impact of hardware impairments on sensing performance is considerably less explored. In this work, we study the interplay between hybrid beamforming architectures, hardware impairments, and sensing and communications performance. Additionally, we define the concept of double-isotropy for pilot-combiner pairs, formalizing the notion of a perfectly energy-fair beam sweep. The multiple start (MS) space alternating generalized expectation maximization algorithm (SAGE) is also introduced, aimed at addressing the optimization issues arising from parametric channel estimation (PCE) in hybrid beamformed systems. We then provide a set of numerical results assessing the impacts of beamformer architecture and ADC resolution on PCE, sensing, and communications performance. The results show that medium resolution ADCs lead to the most power efficient configurations, with the best tradeoff between power consumption and performance for the majority of beamforming architectures. Additionally, fully digital beamforming architectures with high resolution converters can often be substituted for a hybrid beamformer setup with medium resolution converters without significant performance loss at a lower power consumption and overall hardware cost.

Power leaking directly from transmitting into receiving radio-frequency chains is a key challenge in the realization of monostatic sensing applications with multi-antenna communication front-ends, to which a promising solution is digitally precoding transmitted signals for improved leakage suppression. While digital transmit precodings perform well in theory, real-world deployments typically exhibit severely degraded leakage suppression. This work investigates quantization noise as a primary factor limiting the performance of such precoding schemes. A closed-form solution predicting the impact of quantization noise on the performance of arbitrary digital joint leakage estimation and leakage suppression precodings is derived, numerically analyzed, and validated in a hardware testbed.

The Frequency Range 3 (FR3) band is attracting increasing attention due to limited lower-frequency spectrum and growing mobile communication demand. This study experimentally investigates channel characteristics in Urban Macro (UMa) scenarios at 8 GHz and 15 GHz using a large-scale MIMO platform with time-division multiplexing (TDM). Key parameters, including root mean square (RMS) delay spread (DS) and angular spread (AS), were extracted and compared with 3rd Generation Partnership Project (3GPP) TR 38.901. Results reveal clear frequency-dependent behaviors: RMS delay spread remains nearly constant under line of sight (LOS) but decreases from 8 GHz to 15 GHz in non-line of sight (NLOS), indicating reduced multipath dispersion at higher frequencies. Both azimuthal spreads (including ASA and ASD) and elevation spreads (including ESA and ESD) exhibit a corresponding decrease with increasing frequency, demonstrating a consistent trend towards more directional propagation across all angular domains. Capacity analysis indicates that the 15 GHz channel slightly outperforms 8 GHz in both LOS and NLOS scenarios due to more concentrated multipath energy and larger dominant singular values. Higher frequencies exhibit greater directionality, whereas lower frequencies provide broader multipath distributions and more stable performance, offering valuable guidance for multi-band MIMO modeling and 6G system design.

Pixel antennas enable antenna coding, a technique that can provide more degrees of freedom in wave manipulation, to enhance wireless communications. However, acquiring full channel state information (CSI) at the transmitter incurs prohibitive overhead due to the unique hardware constraints from pixel antennas. This paper thus proposes a limited feedback multi-input multi-output (MIMO) system using pixel antennas, where the antenna coder and digital precoder are designed based on pre-defined codebooks and efficient index feedbacks. We first derive the optimal digital precoder under practical power constraints that provides insights on simplifying the joint codebook construction for antenna coder and digital precoder. We then develop a low-complexity offline codebook construction algorithm that enables subsequent codebook designs for the antenna coder and digital precoder. Simulation results demonstrate that the proposed scheme significantly outperforms unconstrained MIMO systems using conventional antennas with fixed configurations.

This paper investigates coherent multiband orthogonal frequency division multiplexing (OFDM) sensing within an integrated sensing and communication (ISAC) framework. We consider an intra-band configuration in which two sensing subbands of equal width are allocated symmetrically within the same OFDM channel, while the central portion remains available for communication. We address the reconstruction of missing frequency-domain samples induced by the spectral gap and the suppression of the resulting grating lobes in the delay profile. To this end, we propose a low-complexity iterative reconstruction method consisting of an initial delay-domain equalization stage and an iterative apodization-based operator with data-consistency enforcement. Performance results for multi-target scenarios show that the proposed approach remains close to the full-band reference for moderate gap sizes and degrades only for larger gaps because of residual grating lobes. Compared with the compressed-sensing-based orthogonal matching pursuit (OMP) baseline, it exhibits a more favorable performance trend as the number of targets increases, especially in the practically relevant low-signal-to-noise ratio (SNR) regime, while offering a complexity scaling that is independent of the estimated number of targets.

The cumulative distribution transform (CDT) is a quantile-based transport representation that exactly linearizes one-dimensional translations of positive densities. We study how this structure behaves under additive perturbations and how it can be exploited for shift recovery. Under a local nondegeneracy condition, we derive a first-order expansion showing that additive noise in physical space induces a nonlocal perturbation in CDT space through the primitive of the noise, weighted by the reciprocal density. This yields an explicit description of transform-domain sensitivity and shows, in particular, that perturbations are amplified in low-density regions. When the physical-space perturbation is modeled as a centered Gaussian random field, the induced first-order CDT perturbation is again Gaussian, with an explicit covariance kernel. We then use this structure to study recovery in CDT coordinates. In the known-template setting, the transport shift is obtained by projection onto the constant mode, giving an explicit estimator together with exactness in the noiseless case and a stability bound under perturbations. In the unknown-template setting, multiple observations permit joint recovery of the shifts and a common template up to the natural constant-mode gauge, leading to a simple de-shift--and--average procedure. We also consider a signed-signal analogue based on the signed cumulative distribution transform (SCDT), where shifts are estimated numerically by feature matching and unknown templates are recovered by alternating alignment and averaging. Numerical experiments validate the perturbation analysis and illustrate effective recovery for both density-valued and signed signals.

A single-radio-frequency (RF) movable array is investigated, in which all movable elements are driven by a single RF chain with equal amplitude and equal phase. The achievable beamforming gain enabled by antenna placement is analyzed. Linear beamforming gain scaling with the number of antennas is shown to be achievable in single-path channels, while coherent-combining conditions and aperture requirements are established for multipath channels. For multiuser transmission, the optimal max-min power allocation is derived in closed form, based on which an element-wise coordinate-search algorithm is developed for antenna placement design. Numerical results validate the analysis and reveal a fundamental tradeoff: beamforming gains can be achieved through antenna placement alone, but only at the expense of increased aperture resources.

Carrier-phase (CP) ranging is a key enabler of high-precision positioning in modern wireless systems. In multi-frequency OFDM-based sensing, phase observations across subcarriers provide information about the underlying propagation geometry. However, in realistic industrial and urban environments, these observations exhibit non-Gaussian and asymmetric characteristics due to deterministic multipath components, violating standard circular statistical assumptions. In this work, we analyze CP-based ranging as an estimation problem over circular phase observations. We show that conventional model-based estimators, such as circular averaging under von Mises assumptions, become biased under 3GPP-compliant propagation conditions. Using a QuaDRiGa-based simulation framework, we evaluate empirical phase distributions in Industrial Factory (InF) and Urban Microcell (UMi) scenarios and quantify their deviation from classical statistical models. To address these limitations, we propose a learning-based estimator that operates directly on empirical phase distributions without assuming a predefined statistical model. Experimental results show improved accuracy compared to classical estimators, particularly under multipath conditions.

The joint rate-distortion framework of Stavrou and Kountouris (IEEE Transactions on Communications 2023) characterises dual-fidelity tradeoffs for semantic communication on stochastic semantic sources. Many task-oriented communication systems instead use designed sources, where the semantic object is a deterministic oracle allocation $φ^(t)$ rather than a stochastic quantity given by nature. We isolate the subclass of designed sources under smooth concave utility with assumptions A1, A2 and Euclidean allocation codomain, and restrict the encoder class to deterministic common-category mappings. Within this subclass the SK exponential-tilting decoder and generalised Blahut--Arimoto iteration specialise to conditional-mean decoding and Lloyd--Max stationarity on $φ^(t)$. When the second fidelity is a monotone single-letter distortion, the joint problem stays inside the SK admissible class; the common-category SK rate is lower-bounded by the max of the corresponding Shannon rate-distortion functions, with equality only when the common-category reconstruction is compatible and RDF-optimal. When the second fidelity is aggregate verification, the joint problem leaves the SK single-letter class and admits a constrained-design feasibility band $R_{\min}(\varepsilon^) \leq R \leq R_{\max}(β^)$ of width $\log_2(K_{\max}/K_{\min})$ bits in partition cardinality. The reduction and the band are scope statements on the SK apparatus, not modifications to it. A smart-grid economic-dispatch example with a non-technical-loss-detection contrast illustrates the band.

Channel estimation in vehicular communication is a crucial element in the advancement of intelligent transportation systems. However, the use of pilot signals in the IEEE 802.11p standard is insufficient for accurate channel estimation in high-mobility scenarios. Data pilot-aided (DPA) estimation helps address this, but suffers from demapping errors. We propose a simplified Temporal Convolutional Network-based estimator (DPA-TCN) trained on a mixed signal-to-noise ratio dataset to improve estimation performance and reduce computational complexity. Our DPA-TCN estimator achieves a bit error rate comparable to a state-of-the-art long-short-term memory network with DPA and temporal averaging (LSTM-DPA-TA) while reducing the complexity of the model by approximately 65%.

Fractional quantum Hall states are a cornerstone of topological physics, hosting fractionally charged quasiparticles with exotic statistics that promise to enable topologically protected quantum information processing. Among these, the Pfaffian state introduced by Moore and Read implements a p-wave pairing structure that supports excitations with non-Abelian exchange statistics. Despite extensive study in electronic systems, direct access to its pairing structure has remained limited. Here we realize a three-particle bosonic Pfaffian state of ultracold $^{87}\mathrm{Rb}$ atoms in an optical lattice subject to a Floquet-engineered synthetic magnetic field. Using a Bayesian-optimized adiabatic protocol, we prepare a state exhibiting Pfaffian pairing correlations. Site-resolved measurements of multi-point density correlations reveal a pronounced suppression of short-range three-body coincidences, reflecting the underlying pairing structure. We further probe the state's transport response through Hall drift measurements. Our results establish a bottom-up approach to engineering non-Abelian topological order and lay the groundwork for future explorations of anyonic braiding in synthetic matter.

Neutrino evolution in dense astrophysical environments is typically described either within a quantum kinetic framework, which neglects the build-up of multi-body correlations, or through simplified many-body calculations that allow significant entanglement to develop. In this work, we compare these two approaches in a simple neutrino-gas configuration, with particular emphasis on the role of non-forward scattering processes. These effects are incorporated either through a collision term in the kinetic description, or by considering the full neutrino-neutrino many-body Hamiltonian. We highlight differences between the two descriptions in both their characteristic timescales and asymptotic behavior. Motivated by the natural suitability of quantum computing for many-body calculations, we further investigate the non-classicality of neutrino evolution, discussing Trotter error scaling, along with the associated costs of constructing quantum circuits in terms of entangling gates and non-Clifford gates. We find that the resources needed for neutrino many-body evolution are on the low end of typical high-energy physics problems and on the mid to high end with respect to quantum chemistry problems. For the full Hamiltonian, resource requirements increase relative to the truncated version. We emphasize the importance of efficient fermion-to-qubit encodings, which are essential for reducing the substantial computational resources required for such simulations.

We consider the problem of purifying noisy qubit unitary channels. Given the ability to apply an unknown qubit unitary channel followed by depolarizing noise, we aim to construct a superchannel that purifies the noisy unitary back to the original unknown unitary. We first provide numerical evidence that sequential strategies can strictly outperform parallel strategies when the number of channel uses is finite, highlighting the fundamental distinction from state purification. We then provide a concrete $\mathrm{U}(2)$-covariant parallel protocol based on a novel entanglement-assisted quantum error-correcting code that suppresses the first-order noise strength as $O(1/n)$ with $n$ channel uses and show this scaling is asymptotically optimal in the low-noise regime, even when sequential strategies are allowed.

The stabilizer fragment of the ZX calculus is amongst the most important fragments of the theory. The closely related Clifford+T fragment is approximately universal (arXiv:1705.11151). Additionally, the stabilizer calculus can be described by a small collection of rewrites, most of which have been shown to be necessary (arXiv:1709.08903). However, two rules, describing the red/green compact-structure coincidence and the important bialgebra law, had not been shown to be necessary. We present a countermodel-style argument showing that both of these rules are individually necessary relative to the connectivity meta-rule of Backens--Perdrix--Wang (arXiv:1709.08903), and hence establish that the rule set presented in arXiv:1709.08903 has no redundant rewrite rule.

Bell's theorem shows that entangled quantum particles can exhibit correlations that classical particles cannot reproduce without an additional nonlocal resource, such as communication. In this sense, quantum particles are fundamentally more nonlocal than classical ones, and entanglement becomes unavoidable in physics. Here we prove the analogous result within quantum theory itself: indistinguishable fermions transmitted through a quantum network can generate correlations that distinguishable particles or indistinguishable bosons cannot reproduce without additional communication. In the same sense, fermions are fundamentally more nonlocal than bosons or distinguishable particles, motivating fermionic anticommutation and indistinguishability as unavoidable operational resources. Our result further implies that fermions can strictly surpass all qubit-based protocols for certain distributed computing tasks, demonstrating that a complete understanding of information processing requires going beyond qubits to fermionic information carriers - febits.

We theoretically investigate spin- and valley-resolved quantum transport in monolayer tungsten diselenide (WSe$_2$) described by an effective massive Dirac Hamiltonian. Particular attention is devoted to a finite barrier region characterized by simultaneously modulated Fermi velocity and scalar potential. The barrier velocity $v_2$ is related to the external velocity $v_1$ through a velocity ratio $ξ=v_2/v_1$, motivated by an optical analogy with the Snell-Descartes law. The exact refraction condition depends on the full spin- and valley-resolved dispersion, and the simple ratio $ξ=v_2/v_1$ is recovered only in the massless, symmetric limit. The interplay of intrinsic spin-orbit coupling in the conduction and valence bands, quantified by $λ_c$ and $λ_v$, with spin- and valley-dependent Zeeman fields, $M_s$ and $M_v$, gives rise to substantial changes in the quasiparticle dispersion, leading to pronounced modifications of the transport characteristics. By solving the Dirac equation and enforcing current-conserving matching conditions at the interfaces, we compute the spin- and valley-dependent transmission probability and conductance. Our results demonstrate that the barrier velocity, scalar potential, incidence angle, incident energy, and barrier width serve as effective control parameters for transport, giving rise to strong anisotropy and resonant tunneling features. Furthermore, we show that both the magnitude and orientation of spin- and valley-polarized currents can be continuously tuned via velocity and potential modulation. These findings establish combined velocity and potential engineering as a powerful theoretical framework for controlling spin-valley physics in two-dimensional transition-metal dichalcogenides.

We consider a movable conducting plate of finite mass, between two fixed ones, whose mechanical degrees of freedom are treated quantum-mechanically and bound to its equilibrium position by a harmonic potential. The movable wall is thus subjected to quantum fluctuations of its position. This creates a system of two sub-cavities separated by the movable fluctuating plate, and two massless one-dimensional scalar fields, one in each sub-cavity. This system is described by an appropriate generalization of the Law Hamiltonian. The presence of the movable wall yields an effective plate-fields interaction, as well as an effective interaction between the field modes. We obtain, at the second order in perturbation theory, the ground state of the interacting system and the reduced density operator of the fields in each sub-cavity by tracing out the wall's degrees of freedom. We calculate the entanglement between two field modes, one in each cavity, by evaluating analytically the negativity; we then evaluate numerically also the total multimode negativity. Our results show that in both cases the fields in the two sub-cavities are entangled, in contrast to the case in which the wall is fixed in space. We discuss the amount of the field entanglement present as a function of relevant physical parameters of the system such as the mass and oscillation frequency of the movable wall, its distance from the fixed walls and the frequencies of the field modes considered.

A quantum repeater segment is a basic building block of a quantum repeater, generating buffered entanglement of quantum memories to connect quantum repeater cells. It also enables the connection between quantum computers. In the implementation we present here, photons emitted from two co-trapped free-space coupled $^{40}$Ca$^+$ ions are converted to the telecom-C band and interfered after transmission over 440$\,$m of optical fiber (220$\,$m per arm), where a photonic Bell measurement is performed to create entanglement between the memories. With this scheme we generate an entangled $\left|Ψ^+\right\rangle$ Bell state with $\ge 68(8)\,$% fidelity, highlighting trapped $^{40}$Ca$^+$ ions as a promising quantum repeater hardware platform.

The standard in Earth-observation tasks today is having near real-time access to surface images in response to changing conditions. For instance, as urban environments interface more with wildlands and wildfires become less predictable, their tracking with satellite resources becomes essential. This requires the coordination of increasingly large constellations of satellites, giving rise to challenging computational problems. With wildfire detection and tracking as a backdrop, we investigate the power of special purpose and novel computing paradigms to tackle the ensuing satellite scheduling problems, making a compelling case for quantum algorithms. We bring quantum scheduling algorithms closer to implementation by examining both the emerging iterative quantum algorithm framework, which comes with analytic guarantees compared to some classical algorithms, and distributed quantum computing methods whose relevance is on the rise as utility-scale problems begin to get solved with quantum computers. Drawing strength from several computing fronts, we develop a distributed/parallelization scheme in conjunction with the quantum algorithm design and apply these techniques to real-world datasets for wildfire detection. While our quantum subprocesses are currently too small to see significant quantum advantage, our results validate the utility of these techniques, and continue forging the path toward distributed quantum computing.

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.

The Ising framework maps the decoding problem in quantum error correction onto ground-state optimization of a classical Hamiltonian, in which $X$-$Z$ error correlations enter as cross terms. Under phenomenological depolarizing noise, the exact joint formulation contains up to 8-body interactions for the toric code and 10-body for the $6.6.6$ color code. These high-order terms degrade solver convergence, inflate runtime, and raise the auxiliary spin overhead when embedding into native 2-body Ising hardware. In this work, we propose the iterative low-order decoding (ILOD) algorithm, which alternates between $X$- and $Z$-type sub-Hamiltonians, approximating cross-type correlations through Bayesian priors that reweight each type's couplings using the other type's inferred error configuration. This halves the maximum body count of interaction terms in the Hamiltonian, accelerating the solver, restoring convergence at larger code distances, and reducing the total spin count for 2-body embedding by a factor of $2.5$. For the toric code, ILOD attains a threshold of $4.73%$ versus $4.83%$ for the joint formulation, with the empirical runtime ratio scaling as $(0.81)^d$. For the $6.6.6$ color code, their thresholds agree within statistical uncertainty for small code distances, and ILOD remains convergent for larger distances where the joint formulation fails to converge despite a larger annealing budget.

The origin of cosmic acceleration remains a central problem in cosmology, commonly attributed to a cosmological constant within the $Λ$CDM model or to dynamical dark energy. Here, we develop an alternative approach in which acceleration emerges from quantum post-selection, a standard feature of quantum theory that is not usually incorporated into cosmological modelling. While quantum theory admits both pre-selected and post-selected ensembles, quantum cosmological models are almost exclusively formulated in terms of initial conditions. Building on previous work on post-selected quasiclassical dynamics, we construct a minimal predictive cosmological model in which post-selection and coarse-graining generate effective late-time acceleration without introducing a cosmological constant, dark energy, or modifications of general relativity. The resulting expansion history is highly constrained theoretically and depends on at most two parameters beyond standard Friedmann evolution. Confrontation with type Ia supernova and cosmic chronometer data yields statistically competitive fits while naturally avoiding the coincidence problem. The model also reproduces the standard radiation- and matter-dominated behaviour at early times and predicts a present-day jerk parameter significantly different from the $Λ$CDM value. These results suggest that cosmic acceleration may arise as a macroscopic quantum cosmological effect rather than from additional cosmological fluids or modified gravitational dynamics.

Reference frame mismatch among communication parties introduces errors in quantum cryptographic protocols. As the number of participants increases, aligning reference frames becomes increasingly difficult, complicating multipartite quantum cryptographic implementations. Here, we theoretically and experimentally investigate multipartite reference-frame-independent (RFI) quantum cryptographic communication using Greenberger-Horne-Zeilinger (GHZ) states. We generalize the bipartite RFI security parameter $C$ to an $N$-party parameter $C_N$ and derive the asymptotic secret key rate expressed solely in terms of experimentally accessible quantities. We analyze the key rate under global and local depolarizing noise models and find that increasing the number of parties $N$ enhances robustness against global depolarizing noise while increasing vulnerability to local channel noise. We also present a proof-of-principle experimental demonstration of four-party RFI quantum cryptographic communication using four-photon GHZ states, confirming the reference-frame invariance of both the $C_4$ parameter and the secret key rate under various reference frame rotations.

Characterization of time-frequency (TF) quantum states requires reliable reconstruction of their TF distributions. However, imperfect transmission or measurement channels can distort reconstructed joint spectral intensities (JSIs), especially when the underlying perturbation mechanism is unknown. Here, we experimentally demonstrate a reconstruction and correction framework that uses a TF grid state as an intrinsic frequency-domain reference. By analyzing the displacement of the grid points, a Gaussian process regression model is employed to reconstruct a correction mapping for the nonlinear coordinate deformation without assuming a prior physical model of the distortion. The learned mapping reduces the residual coordinate deviation of the TF grid state by approximately a factor of 11 and, when applied to an independent frequency-entangled test state, improves the Gaussian-shape fidelity from 76.2\% to 90.0\%. These results establish TF grid states as practical metrological resources for diagnosing and correcting distortions in TF quantum systems, providing a pathway toward distortion-resilient quantum communication and high-dimensional quantum information processing.

A central principle in quantum machine learning is that an ansatz should be expressive enough to represent the quantum data of interest. Yet, the expressibility is statistically meaningful only insofar as it can be learned from finitely many copies of an unknown quantum state. In this work, we develop an information-theoretic Occam theory for quantum data generated by finite-size quantum circuits. For the class $S_{n,G}$ of $n$-qubit pure states preparable with at most $G$ two-qubit gates, a metric-entropy argument gives the realizable sample law $\widetildeΘ(G/ε^2)$ in the circuit-limited regime. For an arbitrary source $\hatρ$, we introduce the best $G$-gate approximation error $d_G(\hatρ)$ and the approximate circuit complexity $C_η(\hatρ)$. We prove an agnostic quantum Occam theorem: with $M$ copies, one can learn up to the best $G$-gate approximation error plus a statistical penalty $\widetilde{O}(\sqrt{G/M})$. We then remove the need to know $G$ in advance through an adaptive model-selection theorem whose oracle inequality selects the circuit complexity justified by the data. Matching lower bounds yield a sample-supported expressibility law: at trace-distance accuracy $ε$, $M$ samples can support only $G_{\rm supported} \simeq Mε^2$ gates, up to logarithmic factors and tomography saturation at $2^n$. Thus, the circuit complexity becomes an adaptive statistical resource rather than a static promise. Our framework turns bounded circuit complexity into a model-selection principle for quantum machine learning.

The distribution of the effective inspiral spin ($χ_\mathrm{eff}$) of the binary black holes detected by LIGO-Virgo-KAGRA can shed light on their formation pathways. We analyze the GWTC-5.0 dataset with two models-one flexible, one fully parametric-that jointly describe $χ_\mathrm{eff}$ and primary mass. We clarify that the previously-reported skewness in the $χ_\mathrm{eff}$ distribution is better understood as additional structure beyond a non-skewed Gaussian bulk centered at small $χ_\mathrm{eff}$. This additional structure extends to larger $|χ_\mathrm{eff}|$, a result previously reported using GWTC-4.0 data. We measure the asymmetry of the distribution of $χ_\mathrm{eff}$ outside the Gaussian bulk from the data. With both the parametric and the flexible analyses, we find tentative evidence for a mass-dependent excess of positive $χ_\mathrm{eff}$ over negative ones outside the Gaussian bulk. Only at $m_1 \in [46,65]\,M_\odot$ do the data require a negative $χ_\mathrm{eff}$ component outside the Gaussian bulk, with $23\text{:}1$ odds. If $χ_\mathrm{eff}$ outside the Gaussian bulk are produced by hierarchical mergers-as it has been suggested-then a fraction of those mergers may be produced in environments that can generate a surplus of binaries with positive $χ_\mathrm{eff}$, such as the disks of active galactic nuclei.

Single-wall carbon nanotubes (SWCNTs) are narrow ribbons of graphene with atomically precise edges and a single quantum transport channel, at experimentally-relevant dopings. This makes them ideal systems to harness quantum transport straintronics (QTS), i.e. using mechanical strain to control accurately quantum transport. We present QTS data from three single-wall carbon nanotube quantum dot (SWCNT-QD) transistors over a broad range of in-situ tunable and reversible uniaxial strain ($Δ\varepsilon_\text{mech}\approx$ 0 to 3 %). We first present the nanofabrication of the suspended SWCNT transistors whose channel lengths are $\approx$ 30 nm. The channels are strained by moving gold clamps holding firmly the nanotubes. We present detailed charge transport data, $dI/dV_{\text{B}} - V_{\text{B}} - V_{\text{G}}$ and $dI/dV_{\text{B}} - V_{\text{B}} - Δ\varepsilon_\text{mech}$, showing a large mechanical-gating effect of the SWCNT-QDs. The precise reversibility of the data, and their agreement with QTS theory, confirms that the tubes are strained elastically. We demonstrate that the mechanical control of the QD doping is not due to capacitive-gating effects, but to quantitatively predictable bandstructure changes including a strain-tunable bandgap. This precise mechanical control of the doping and bandgap of SWCNT-QDs could find applications in qubits, condensed matter physics, and homojunction molecular transistors.

Randomness amplification, the task of extracting uniform private bits from biased seeds that may be partly known by a malicious third party, is of central importance in cryptography. The highest security in this task is provided by a class of quantum protocols known as device-independent, which however are challenging to integrate into scalable devices. Semi-device-independent (SDI) protocols are a promising alternative that guarantees security under few natural assumptions, such as bounds on the amount of energy used by the devices. Here, we provide the first demonstration of SDI randomness amplification on an integrated silicon photonic chip, achieving a throughput rate of 20 Mbps suitable for practical applications. This rate is achieved through a novel technique for SDI entropy certification, which delivers strictly tighter von Neumann entropy bounds compared to existing methods and remains valid even if the preparation and measurement devices share quantum correlations. Overall, the methods developed in this work enable the integration of SDI technology into portable telecom devices, opening up a new generation of quantum cryptographic hardware.

Fiber optical cavities offer small mode volumes and correspondingly strong light-matter interactions in an open Fabry-Perot geometry. However, existing fabrication techniques do not reliably produce substrates with surface profiles amenable to high mode matching between the cavity mode and fiber core, thereby limiting the achievable collection efficiency. Here we present a technique to fabricate fiber mirror substrates while using $\textit{in situ}$ reflectometry to constrain the achievable mode matching prior to coating. By measuring the back-reflection from freshly cleaved fiber tips, we pre-select 138 fibers compatible with 96.5-99.5% mode matching, and after a single CO$_2$ laser ablation pulse, these fibers remained compatible with 95.3-99.2\%. This simple technique provides rapid feedback during each stage of substrate fabrication, greatly enhancing the yield of viable fiber mirror substrates prior to (expensive) coating runs.

In this chapter we review some results describing the high-frequency eigenmodes of the Laplacian on compact manifolds, or Euclidean domains, for which the geodesic flow is chaotic. We focus on the macroscopic distribution of these eigenmodes, which is described by the concept of semiclassical measure. The main result on the question is the Quantum Ergodicity theorem, originally due to Schnirelman. We provide the detailed proof of this theorem, including the adjustments necessary to treat the case of manifolds with boundary. We also discuss the Quantum Unique Ergodicity conjecture, and some progress towards this conjecture for strongly chaotic (Anosov) systems. In particular, we describe the constraints on admissible semiclassical measures, in terms of their Kolmogorov-Sinai entropy, as well as more recent delocalization results.

Any quantum computation consists of a sequence of unitary evolutions described by a finite set of Hamiltonians. For the case where this set consists of only products of Pauli operators, known as Pauli strings, we provide a necessary and sufficient condition for it to generate $\mathfrak{su}(2^n)$, i.e., to be universal for quantum computation on $n$ qubits. When combining Pauli strings with a general Hamiltonian, we show a sufficient (and in certain circumstances even necessary) condition for universality based on the Pauli-basis expansion of the Hamiltonian. As an application of these results, we prove two corollaries: (i) a necessary and sufficient condition for the universality of a general Hamiltonian given arbitrary single-qubit control on all qubits, and (ii) the universality of an XYZ Heisenberg Hamiltonian with local control of just two adjacent qubits.

In this work, we systematically study the gravitational lensing properties and observational signatures of compact boson stars. Unlike black holes, the photon effective potential of a compact boson star develops a nearly flat region, whose width increases with the compactness of the star. This flat structure significantly broadens the range of impact parameters that can produce large-angle deflections, leading to noticeably wider lensing rings of all orders. Photons constituting these rings traverse more complex paths, rendering the resulting images more sensitive to the spatial distribution of the accretion flow. Ray tracing results show that, compared to black hole models, the image topology and visibility amplitudes of compact boson stars exhibit a stronger dependence on the accretion flow structure. These results highlight qualitative differences in the observational properties of compact boson stars and black holes.

We solve exactly the relativistic finite-difference equation for the quantum three-dimensional ring-shaped Quesne oscillator potential. Our investigation is based on a finite-difference version of relativistic quantum mechanics. So-called relativistic configurational r-space is a key concept here. We show that the radial wavefunctions and angular wavefunctions are expressed through the continuous dual Hahn polynomials and Jacobi polynomials, respectively. A discrete energy spectrum has been found. The radial wave functions and energy spectrum have the correct nonrelativistic limit. We also build a dynamical symmetry group SU (1, 1) for the radial part of the equation of motion, which allows us to find the energy spectrum purely algebraically.

In quantum mechanics, a measurement applied to a state in general produces some amount of intrinsic randomness. This is not only a fundamental feature of the theory, but is also at the basis of any quantum process to generate random numbers. The simplest of such processes consists of a single, fully charaterized, measurement acting on a single, fully characterized, state. Unfortunately, no general method to estimate the intrinsic randomness produced in such setups is known. In this work, we address this issue by presenting a semidefinite programming formulation of the maximum probability with which an adversary, Eve, can guess the outcomes of characterized but untrusted prepare-and-measure setups. We then present several applications of this construction. First, we apply our method to a variety of specific setups, allowing us both to benchmark the approach and, more importantly, to determine the exact amount of certifiable randomness in scenarios where only upper bounds were previously available. Then, we show that the presence of entanglement between the device preparing the state and the measurement strictly increases Eve's predictive power, already in the most elementary setup of a binary measurement acting on a qubit state.

The harmonic oscillator is considered within generalized frameworks using complex and quaternionic numbers. The classical oscillator is considered in terms of a complex position function, and quantum oscillators are examined in terms of complex wave functions, and in terms of quaternionic wave functions as well. Both of the quantum solutions are obtained within the real Hilbert space formalism. The results reveal the complex and quaternionic descriptions as suitable frameworks for non-stationary processes, including damped oscillations, forced oscillations, and additionally self-interacting processes that cannot be appropriately described otherwise.

Entanglement is a key bottleneck limiting the efficiency of tensor-network and quantum simulations of molecular electronic structures. Here, we systematically assess and extend Clifford disentanglers as a structure-preserving approach to entanglement reduction: they can modify the entanglement structure of qubit wavefunctions while retaining the Pauli-string form of qubit Hamiltonians. To enable a practical search over Clifford transformations, we classify Clifford operators by their action on the Schmidt spectrum across a bipartition, reducing the two- and four-qubit search spaces to 20 and 91392 representatives, respectively. Embedded in an iterative Clifford-augmented matrix product state framework, these transformations reduce the energy errors at fixed bond dimension for the molecular test cases studied and mitigate the dependence on orbital orderings and fermion-to-qubit mappings. We further show that Clifford disentanglers can also benefit quantum simulations such as the shallow-circuit variational quantum eigensolver calculations. Together, these results establish Clifford disentanglers as a useful structure-preserving entanglement-engineering tool for tensor-network and quantum simulations of molecular electronic structure, while also clarifying their correlation dependence and motivating future developments.

We introduce a geometrically constrained hidden-variable framework that generates a family of correlations parametrized by a boundary function, within which the quantum singlet correlation appears as a particular member. Exact expressions for the correlation function are derived. Several structural results are established, including admissibility conditions, symmetry properties, a universal stationary point of the associated CHSH function, and an exact relation between the CHSH value at $ν=π/4$ and a geometric contrast measure defined on the underlying hidden-variable distributions. Rather than treating the quantum singlet correlation as an isolated target to be reproduced, the present framework places it within a broader geometric structure of correlations. These results suggest the existence of a nontrivial geometric structure underlying the family of correlations and motivate the search for a principle capable of selecting the quantum singlet solution from within that family.

Characterizing quantum processes is essential for hardware benchmarking, error diagnosis, and algorithm verification. While recent work [PRX QUANTUM \textbf{4}, 040337 (2023)] extended classical shadows from quantum state to quantum process, enabling efficient single-channel $\mathcal{E}$ property prediction, its applicability to composite processes $f(\mathcal{E}_1, \mathcal{E}_2,\cdots, \mathcal{E}_k)$ remains unexplored. We introduce shadow engineering, a framework encoding the classical shadows of processes into sparse transfer matrices to predict $f(\mathcal{E}_1, \mathcal{E}_2,\cdots, \mathcal{E}_k)$ properties with proven polynomial sample complexity, matching single-channel efficiency while exponentially lower than quantum process tomography. Crucially, this approach repurposes existing $\mathcal{E}_m$-shadow data without physical execution of $f(\mathcal{E}_1, \mathcal{E}_2,\cdots, \mathcal{E}_k)$, enabling flexible quantum process characterization with minimal hardware overhead. We demonstrate the framework's effectiveness and practicality on a superconducting quantum processor for typical applications such as error mitigation and Hamiltonian dynamical simulation. This framework unlocks new capabilities for predicting complex quantum behaviors without physical re-execution, with immediate applications in near-term device calibration and quantum simulation.

The central prerequisite of any fault-tolerant quantum architecture is a quantum memory: a block of encoded physical qubits whose logical state is actively preserved against noise across many rounds of error correction. In neutral-atom Rydberg arrays, realizing such a memory is obstructed not by the entangling gates themselves, which are already fast and high-fidelity, but by the auxiliary operations that a conventional error-correction cycle requires: mid-circuit fluorescence measurement, inter-zone atom transport, and locally focused single-qubit addressing. Each of these introduces latency, atom loss, or optical crosstalk that exceeds the cost of the underlying gates by orders of magnitude. These costs accumulate cycle after cycle, progressively degrading the very logical information the code is meant to protect. Here we propose a protocol that stabilizes a toric-code quantum memory without moving, measuring or local addressing atoms. The key is to use a three-species Rydberg atom array for the complete stabilizer cycle, including syndrome extraction, coherent correction, and ancilla reset, under global, species-selective laser pulses. Numerical simulation of a $4 \times 4$ rotated toric code shows a longer qubit lifetime when the physical error rate is below a pseudo-threshold $p^\star \approx 0.034$. The scheme offers a concrete, hardware-efficient route to topological quantum memory in neutral-atom platforms.

The quantum information lost in open evolutions cannot be fully recovered, but partial recovery is possible. The Petz recovery map guarantees almost optimal recovery, notably if the chosen reference state is close to the real one. This map has been widely used in theoretical studies, but has been the object of only a handful of experimental realisations, typically under a single fixed noise model. In this work, we describe and implement the Petz recovery map for a versatile class of qubit channels with tunable decoherence and dissipation. The setup we realize is also the first experimental example of ``tabletop reversibility'': for a good range of choices of the reference state, the Petz recovery map can be implemented with the same devices as the forward dissipative evolution, whose effect it is partially undoing. Our results demonstrate that the Petz recovery map can be resource-efficiently realized without requiring complex ancillary resources, providing a feasible pathway for mitigating information loss in quantum systems.

Quantum systems face a fundamental trade-off: they must remain decoupled from the environment to maintain long coherence times, yet they require interactions with the environment to be accessible for measurement. As a prime example, emitter arrays coupled to waveguides facilitate collective modes that, owing to interference, can suppress radiation into the waveguide. While complete destructive interference creates perfectly dark states with infinite lifetimes, their inherent decoupling makes them unmeasurable in standard waveguide quantum electrodynamics. Consequently, current approaches must rely on system non-idealities that permit measurement but limit the coherence times. In this work, we lift this limitation by proposing the use of weakly squeezed light generated in \{chi}(2) nonlinear waveguides for the spectroscopy of completely dark states. We show that the fluorescence spectrum probes transitions between the dressed dark states of the emitter array. This work paves the way towards the measurement and control of dark states, with applications for robust quantum memories, computation, and communication.

We explore an array of quantum emitters as non-equilibrium probes, coupled to a one-dimensional photonic waveguide, aiming to estimate its properties such as wave number which encodes the waveguide frequency and dispersive characteristics. By considering transient dynamics following initial excitation, we show that the quantum Fisher information (QFI) can be significantly enhanced through careful emitter positioning. For two-emitter probes, optimal spacing stabilizes populations and coherences in the single-excitation subspace, suppressing super radiant decay and extending both the magnitude and longevity of QFI. Randomized emitter configurations also reveal that vanishing waveguide-mediated cross decay maximizes both achievable sensitivity and the temporal duration over which information about the parameter remains accessible. Extending to multipartite probes, we demonstrate that the maximum QFI and its temporal integral scale with system size, exceeding the Heisenberg limit for all positioning strategies. Our results highlight the potential of waveguide-coupled emitter arrays as versatile quantum sensors, where collective radiative dynamics can be harnessed to achieve tunable, long-lived, and enhanced precision.

Spherically symmetric, general relativistic Bondi accretion is known as the Michel flow. The stationary integral transonic solutions for the Michel flow has been constructed for multi-component accretion described by an equation of state where the adiabatic index varies with the radial distance along which the streamlines are studied, and the corresponding phase portrait spanned by such radial distance and the flow Mach number has been obtained. Borrowing the techniques used in the dynamical systems theory, the nature of the transonic points of the aforementioned flow has been classified. The steady state flow has been perturbed to study the stability of the stationary solutions, and it has been found that such flows are stable under the (linear) radial perturbation. As a consequence of the stability analysis, the corresponding acoustic space time embedded within the accreting matter has been obtained, and the horizon of the metric of such sonic space time has been identified by constructing the causal structure with the help of the Carter-Penrose diagrams. In this way, the accreting black hole systems in the general relativistic set up has been investigated from various different perspectives - from its astrophysical aspects, from the dynamical systems point of view, as well as within the realm of the classical analogue gravity phenomena.

We investigate the critical regime $κ(R,T)=0$ in $κ(R,T)$ gravity. While most studies assume a non-vanishing effective gravitational coupling, the existence of critical hypersurfaces where $κ$ vanishes is a generic feature of many admissible coupling functions. We show that the apparent singularity of the non-conservation equation is an artifact of a rewritten form of the conservation law and that the fundamental equations remain regular at $κ=0$. We further analyze the structure of critical hypersurfaces, derive the associated compatibility condition $(\nabla^μκ)T_{μν}=0$, and discuss their interpretation as gravitational screening surfaces separating attractive and repulsive gravitational phases. The existence of critical coupling hypersurfaces also obstructs a global Einstein-frame description, distinguishing $κ(R,T)$ gravity from theories based solely on algebraic redefinitions of the energy-momentum tensor. Possible cosmological and astrophysical consequences are briefly explored.

We study the effects of dimension-four and dimension-six nonminimal Higgs couplings to the Ricci scalar $R$ in the $R^2$-Higgs inflation model in light of the recent ACT and SPT observations. We show that the dimension-six operators $|Φ|^2 R^2$ and $|Φ|^4 R$ can accommodate the enhanced scalar spectral index $n_s$ preferred by the combined CMB+BAO analyses. Using a doubly covariant formalism, we find that the same region of parameter space that explains the observed value of $n_s$ can also induce rapid preheating through the production of the Goldstone modes. If thermalization proceeds efficiently through this preheating mechanism, it may help match the inflationary scale with the CMB reference scale.

The Traveling Salesman Problem (TSP) is a classical NP-hard problem in combinatorial optimization, where determining the shortest route among a set of cities becomes computationally prohibitive as the problem size increases. This work explores quantum computing as an alternative approach to address this complexity. Unlike existing methods that primarily rely on quantum annealing, we propose a quantum iterative framework integrating Quantum Phase Estimation (QPE) and Grover's search algorithm. Route costs are encoded as quantum phases, enabling QPE to efficiently evaluate them, while Amplitude Amplification, implemented via the Grover-Long algorithm, iteratively refines the solution space toward the optimal route. A proof-of-concept case study on a small-scale TSP instance demonstrates the feasibility of this approach and its potential for scaling to larger optimization problems. Furthermore, under an expectation-based analysis, the algorithm exhibits an expected computational complexity of $O(\frac{m^2\log_2(m)\log_2(1/ε)}{\sqrtε})$ which depends on the error tolerance parameter $ε$. This estimation omits the initialization term, which we expect future refinements to render subdominant to Phase Estimation.

We study statistics of atypical measurement outcomes in the steady states of driven open quantum systems. In equilibrium, the probability distribution over the phase space, as encoded in, e.g., the Wigner function, is analytic in the phase-space coordinates. We show that this property is generically lost in driven dissipative systems: their {\it large-deviation function} develops lines and surfaces across which its derivatives are discontinuous. As an illustrative example, we consider a parametrically driven Kerr oscillator coupled linearly and/or nonlinearly to a dissipative bath. Rare fluctuations in the amplitude and phase of the induced oscillations are governed by semiclassical instanton trajectories of the corresponding Keldysh-Lindblad action. We demonstrate that a given fluctuation can be realized through multiple distinct instanton trajectories. The competition between these trajectories leads to abrupt switching of the dominant instanton and, consequently, to non-analytic features in the large-deviation function.

We establish a rigorous mathematical bridge from quantum many-body path integrals to the cusp catastrophe model by Lyapunov-Schmidt reduction, which provides a theoretical foundation for analyzing superconducting phase transition using the catastrophe theory. First, it is proved that, near the critical point the infinite-dimensional effective action is diffeomorphic to a finite-dimensional catastrophe. Secondly, starting from Ginzburg-Landau free energy functional, the Euler-Lagrange partial differential equation can be reduced to the cusp catastrophe model. Thirdly, the fermionic imaginary-time path integral to the cusp catastrophe is derived through the Hubbard-Stratonovich transformation, Matsubara frequency expansion, and Grassmann algebra. Furthermore, we connect this framework with the adsorption potential theory we proposed, elucidating the catastrophic topological nature of the electron pairing mechanism in high-temperature superconductivity. The precise microscopic derivation of the adsorption potential from first-principles electronic structure calculations would strengthen the predictive power of the theory.

In closed quantum systems, local information spreads throughout the entire system and becomes highly complex under unitary evolution. In contrast, when the system is embedded in an environment, system-environment coupling can transfer information from the system into the environment, thereby reducing the rate of complexity growth within the system. This leads to the environment-induced scrambling transition established in previous works. In this work, we identify entropy injection from the environment as a different physical process that instead enhances many-body chaos. Our setup consists of coupling a system that is already in equilibrium with one environment to another environment, which serves as an entropy reservoir and drives the system into a non-equilibrium state. When entropy flows into the system through either heat transfer or particle transfer, the effective Hilbert space explored by the system enlarges, a mechanism that can enhance many-body chaos. We explicitly demonstrate this idea by constructing a solvable complex Brownian SYK model, in which both the relaxation toward the steady state and the steady-state quantum Lyapunov exponent can be computed analytically. Our results provide a controllable mechanism for tuning quantum scrambling through entropy flow in quantum many-body systems coupled to environments.

Grover's algorithm achieves a quadratic speedup for unstructured search given a global oracle for the target set. In many applications, however, the target set is specified as the intersection of multiple constraint sets. Constructing a global oracle for the intersection can be costly, whereas the individual constraint oracles are often much simpler to implement. We study a randomized Grover search algorithm that directly uses these constraint oracles. At each iteration, one of the corresponding Grover operators is selected at random. For the two-operator case with uniform sampling, we prove that the success probability approaches one after \[ Θ\left(\frac\pi4\sqrt{\frac{N}{r}}\right) \] iterations, where $r$ is the size of the intersection. Thus, the algorithm achieves the same asymptotic query complexity as standard Grover search but without requiring a global oracle. We then generalize the analysis to arbitrary sampling distributions and an arbitrary number of Grover operators through an auxiliary operator that approximates the expected Grover evolution, while retaining the same asymptotic complexity. We further show that highly biased sampling distributions can still achieve near-unit success probability, enabling cheaper Grover operators to be used more frequently. Finally, we prove asymptotic optimality and support the theoretical results with numerical simulations.

In this paper, we investigate energy extraction via the Comisso-Asenjo (CA) magnetic reconnection process near a coupled $\mathcal{N}=2,\,U(1)^2$ gauged supergravity Black Hole (BH). Our study focuses on the combined impact of the independent parameter set $p_i\in(N_g,g,v,e)$ with the spin parameter $a$ on the extracted energy ($ε_{\pm}$), efficiency ($η$), and extracted power ($\mathcal{P}_{CA}$), aiming to identify optimal combinations where energy can be extracted with higher efficiency in certain cases at lower spin $(a\sim0.39)$ than the Kerr extremal case $(a\sim1)$. Using the spacetime parameters, we explore various cases leading to distinct spacetimes and provide an extended comparison with the Kerr Black Hole (KBH). We also examine the influence of the orientation angle ($ξ$) and magnetization parameter ($σ_0$) on both efficiency and extracted power. Investigating low $[\,\forall p_i<0.2 \land N_g<0.08\,]$, mid $[\,\exists p_i\ge0.5 \land N_g\in(0.08,0.15)\,]$, high $[\,\exists p_i>0.7 \land N_g\in(0.16,0.23)\,]$, and mixed $[\,\forall p_i\in(0,1) \land N_g\in(0,0.23)\,]$ parameter combinations, we explore only extremal cases for all spacetime parameters and demonstrate that the extremal Kerr efficiency limit ($η>1.495$) can be exceeded. The statistical Kendall's Tau approach allows us to identify the key independent parameters acting as boosters or dampers in the energy extraction process and to visualize the relationship between $(N_g,g,v,e)$ and the physical outputs $(a_{\rm ext},r_E,r_{\rm ergo},ε_{\pm},η,\mathcal{P}_{CA},R_η,R_{\mathcal{P}})$. Furthermore, we show that the observable Lundquist number $S_{\rm obs}$ in rotating BH spacetimes acquires an observer-dependent angular dependence through the lapse function $(α)$. This leads to deviations from the standard Sweet-Parker scaling when expressed in terms of observable quantities.

We study the dynamics of quantum correlations in the hydrogen hyperfine spin system subject to Markovian phase noise. Treating the electron and proton spin degrees of freedom as an open two-qubit system governed by an isotropic hyperfine Hamiltonian and local dephasing, we obtain the exact time-dependent density matrix and derive analytical expressions for the full X-state family. We compute concurrence($C$), trace-distance measurement-induced nonlocality (Trace MIN--$\mathcal{N}_1$), and average steering coherence (ASC) in closed form and establish their strict ordering $ C(t)\leq \mathcal{N}_1(t)\leq \mathrm{ASC}(t) $ at all times. Entanglement is identified as the most fragile resource, undergoing sudden death at a finite time. Trace MIN exhibits dephasing-immune freezing for states with nonzero population imbalance, while ASC is the most robust quantity, persisting longest in every scenario studied.We additionally demonstrate that the dephased thermal hyperfine state serves as a resource for quantum teleportation, deriving a closed-form expression for the average fidelity and establishing that the teleportation advantage window coincides exactly with the entanglement survival interval, $\mathcal{F}_A > 2/3 \Longleftrightarrow \mathcal{C} > 0$, for the full X-state family with maximally mixed marginals. We identify four distinct dynamical regimes and map all three correlation measures onto directly measurable Pauli spin correlators, enabling experimental reconstruction of the full hierarchy without full state tomography.

Gravitational waves from extreme-mass-ratio inspirals (EMRI) provide a direct probe of the strong-field geometry of black holes. Motivated by this, we study the motion of test particles and the resulting gravitational wave emission in the spacetime of a charged black hole inspired by loop quantum gravity (LQG), where the classical singularity is replaced by a smooth transition surface arising from the LQG polymerization, in which its radius is set by the LQG area gap condition. As a result, the polymerization parameter $δ_b$ is uniquely determined by the mass $M$ and charge parameter $Q$, so that all cases examined in this work contain LQG correction. By constructing the effective potential, the innermost stable circular orbit (ISCO) and the marginally bound orbit (MBO) are determined. Periodic orbits are classified using the Levin-Perez-Giz zoom-whirl taxonomy, showing how the orbit topology shapes the waveform, so that each closed trajectory is labeled by the triple integer $(z, w, v)$ and located through the rational frequency ratio $q = ω_φ/ω_r - 1$. Within the quadrupole approximation, the gravitational waveforms for an EMRIs are estimated, and the resulting polarizations are obtained in the time-domain and frequency-domain. The resulting polarizations in the time-domain exhibit a zoom-whirl morphology, with the waveform amplitude and phase dependent on the LQG parameter. The characteristic strain peaks in the millihertz band for all values of the charge parameter $Q$, and they exceed the projected sensitivities of LISA, Taiji, and TianQin, suggesting that future observations could place meaningful constraints on the LQG polymerization parameter in the strong-field regime.

During the evolution of a binary black hole, energy and angular momentum are exchanged between the orbital motion and the individual black holes through horizon absorption, modifying both the binary dynamics and the black hole masses and spins. This leaves an imprint on the emitted gravitational waves that may be relevant for the accurate modeling of signals observed by current and future detectors, while also offering a probe of the nature of compact objects. In this work, we derive, for the first time and at leading order in the post-Newtonian expansion, the effect of horizon absorption in binary black hole inspirals with both orbital eccentricity and spin-induced precession, and we incorporate these corrections into the pyEFPEHM waveform model. We then quantify their impact through analytical estimates of the orbital dephasing, waveform mismatches, and Bayesian parameter-estimation studies. The effect is largest for systems with large spin components (anti-)aligned with the orbital angular momentum ($|\vecχ_i \cdot \hat{l}| \sim 1$), highly unequal mass ratios ($q=m_2/m_1 \ll 1$), and long inspirals spanning a wide frequency range ($\log(f_\mathrm{max}/f_\mathrm{min}) \gg 1$). For such systems, neglecting horizon absorption biases the recovered binary parameters at moderate signal-to-noise ratios. In quasi-circular binaries these biases largely absorb the effect, rendering it difficult to detect. In eccentric binaries, however, the richer signal morphology breaks this degeneracy, making horizon absorption potentially measurable in high signal-to-noise-ratio events.

The appeal of adiabatic quantum computing lies in its intrinsic robustness against various technical imperfections, making it attractive for many quantum information applications. However, it faces a fundamental challenge: accelerating the adiabatic operations while preserving adiabaticity within the qubit coherence time. In this article, we propose an electromagnetically induced transparency-based adiabatic CNOT gate protocol which harnesses atomic hyperfine intermediate states (HISs) to speed up the adiabatic evolution. The HISs, naturally-existed in two-photon transitions, often need to be suppressed due to their significant decay errors. In contrast, this paper introduces a novel method that utilizes appropriately chosen HISs not only to enhance the adiabaticity in STAY pathway but also to accelerate the population transfer in TRANSFER pathway. Through pulse optimization, we achieve adiabatic gate fidelities exceeding 0.9991 within 0.3903 μs in realistic Cs atomic setups. To demonstrate the generality of protocol we further assess the impact of decays from multiple HIS and extend our model to arbitrary number of states, providing a practical route toward fast and robust adiabatic quantum gates in Rydberg-atom platforms.

Quantum cloud platforms are poised to deliver powerful computing capabilities, but users have no direct means to verify which physical device executes their workload. This lack of transparency enables hardware substitution attacks, where a malicious adversary could redirect a job to a substituted or inferior processor. We present a general authentication framework that addresses this problem by constructing multi-dimensional quantum fingerprints from raw measurement data. Without any curve fitting, we directly concatenate the raw statistics of complementary experiments into a high-dimensional feature vector that preserves subtle device-specific information. A Mahalanobis nearest-neighbor classifier achieves 100\% benign authentication accuracy on three superconducting processors over a three-week chronological split. The classifier naturally yields an authentication confidence $C_{\mathrm{claimed}}$ which reveals device-specific safety margins and motivates per-device alert thresholds. We assess the framework's robustness under two distinct scenarios. Under additive isotropic Gaussian noise, $C_{\mathrm{claimed}}$ decays predictably at a rate explained by inverse covariance traces, enabling an early warning mechanism. Against white-box adversarial perturbations, the same confidence threshold detects $L_2$ targeted attacks with near-perfect success and reveals device-dependent empirical thresholds for $L_\infty$ attacks, while untargeted and sparse attacks are ineffective. The proposed framework thus unifies fingerprint extraction, drift-resilient authentication, proactive health monitoring, and adversarial defense, offering a practical step toward trustworthy quantum cloud computing.

We investigate the atomic transition probabilities in atom-mirror set-ups that are in circular motion. In one scenario, the atom is in circular motion inside a static cylindrical mirror. In the other scenario, the cylindrical mirror rotates around its central axis while the atom remains static. We report structural similarity in the atomic transition probabilities between these two cases -- these probabilities are equivalent upon interchanging the field frequencies between the two scenarios. We interpret such an observation as a semi-classical phenomenon analogous to the classical Mach's principle.

Gravitational-wave interferometers exhibit a wide variety of short-duration non-Gaussian transients, commonly referred to as glitches, that complicate the detection of astrophysical signals, bias parameter estimation, and detector characterisation. Existing machine-learning approaches classify glitch morphologies but do not provide a complete mechanism to segment and extract these disturbances from the strain data. We introduce a wavelet-based, saliency-guided framework for the supervised extraction of transient noise. Candidates are first pre-tagged using Uniform Manifold Approximation and Projection, which is also used as a diagnostic of the learned representations. A traditional learning model operating on Continuous Wavelet Transform spectrograms then identifies relevant time-frequency regions through saliency maps. These saliency patterns are transferred to an invertible multiresolution representation via the Discrete Wavelet Transform, where adaptive coefficient masking enables exact reconstruction of both glitch-only and glitch-suppressed waveforms. We demonstrate effective extraction across several representative glitch families, including 'Whistle' and 'Scattered-Light' transients, and show robustness in challenging regimes such as low signal-to-noise events and partially overlapping structures, where classical thresholding or band-limited filtering methods typically fail or introduce leakage. The proposed framework offers an interpretable and computationally efficient approach to transient-noise extraction, establishing a foundation for scalable applications to larger glitch catalogs and future observing runs.

Quantum batteries provide a useful setting for exploring nonequilibrium many-body effects in energy storage. Here we investigate the optimization of a quantum battery based on the central-spin model. We identify two complementary structural indicators associated with the effective charging dynamics: one yields an upper bound on the average charging power, while the other characterizes the buildup of stored energy. We show that these two indicators are jointly optimized at a distinguished initial charger excitation number, which selects a particular Dicke sector of the model. At this common optimal point, the effective charging Hamiltonian becomes exactly mirror symmetric, suggesting mirror symmetry as a useful structural indicator for optimizing quantum batteries. We further show that the corresponding optimal dynamics can be closely approximated by product initial states, in particular by spin coherent states whose excitation-number distribution is centered at the symmetry-selected point. Our results establish a direct connection between charging performance, optimal-state structure, and emergent symmetry in the central-spin quantum battery, and suggest symmetry as a useful organizing principle for efficient charging in interacting many-body quantum systems.

The Grover mixer quantum alternating operator ansatz (GM-QAOA) employs the Grover mixer to confine the quantum evolution to the feasible subspace defined by the problem. Its mixing unitary, however, requires a global multi-controlled phase-shift gate acting on all qubits, resulting in substantial circuit overhead on near-term quantum devices. In this work, we propose locally acting Grover mixers tailored to initial states that admit a product structure over disjoint qubit subsystems, which may be obtained by encoding only a subset of problem constraints into the initial state preparation. The proposed method preserves the search space defined by the initial state while significantly lowering implementation cost, as the global multi-controlled phase-shift gate is replaced with local operations on disjoint subsystems. Numerical simulations on the exact-cover problem and the traveling salesman problem (TSP) demonstrate that the proposed method achieves convergence behavior comparable to that of the original GM-QAOA, while using shallower circuits with fewer gates. We further compare two constraint encoding strategies for the TSP, encoding only a subset of constraints versus all constraints into the initial state preparation, and show that the former combined with the proposed mixer yields markedly more compact circuits at the point where comparable solution quality is achieved.

Non-invertible symmetries in quantum many-body systems generally give rise to sequential unitary circuits that move symmetry defects. In this paper, we investigate invariants defined by sequences of such circuits, which move non-invertible defects and generate a Berry phase evaluated on quantum states with defects. We show that this Berry phase generally defines an invariant under local deformations, provided that the sequential circuits preserve the locality of those deformations. This invariant also rules out a short-range-entangled state that preserves the non-invertible symmetry, thereby signaling the 't Hooft anomaly of a non-invertible symmetry purely in terms of unitary operators acting on a state. We then apply this framework to loop excitations in three spatial dimensions and identify a new loop excitation in the (3+1)D $\mathbb{D}_4$ topological order, which we dub a non-Abelian fermionic loop. Using the invariant of sequential circuits, we characterize the statistics of non-Abelian fermionic loops. In addition, we find a new (3+1)D mixed topological order with a single non-Abelian fermionic loop, whose long-range entanglement is protected by an invariant of sequential circuits.

Single-photon optical nonlinearities face a fundamental trade-off: maximum nonlinearity requires both spectral resonance (narrow bandwidth) and high peak intensity (short duration), constraints that are incompatible due to the time-energy uncertainty relation. We demonstrate experimentally that this limitation does not need to exist in cases involving post-selection. We measure a cross-phase shift (XPS) produced by a resonant photon from a narrow-band source that is first transmitted through a cold atomic cloud and then localized in time through detection. The peak size of this XPS is greatly enhanced compared to that of Gaussian single-photon-level pulses without post-selection, benefiting from the narrow bandwidth of the resonant prepared state and the high intensity of the post-selected state simultaneously. We measure enhancements in the peak XPS of 6$\pm$1 at an optical depth (OD) of 2.4$\pm$0.1, and our results are in qualitative agreement across a range of optical depths with the recently developed weak value theory of atomic excitation [Thompson et al., APL Quantum 2, 036108 (2025)] for such post-selected photons. This work uncovers new consequences of having simultaneous knowledge of frequency and time, raising new foundational questions about how a particle behaves, and interacts with other systems, when its preparation and post-selection are non-commuting.

Accurate calibration of the gravitational-wave strain h(t) is essential for both detection and astrophysical inference. In operating detectors, slow temporal variations in the sensing response are tracked using calibration lines, but practical constraints can prevent those lines from being injected at frequencies that are favorable for precise estimation of sensing-side parameters. We present a statistical framework for preselecting evaluation frequencies under such constraints. We apply this framework to KAGRA data from the first part of the fourth LIGO-Virgo-KAGRA Observing Run, for which the nominal cavity-pole frequency was about 18 Hz, while the sensing-side calibration line used in practice was injected at 32.7 Hz. For each candidate evaluation frequency, we construct the sensing function, quantify its segment-wise statistical uncertainty from empirical percentiles of the sample distribution, and rank the candidates using a score that combines the interval widths of the amplitude and phase. When a 1% amplitude interval width and a 1 degree phase interval width are weighted equally, 244 Hz is selected in all 4096 s analysis segments throughout the analyzed period. Relative to the reference frequency of 32.7 Hz, the amplitude interval width is reduced to about one quarter over a broad frequency range, while the phase interval width remains broadly comparable. We also assess the discrepancy introduced by frequency translation separately. These results suggest that the proposed method provides a useful statistical preselection framework for evaluation frequencies under practical operational constraints.

Decoders for quantum error correction (QEC) experiments rely on detector error models (DEMs), which encode, for each error, its probability and the detectors and logical observables it flips. Here we show that estimating DEM event probabilities from experimental syndromes is feasible, avoids independent device benchmarking, and produces useful decoder priors for estimating and reducing decoded logical error probabilities. We evaluate our methods using open-source data from surface-code memory experiments performed on Google's Willow chip, and we carry out analogous surface-code experiments on IBM's \texttt{ibm\_miami} processor. Despite the different physical error scales of the Google and IBM devices, in both cases our estimated DEMs improve logical error probabilities relative to baseline device-informed DEMs, typically at the $5\%-10\%$ level and with larger gains in some IBM cases, without additional calibration circuits, decoder fine-tuning, or supervised fitting to logical outcomes.

Mechanical strain is a powerful resource for tuning quantum systems, but existing piezoelectric strain cells are generally optimized for fragile, high-aspect-ratio single crystals rather than the thick, square-profile chips typical of semiconductor quantum devices. Furthermore, adapting these cells for qubits requires accommodating dense RF and DC wiring while maintaining strict electrical isolation from high-voltage piezo actuators. Here, we present a piezoelectric uniaxial strain cell designed to homogeneously strain thick, square-profile substrates. We introduce a highly symmetric dual-chip loading configuration that effectively suppresses flexural deformation and shear stress. The cell integrates a high-density RF/DC interposer to support standard wire bonding and encloses the actuators in a grounded Faraday cage to prevent unwanted Stark shifts in the device layer. Finite element simulations confirm that combining stiff actuators with this symmetric mounting drastically improves strain homogeneity. Finally, we validate the apparatus experimentally by applying uniaxial strain to a 200 $μ$m thick silicon die. Surface strain measurements demonstrate an applied strain of 215 $με$ for 200 V applied piezo bias.

Barcode scans, clear phone calls, reliable data storage, satellite communication, and large-scale quantum computation are all made possible by error correction. We present a handbook version of The Error Correction Zoo, a curated reference of methods for protecting classical or quantum information from errors during storage and transmission. The handbook includes descriptions of these error-correcting codes and a classification according to the symbols they use. It also catalogues relations among codes and related objects such as sphere packings, lattices, designs, groups, and classical and quantum phases of matter. The collection is intended both as a rigorous reference and as a practical aid for tracing the web of code relationships and uncovering new connections.

Efficiently decomposing finite element stiffness matrices into the Pauli basis is challenging due to the exponential growth of Pauli strings with problem size. A naive Pauli expansion requires $Θ(8^{\lceil \log_2 N \rceil})$ operations, where $N$ denotes the number of degrees of freedom, rendering direct decomposition infeasible for large systems. Existing approaches exploit algebraic sparsity or operator structure but do not incorporate the geometric organization intrinsic to finite element discretizations, and consequently exhibit poor scaling for stiffness matrices. To address this problem, we introduce PHASE, a hierarchical, geometry-aware Pauli decomposition algorithm that leverages recursive mesh partitioning to organize element contributions across multiple spatial scales. PHASE employs a hybrid strategy that combines full- and reduced-space Tensorized Pauli Decomposition with Fast Walsh-Hadamard Transform-based aggregation to assemble global Pauli coefficients efficiently. We show that this approach yields a dimension-dependent reduction in the exponential scaling exponent of Pauli assembly asymptotic complexity relative to existing methods, reducing the cost from $2^{2{\lceil \log_2 N \rceil}}$ to $2^{γ_d{\lceil \log_2 N \rceil}}$ with $γ_d < 2$ under standard mesh regularity and balanced partition assumptions. These results substantially improve the feasibility of quantum-compatible operator synthesis for large-scale finite element models.

The Linear Combination of Hamiltonian Simulation (LCHS) framework simulates dissipative linear dynamics by representing time evolution as an integral over unitary operators, which is discretized by quadrature and implemented via Hamiltonian simulation. While existing analyses achieve near-optimal scaling in time and precision using norm-based quantities of the dissipative generator, we show that implementing the Hamiltonian simulation steps with Multi-Product Formulas (MPFs) yields commutator-sensitive error and complexity bounds. We demonstrate that the quadrature rule affects not only discretization error but also commutator structure and query complexity. This dependence is quantified through post-quadrature analysis for abstract MPF error profiles and for general time-independent and local Hamiltonians using known commutator-sensitive MPF error estimates. We compare uniform trapezoidal and free-scale sinh--sinh quadrature, showing improved quadrature-cardinality scaling for the latter, and illustrate the framework with applications to fractional diffusion, advection--diffusion, and open quantum systems.

Entanglement-assisted (EA) quantum QC-LDPC codes offer strong error-correction capabilities with structured parity-check matrices, but their practical use depends on efficient encoder circuits and the availability of pre-shared Bell pairs (ebits). In all encoder implementations based on the stabilizer formalism, the dominant contribution to this complexity comes from the use of controlled gates. In this paper, we adopt the Sharma-Kumar-Garani (SKG) encoder construction. We formulate the encoder optimization as a search over GF(2) row operations that decompose the binary matrix derived from its CNOT sub-sequence. We solve this problem using a beam search algorithm guided by a Hamming-distance heuristic. For the tested EA quantum QC-LDPC code families, the proposed method achieves CNOT-count reductions of 7.3-34.0% relative to the SKG baseline encoder. The optimized circuits also yield lower CNOT counts than Patel-Markov-Hayes synthesis on all tested instances and are verified by stabilizer-tableau simulation. These results show that substantial encoder simplification is possible for structured EA QC-LDPC codes.

We introduce a construction of Pauli Hamiltonians with exactly known ground-state energies, intended as reference instances for ground-state energy estimation algorithms. The construction embeds a planted block-product state as the simultaneous ground state of a sum of frustration-free local clauses on overlapping supports, exposes the resulting model only as a polynomial-size linear combination of Pauli operators, and admits optional Clifford conjugation that preserves the spectrum. The framework subsumes classical planted constraint-satisfaction problems as a diagonal special case, providing a direct embedding channel through which classical hardness properties can be inherited. Open-source software, certification keys, and example instances are made publicly available.

We perform a Hamiltonian constraint analysis of metric $f(R)$ gravity in the Jordan frame and show that the regular constraint classification degenerates on singular phase-space surfaces located at $f'(R)\!=\!0$ and $f''(R)\!=\!0$. We then study the perturbative implications of these surfaces. For exact backgrounds satisfying $f(R)\!=\!0$ and $f'(R)\!=\!0$, the linearized spectrum is empty; the known pure $R^2$ result is therefore a special case of a more general degeneracy in $f(R)$ gravity. We also show that FLRW trajectories in the Starobinsky model can cross the surface $f'(R)=0$, but that inhomogeneous perturbations develop a degenerate constraint structure at the crossing. The resulting crossing condition is better interpreted as a regularity condition for perturbative evolution than as an ordinary constraint within the Dirac--Bergmann algorithm. Together, these results distinguish backgrounds that lie entirely on a singular surface from backgrounds that cross one dynamically, and show that the two situations lead to different perturbative degeneracies.

We show that isotropic random walks on the complex projective space provide a canonical and analytically tractable stochastic-geometric framework for the exploration of quantum-state space. The approach combines harmonic analysis on compact rank-one symmetric spaces with stochastic pure-state evolution and yields explicit analytical expressions for transition kernels, fidelity statistics, and geometric observables associated with the Fubini--Study metric. In particular, the framework provides a solvable reference model for isotropic depolarization and Haar equilibration, reproducing Haar-random fidelity statistics and the invariant measure on projective Hilbert space without specifying a microscopic Lindblad generator. In the short-time regime, the stochastic evolution converges to Brownian diffusion generated by the Fubini--Study Laplace--Beltrami operator, while the long-time limit exhibits concentration-of-measure behaviour characteristic of high-dimensional random quantum states. We further derive analytical and asymptotic results for the first-passage-time problem, including closed-form expressions in the Brownian limit for the mean first passage time and the long-time tail of the first-passage-time distribution. For high-fidelity target states, the mean first passage time exhibits a strong dimension-dependent divergence originating from the concentration properties of the Fubini--Study geometry.

We recast Einstein's equations as ordinary integro-differential equations for the worldlines, integrating out the gravitational field by means of the Schwinger-Keldysh path integral. The same framework allows the gravitational waveform to be computed for unspecified orbits. The two computations are independent: solutions of the equations of motion can then be inserted to reconstruct the waveform for generic orbits. The derivation of the equations of motion does not require a map between scattering and bound-orbit observables. Thus, it could be implemented within an Effective One-Body-inspired framework, with the advantage that retardation and radiation effects are automatically included: no separation between potential and radiation modes is required. Conversely, the waveform computation may provide an alternative to the Effective One-Body approach, if supplemented by suitable resummation schemes. We emphasise that computations in this framework bypass the need for integration-by-parts identities, which are the main technical bottleneck in the computation of observables. In this paper, we outline the general framework and present a computational strategy at leading and next-to-leading order in the weak-field expansion.

Wormholes that are traversable in principle offer fascinating insights into general relativity, yet they typically require exotic matter and suffer from stability issues. We construct a thin-shell wormhole by gluing two copies of a quantum-corrected, regular spacetime obtained from string T-duality. This regularisation replaces the classical curvature singularity with a smooth core and introduces a fundamental length scale $l_0$. For the static configuration, we derive the surface stresses and show that, unlike the Schwarzschild case, the null and strong energy conditions can be satisfied for sufficiently large throat radii. A linearised stability analysis reveals a rich landscape: close to the minimum allowed throat radius the configuration is unstable; at intermediate radii ($a \sim l_0$) the geometric stability threshold becomes negative, yielding a window of \emph{unconditional stability} where any convex surface mass function suffices; at large radii the wormhole recovers Schwarzschild-like behaviour and stability requires a stiff equation of state. The T-duality scale $l_0$ is thus not merely a regulariser but a key physical parameter that opens a novel region of unconditional stability absent in classical thin-shell wormholes. Our results suggest that quantum-gravity-motivated modifications can simultaneously cure singularities and make traversable wormholes dynamically viable, providing new targets for gravitational-wave astronomy and theoretical studies of exotic compact objects.

Rolling stock planning is a complex optimization problem in railway management that involves assigning physical trains to scheduled trips while minimizing operational costs. In this work, we address a specific instance of this problem featuring 190 trips over two days, subject to constraints such as mandatory maintenance stops. We reformulate the problem as a Maximum-Weight Independent Set (MWIS) problem on a graph where nodes represent feasible train cycles. To handle the computational complexity of the large search space, we propose a hybrid divide-and-conquer algorithm. This approach iteratively selects subgraphs and solves the MWIS problem using various solvers, including exact classical methods and the Quantum Approximate Optimization Algorithm (QAOA). We evaluate the algorithm's performance by comparing these methods and analyzing the scaling with respect to subgraph size, with QAOA assessed through both classical simulation and execution on a quantum device (IQM Emerald). Our results indicate that increasing the subgraph size generally improves solution quality, demonstrating that the hybrid framework can effectively bridge the gap between polynomial-time approximate solvers and exponential-time exact methods.

We develop a perturbative framework for a black hole embedded in a generic, possibly anisotropic, matter environment under spherical symmetry. Our approach extends previous analyses restricted to vanishing radial pressure or to perturbative matter configurations. Within this framework, we derive an analytical generalization of the Einstein cluster that incorporates a polytropic radial pressure, and we investigate the properties of this solution. We show that both the geodesic structure and the axial quasinormal-mode spectrum remain predominantly governed by an overall gravitational redshift effect, while the radial pressure systematically enhances the environmental corrections. In contrast, the tidal Love numbers are substantially more sensitive, and can exhibit order-unity deviations, including vanishing and negative strictly static magnetic Love numbers for sufficiently large anisotropy. We present the full linearized equations, which can be applied to various extensions, including ringdown analysis and extreme-mass-ratio inspirals.

We present a comprehensive theoretical investigation of dissociative recombination (DR) and resonant ion-pair (RIP) formation in $\mathrm{HeH^+}$ isotopologues using time-dependent wave-packet propagation methods. Nuclear dynamics are treated on a set of 23 coupled electronic states, including $^2Σ$, $^2Π$, and $^2Δ$ symmetries, in both adiabatic and strictly diabatic representations, with rotational couplings explicitly included. Reaction cross sections are computed over collision energies ranging from 0 to 50 eV. The results reveal that inclusion of a large manifold of resonant states and rotational couplings significantly enhances the DR cross section relative to earlier theoretical studies. In the diabatic representation, $^2Σ$ states dominate the recombination dynamics, while in the adiabatic representation, $^2Π$ and $^2Δ$ states contribute significantly at low collision energies. For RIP formation, two different diabatization schemes yield systematically larger cross sections than previous models, highlighting the sensitivity of ion-pair production to electronic coupling structure. Isotopic effects are examined, showing a clear inverse dependence of cross section magnitude on reduced mass. The present results underscore the importance of multi-state coupling and nonadiabatic effects in accurately describing electron-molecule collision processes in primordial and astrophysical plasmas.

Multiparameter quantum estimation faces two fundamental obstacles: sloppiness, i.e., anisotropy of the quantum Fisher information matrix (QFIM) that renders some parameter directions insensitive, and incompatibility, the non-commutativity of optimal measurements for different parameters. The trade-off bound $C_T$ captures their joint impact on precision, but it has remained unclear how the distribution of incompatibility across parameter planes affects its overall cost. Here we separate the total amount of incompatibility from its location. We introduce a dimensionless quantity $G_n^{(F)}$ that measures the alignment between the incompatibility distribution and the eigenvalues of the QFIM, and show how the Frobenius scale of the incompatibility contribution factorizes. We obtain a bound and prove the incompatibility cost lies between this bound and a rank-dependent multiple thereof. We also prove that at fixed sloppiness, or equivalently fixed Fisher volume, concentrating incompatibility into a single parameter plane reduces the optimized trade-off cost because the Fisher geometry can then be reshaped to allocate more Fisher area to that plane. A qutrit $SU(2)$ encoding numerically confirms that states with larger incompatibility strength can nevertheless incur a smaller cost if the matching factor $G$ is sufficiently small. Our results establish that the distribution of incompatibility relative to the Fisher eigenbasis is a central diagnostic for multiparameter estimation, beyond the total incompatibility strength.

As distributed quantum computing (DQC) offers a leading path towards scalable quantum computation, the ability to benchmark distributed algorithms under realistic conditions becomes critical for system co-design. However, without access to physical systems, researchers lack tools to evaluate distribution protocols. We introduce Q-DICE (Quantum Distributed Interconnect Compiler and Emulator), a hardware-aware emulation environment for benchmarking distributed quantum circuits on classical simulators and on NISQ-era monolithic hardware. This work provides three core contributions: (1) a programmatic scheme to construct distributed QPU backends, utilizing two novel techniques - QPU slicing and stitching - to facilitate distributed circuit mapping, (2) a methodology for modeling nonlocal link noise using physically motivated Kraus operators and stochastic error channels, and (3) a boundary-aware circuit mapping algorithm enforcing distributed QPU topology constraints during transpilation. Together, these components constitute a distribution-aware compiler and noise-modeling engine that faithfully enforces the physical limitations of distributed quantum hardware within existing execution environments. We validate Q-DICE against a multitude of experimentally demonstrated quantum circuits, including a distributed Grover's search on optically linked trapped-ion hardware, achieving a worst-case fidelity deviation of 4% between simulated and experimental results. These findings demonstrate Q-DICE's capacity to accurately reproduce real distributed quantum system behavior across platforms, streamlining experimentation with distributed quantum algorithms and architectures.

Non-Hermitian quantum systems exhibit qualitatively distinct physical behavior compared to Hermitian systems, a prime example being spectral singularities known as exceptional points. Their relevance in, e.g., quantum sensing, unidirectional transport, and robust lasing makes it important to be able to identify exceptional points through observable features of a many-body system. Here, using as an example a one-dimensional complex XY spin chain realizing both rotation-time RT- and parity-time PT-symmetric regimes, we develop a framework for detecting exceptional points based on the distinction between left and right eigenvectors of the Hamiltonian, which in a non-Hermitian system are no longer the adjoint of each other. We first show that a global measure constructed from the difference between the Hamiltonian and its adjoint locates exceptional points via distinct non-analytic behavior. At the level of observables, differences in local spin correlations evaluated on the right and left eigenstates provide a reliable static detection scheme. In contrast, static bipartite entanglement measures fail to capture this distinction, urging us to study the quantum dynamics of the model. Following a sudden quench, we demonstrate that the time-averaged right-left entanglement entropy difference directly encodes signatures of the exceptional point. In the RT-symmetric regime, it exhibits a pronounced peak at the exceptional point, whereas in the PT-symmetric regime it behaves as an order-parameter-like quantity, remaining finite in one phase and vanishing at the transition. Our results establish a direct link between the structure of non-Hermitian eigenstates and observable signatures of exceptional points, providing a practical route to identify them in existing quantum simulators.

While Hamiltonian Lattice Gauge Theory is gaining traction, today's limited numerical capacity leaves simulations affected by discretization errors. This motivates the implementation of renormalization group (RG) techniques to find discretization-error-free operators. To this end, we introduce VAPOR, a variational quantum algorithm that decomposes operators into Pauli strings, identifies RG flow orbits, and determines fixed points of a naively discretized operator. We illustrate this using a toy model of a kinematic operator in a symmetry-restricted SU(2) Yang-Mills theory.

We conduct a global general relativistic neutrino-radiation-transfer magnetohydrodynamics simulation of a $1.35$-$1.35M_\odot$ binary neutron star with the unprecedented spatial resolution of $6.25$\,m on the Japanese supercomputer FUGAKU. The total consumed CPU time is $\approx 530$ million core hours. We initialize the binary neutron star's magnetic field to be $3.16\times 10^{12}$~G at maximum, which is compatible with the upper end of the observed binary pulsars. We demonstrate that the Kelvin-Helmholtz instability that emerges when the two neutron stars touch amplifies the magnetic field to an expected electromagnetic saturation energy of $\sim 10^{50}$~erg within $3$~ms after the merger. The spectral analysis indicates that the Kazantsev and Kolmogorov spectra are reproduced in the magnetic and kinetic power spectral densities, respectively. We also find that it induces stellar-scale magnetic field amplification by at least a factor of $316$. We conclude that a magnetar may form at least temporarily following neutron star mergers in a few ms.

We investigate the existence and implications of ``bouncing geodesics'' in asymptotically flat Schwarzschild and Schwarzschild--de Sitter black holes. These trajectories, which probe the high-curvature regions near the black hole singularity, correspond to specific ``bouncing singularities'' in the bulk retarded Green's function. We provide a precise description of these singularities by combining the local Hadamard form with the global propagation of singularities theorem. We then derive the critical times at which the bulk retarded correlator becomes singular, considering all possible anchorings of the bouncing geodesics, including null infinity and the cosmological horizon. Finally, for black holes enclosed in a reflecting cavity, we establish a universal connection between the locations of the bouncing singularities and the spectrum of cavity quasinormal modes (QNMs) by deriving a cavity version of the thermal product formula, analogous to the one known for anti-de Sitter black holes. This relation allows one to extract information about the black hole interior from the asymptotic QNM spectrum measured at a reflecting hypersurface, even when the cosmological constant is zero or positive. We confirm this prediction through explicit examples by computing the cavity QNMs of scalar and electromagnetic fields, as well as gravitational waves, in spacetimes with asymptotically flat and de Sitter black holes.

Three-body interactions exhibit phases of strong chaotic evolution as well as hierarchical motion where one body separates from a binary and follows a hyperbolic or elliptic trajectory around it. The binaries produced during phases of hierarchical motion may lead to gravitational wave (GW) inspirals, but this depends on the outcomes of the chaotic states. In this paper we re-derive the elliptic outcome distribution using equilibrium statistical mechanics and explore it together with the hyperbolic distribution. When comparing to N-body simulations, we find that we can reduce the elliptic outcome model to one free parameter instead of the previously used two and that the predicted disintegration probabilities agree except for very low angular momentum triples. We then use both outcome distributions along with a star cluster model to design a Monte Carlo algorithm for repeated binary-single scatterings within dense star systems. We explore star cluster masses of $[10^5 - 10^7] M_{\odot}$, with the goal of quantifying observably eccentric merger (OEM) GWs, visible to instruments such as LIGO and Virgo. Assuming an OEM detection sensitivity of $f_{\rm min}=10 {\rm Hz}, e_{\rm min} = 0.1$, we find the elliptic OEMs are about $\sim (32 - 63)\%$ of the total elliptic mergers and that the total cluster mass greatly impacts the fraction of ejected binaries. The OEM to total merger fraction (OEM fraction) is found to be $(2.6 - 4.4)\%$. Considering the detection sensitivity that GW interferometers have today $(f_{\rm min} \simeq 34.4 {\rm Hz})$ we obtain the OEM fraction in the $(1.6 - 3.1)\%$ range.

This work investigates slowly rotating traversable wormholes supported by string fluids whose properties vary with distance from the throat. This radial variation allows the matter to transition smoothly from a de Sitter-like core near the center to a string-dominated environment further out, producing a regular, horizon-free, and asymptotically flat spacetime. By letting the transverse pressure depend on radius, the fluid naturally adapts to the surrounding geometry, resulting in a well-behaved energy density and shape function. Even modest rotation introduces frame-dragging effects that gently twist photon paths, creating subtle differences between co-rotating and counter-rotating trajectories. These effects are strongest near the throat, while at larger distances the spacetime is largely governed by the static gravitational potentials. Circular photon orbits reveal that the interplay of the redshift function, wormhole shape, and rotation shapes the photon-sphere structure. Different radial profiles of the string fluid generate distinctive photon-ring patterns, offering potential observational signatures of both the rotation and the internal matter distribution. Overall, radially varying string fluids provide a flexible and physically consistent source for traversable wormholes, bridging smoothly between vacuum-like and string-dominated regions while maintaining regularity and supporting slow rotation. This study highlights how anisotropic matter can influence both curvature and light propagation, providing a realistic framework for horizonless exotic spacetimes and suggesting new avenues to explore subtle observational effects around traversable wormholes.

Resonant $Z$-boson production is a cornerstone of precision electroweak physics, with its vacuum line shape set by the $Z$ mass, width, and collision kinematics. We show that a strong laser field can significantly alter this picture. By treating the field nonperturbatively, we find that laser dressing of the incoming fermions alters the effective collision kinematics and opens laser-photon exchange channels, including multiphoton processes, in $e^{+}e^{-}$ collisions. As a result, the $Z$-resonance profile develops distinct intensity-dependent regimes, evolving from the vacuum limit to saturation at intermediate field strengths and to an approximately quadratic enhancement at higher intensities. Additionally, the polarization composition of the produced $Z$ bosons is redistributed. In particular, at high intensities the laser-induced contribution can compensate the intrinsic chiral asymmetry of the electroweak interaction, leading to nearly parity-balanced $Z$-boson production. Our results identify that strong classical fields can dynamically control electroweak resonance phenomena, opening a bridge between strong-field QED and high-energy collider physics.

Recent results from the DAMIC-M and PandaX collaborations have excluded the standard freeze-in production of dark matter for masses in the range $3~\mathrm{MeV} \lesssim m_χ\lesssim 1~\mathrm{GeV}$ in the context of extensions of the Standard Model featuring an additional ultra-light $U(1)_{\rm X}$ gauge boson. In this work, we analyze the constraints induced by DAMIC-M and PandaX results on the reheating temperature in freeze-in models at stronger coupling, or when a non-thermal source (such as inflaton decay) comes into play. We identify viable scenarios in which the DM relic abundance is correctly reproduced while evading current experimental bounds on the electron-scattering cross section, $\overlineσ_\mathrm{e}$. In particular, we show that for reheating temperatures below the electroweak scale, Boltzmann suppressed production can be compensated by stronger couplings, bringing freeze-in scenarios within present experimental reach. Finally, we study a hybrid scenario in which a small branching ratio of inflaton decay seeds a nonzero initial dark-matter abundance. We show that such contributions can significantly modify freeze-in predictions across broad regions of parameter space, offering an additional pathway for probing extremely feeble interactions.

We critically examine the expectation that in a fundamental quantum field theory, dimensionless couplings in the Lagrangian density should all be of order unity. We propose a measure to quantify the adherence of a theory to this: the spread (the ratio of the largest to the smallest of the magnitudes) of such dimensionless couplings, obtaining various closed-form results. If we take independent identically distributed (IID) couplings to parameterise our uncertainty on the values of the order unity couplings, ratios of couplings can be much larger than one might naively expect. For a theory with 20 IID unit normal couplings, the probability that the magnitude of the ratio of two of them is greater than 100 is 0.29, for example. Even when the IID couplings have exponentially suppressed tails, the distribution of ratios of order one couplings has fat power-law tails which grow with the number of independent couplings.

Extensive systems have a simple thermodynamic signature: the logarithm of the partition function scales homogeneously with the size of the system. We show that the failure of this scaling, measured by the replica energy ${\cal E}$, provides a useful bridge between statistical mechanics and quantum field theory. The associated differential operator $(1-\frac1d L\partial_L)$ removes the leading bulk contribution to $W=\log Z$ and isolates the part that is sensitive to boundaries, topology, defects, long-range forces, or other sources of nonadditivity. In quantum field theory this thermodynamic idea has two closely related uses. For ordinary finite-volume or spherical partition functions, suitable higher-order versions of the same filter remove local counterterms and extract universal fixed-point data such as the central charge, the sphere free energy $F$, and the Euler anomaly coefficient $a$. For replica geometries with entangling defects, the same filtering principle gives the renormalized defect free energy. In $2+1$ dimensions, its $n\to1$ limit is precisely the entropic $F$-function. We use this perspective to distinguish ordinary finite-size corrections, topology-dependent constants in gapped phases, subextensive fracton degeneracies, and genuinely nonextensive systems with long-range interactions such as self-gravitating thermal matter. Replica energy therefore offers a common thermodynamic language for additivity, defect free energies, and renormalization-group irreversibility.

We study the next-to-leading order electroweak corrections to $e^+e^- \rightarrow W^+W^-$ from dimension-six two-fermion top-quark operators in the Standard Model Effective Field Theory. We compute analytical and numerical results for future electron-positron colliders, focusing on the proposed LEP3 and FCC-ee that will operate at centre-of-mass energies below the $t \bar{t}$ production threshold. We compare the indirect sensitivity arising from virtual corrections to $WW$ production to that from $ZH$ production, and to the current constraints from LEP and LHC data. We show that NLO corrections can provide competitive sensitivity to these operators. This work represents a first step towards the systematic computation of electroweak corrections to $W$-pair production at lepton colliders in the SMEFT, whose impact can then be properly assessed in global analyses.

In the Born-Oppenheimer approximation for QCD, the exotic hidden-charm tetraquark meson $χ_{c1}(3872)$ is a near-threshold bound state in Born-Oppenheimer potentials associated with an isospin-0 adjoint meson. The $χ_{c1}(3872)$ is the $1^{++}$ member of a heavy-quark spin-symmetry multiplet whose other members have $J^{PC}$ quantum numbers $0^{++}$, $1^{+-}$, and $2^{++}$. We introduce a simple model for the Born-Oppenheimer potentials that interpolates between the adjoint-meson potential at short distances and the triplet-meson-pair potential at large distances. We take into account the spin splittings of charm mesons nonperturbatively for the first time by solving the diabatic Schrödinger equation. We also take into account the spin splittings of the adjoint meson as well as a narrow avoided crossing with the quarkonium potential. We tune the energy of $χ_{c1}(3872)$ to the $D^* \bar{D}$ threshold and then calculate the spin splittings of the other members of the multiplet and their decay widths into charm-meson pairs. We also calculate the energies and decay widths of the corresponding multiplet of hidden-bottom tetraquarks. These calculations provide a template for the quantitative analysis of all hidden-heavy hadrons using the Born-Oppenheimer approximation for QCD.

We develop a general framework for matching KSVZ-like ultraviolet completions featuring vector-like fermions and a spontaneously broken $\mathrm{U}(1)_{\mathrm{PQ}}$ symmetry onto the Standard Model Effective Field Theory and the low-energy axion-like particle effective theory. The framework applies to arbitrary vector-like fermion representations and PQ-charge assignments, and systematically captures the effective interactions generated in both sectors. We then perform a comprehensive phenomenological analysis of the resulting SMEFT operators, based on global fits to electroweak precision, Higgs, and flavor observables, obtaining robust bounds on the corresponding Wilson coefficients that are largely independent of the details of the ultraviolet realization. These constraints can subsequently be translated into the QCD axion and ALP parameter space, providing indirect probes of ALP couplings. We further investigate several representative examples of the interplay between the SMEFT and ALP sectors, illustrating how direct ALP searches and indirect precision and flavor observables provide complementary information on the same underlying dynamics. We find that, over large regions of parameter space, indirect constraints derived from the SMEFT analysis dominate over direct ALP probes, except in scenarios where the PQ-charge assignment permits mass mixing with Standard Model fermions. Overall, our results establish a unified framework for connecting ultraviolet completions, SMEFT analyses, and ALP searches, enabling both the interpretation of existing constraints and the exploration of future signals within a common theoretical setting.

In this paper, one constructs a nonlocal extension of the Rarita-Schwinger theory for spin-$3/2$ fermions. Two classes of analytic form factors are considered: scalar form factors $f(\Box)$ and Dirac-operator form factors $f(\slashed{\partial})$. The massless theory is treated together with a covariant nonlocal gauge fixing, which allows the propagator to be written directly in terms of the spin-$3/2$ projector. In the massive theory, we show that the free Rarita-Schwinger constraints remain intact for analytic form factors, so that the unphysical spin-$1/2$ sector does not become dynamical. For $f(\Box)$ the tensor-spinor structure of the propagator is the same as in the local theory, while the pole equation is deformed by the scalar form factor. For $f(\slashed{\partial})$ the physical modes obey a nonlocal Dirac-type equation, leading to modified dispersion relations that can be written explicitly for exponential form factors. We discuss the conditions under which the construction is ghost-free at the free level and identify the natural limitations that must be addressed before interactions are introduced.

We use effective field theory to factorize production matrix elements that appear in the NRQCD framework for quarkonium cross sections. By applying a Hubbard-Stratonovich transformation and appropriate field redefinitions, we show that the soft and ultrasoft sectors of NRQCD can be decoupled from the heavy quark and antiquark fields in a hybrid vNRQCD/pNRQCD Lagrangian at leading order in the velocity power-counting. This enables us to re-factorize quarkonium production matrix elements in terms of matrix elements of color-singlet composite fields, which we can write as the wave-function at the origin, and state independent vacuum correlators of chromo-electric and chromo-magnetic gluon fields. This approach verifies powerful relationships between the LDMEs of different S-wave quarkonia originally derived using pNRQCD. Additionally, it allows us to derive new relationships for the production matrix elements used in the transverse momentum dependent factorization (TMD) framework, known as TMD soft transition functions, providing a much stronger set of constraints on these nonperturbative operators. This work significantly advances our understanding of quarkonium production, particularly in the TMD framework.

Motivated by earlier approaches, we develop a worldsheet framework for computing correlation functions in the single trace $T \overline{T}$-deformed tensionless AdS$_3$/CFT$_2$ duality. By describing the deformed bulk theory as a Berkovits-Vafa $\mathcal{N}=4$ topological string, we obtain a consistent definition of physical states and correlation functions, yielding a tractable setup for testing aspects of holography beyond AdS/CFT. We construct deformed physical vertex operators and compute their tree-level two-point functions exactly. We discuss the relation of our results to previous proposals for $T \overline{T}$-deformed two-point functions obtained from alternative worldsheet approaches, JT gravity, and perturbative field theory computations.

The Schiff moment of a nucleus is a symmetry-violating nuclear moment that indicates new physics beyond the Standard Model. We place the limit, $|\mathscr{S}({}^{153}$Eu)$| < 1.7 \times 10^{-8}$ $e\,$fm$^3$ (95\% confidence), on the Schiff moment of the $^{153}$Eu nucleus, using nuclear spin resonances in two ensembles of oppositely-polarized $^{153}$Eu$^{3+}$ ions in a Y${}_2$SiO${}_5$ crystal. This measurement using octupolar nuclei in a mm-scale crystal constrains new physics at the TeV energy-scale.

We study spherical T-duality for iterated sphere bundles. We show that for a class of iterated sphere bundles the cohomological data contained in its Gysin sequences can be repackaged into data for a vanishing Massey product. We further show that if these bundles are endowed with an integral cohomology class of transgressive degree one, then they have a T-dual iterated sphere bundle, namely, the one associated to the same Massey product read backwards.

Using a direct scan of 26 energy points near the $J/ψ$ resonance, we perform the first measurement of the relative phase $φ_{γ, 3g}$ between the strong and electromagnetic amplitudes in $J/ψ\to\bar K^0 K^*(892)^0+c.c.$ and $J/ψ\to K^+ K^*(892)^-+c.c.$. Unexpectedly, the phase in the neutral channel is found to be $\sim 150^\circ$, deviating from orthogonality ($90^\circ$) by 4.2$σ$ and from a relative real amplitude (0$^\circ$ or 180$^\circ$) by 10.0$σ$ or 1.6$σ$, respectively. In contrast, the charged channel phase is consistent with $\sim 180^\circ$ within 1$σ$, exhibiting model-dependent behavior. The corresponding branching fractions are consistent with the world averages but achieve better than twofold improvement in precision. The ratios between the branching fractions of $J/ψ\to\bar K^0 K^*(892)^0+c.c.$ and $J/ψ\to K^+ K^*(892)^-+c.c.$ are also measured. After subtracting the electromagnetic contribution, the corresponding strong amplitude ratios obey isospin symmetry within $1.8σ$. A search for direct $CP$ violation yields asymmetries consistent with zero.

In this work, we apply deformed light-cone null reduction method to a complex Maxwell theory in a manifestly gauge-invariant formulation. We show that the local U(1) gauge structure degenerates in the $c\to 0$ limit: the Gauss law constraint reduces from a restriction on initial data to a conservation law, releasing the longitudinal gauge mode as an independent degree of freedom (d.o.f). This raises the physical field count from $2(d-1)$ to $2d$. We prove a no-go theorem: under the single-mode Kaluza-Klein(KK)-like ansatz, no scaling of the field components can simultaneously preserve nontrivial dynamics and a first-class Gauss law, due to an inherent mismatch between velocity-type and constraint-type contributions in the parent action. Rather than representing the Carrollian electrodynamics derived via group contraction, the free complex scalar theory that emerges is merely an artifact of the truncation procedure at $c\to0$.

Witten recently proposed a background-independent algebraic framework for quantum gravity, wherein an observer endowed with a Hamiltonian defines a diffeomorphism invariant worldline algebra manifested by the modified Hamiltonian constraint. In the semiclassical limit, this construction admits a lift to a von Neumann algebra acting on a Hilbert space defined by geodesic in a fixed background. Motivated by this, we revisit quantization of certain class of deformed CFT Hamiltonian on a cylinder to capture non-perturbative aspects of black holes. We construct a type-I Von-Neuman algebra by imposing conformal boundary conditions on cut-offs near fixed points of Hamiltonian flow, acting on a GNS Hilbert space built from highest-weight representation of `emergent modular Virasoro algebra'. Upon identifying the Hamiltonian with the modular Hamiltonian of a sharp subregion associated to a fixed reference KMS (vacuum) state, the algebra changes to type-III$_{1}$ factor. We also discuss the structure of emergent Hilbert spaces using `open-closed string' duality after incorporating an emergent non-trivial center made out of scalars at fixed points. We further employ this modular quantization of a single holographic CFT to demonstrate how the boundary limit of exact Hartle-Hawking correlator of smooth BTZ background emerge in the strict semiclassical limit in an alternative dual description, while at finite $G_{N}$, the corresponding description is intrinsically non-smooth, featuring both a stretched horizon and a boundary cutoff. The exact correlator has also been precisely reproduced from the vacuum correlators in modular quantization. We further discuss the effect of incorporating gravity by including the center via AdS/CFT on boundary correlators, for which the description of a smooth horizon is replaced by a (stretched) horizon containing explicit microstructures embedded within it.

We revisited the singlet fermionic dark matter model in this work, where a Majorana fermion $χ$ carrying $Z_4$ charge is assumed as the DM candidate. A new singlet scalar $S_0$ with a non-zero vacuum expectation value is also introduced to the SM so that $χ$ can obtain mass after spontaneous symmetry breaking. We focus on the secluded DM region for the model, where interactions between DM and SM particles can be negligible. We have a new Higgs $h_2$ in the model, and the mixing angle of $h_2$ with the SM Higgs will play an important role in determining DM production, depending on the mass hierarchy between the new Higgs mass and the DM mass. We study DM relic density as a function of the model's four free parameters and estimate the viable parameter space under DM relic density constraint as well as direct detection constraint. We focus on the heavy DM mass region, and our analysis indicates that the mixing angle does not necessarily need to be very small in the secluded dark matter scenario, which offers potential for probing such models in future collider experiments.

We present femtoscopy as a new way to search for saturated gluonic matter in inclusive photonuclear processes, such as inclusive ultraperipheral collisions at the Large Hadron Collider (LHC) and the inclusive photonuclear reactions at the Electron-Ion Collider (EIC). As the femtoscopic approaches are sensitive to the space-time structure of the particle emitting source, they are ideal in providing insights also about the initial stage of the collision, where the gluon distributions may impact the effective size of the dipole the quasireal photons oscillate into. This technique demonstrates its capabilities in isolating nuclear shadowing and gluon saturation effects. Finally, we show how a femtoscopic approach is highly sensitive to sub-fermi scale structures typically observed in ultraperipheral collisions, such as gluonic hot spots.

We study the non-leptonic weak decays of doubly charmed baryons (${\cal B}_{cc}$) into singly charmed baryons (${\cal B}_c$) and vector mesons ($V$), denoted as ${\cal B}_{cc}\to{\cal B}_{c}V$. The short-distance contributions are calculated within the naive factorization hypothesis, while the long-distance final-state interaction effects are modeled via hadronic triangle diagrams. Unlike previous approaches, which compute only the imaginary part using the Cutkosky cutting rule, we evaluate the complete loop integrals to obtain both the real and imaginary parts of the amplitudes. These provide the nontrivial strong phases essential for CP violation. The model parameters are determined using experimental data. With this improved calculation method, we predict the branching ratios and decay asymmetry parameters for various decay channels, as well as $CP$ violations for short-distance dominated and singly Cabibbo-suppressed channels. This strengthens our theoretical framework for future study of doubly charmed baryons. Certain decays, primarily driven by long-distance effects, have been calculated; their observation in future experiments could help clarify the role of final-state interactions in charm baryon decays. Therefore, our calculation of ${\cal B}_{cc}\to{\cal B}_{c}V$ provides crucial predictions for branching ratios, decay asymmetry parameters, and $CP$ violation, which are essential for guiding experimental study at LHCb.

We investigate $J/ψ$ photoproduction off the proton, $γp \to J/ψp$, to elucidate the nonexistence of hidden-charm pentaquark signals reported by the GlueX and CLAS12 experiments. Within a coupled-channel rescattering mechanism, we employ the transition amplitudes from a previous coupled-channel analysis that dynamically generates the $P_{c\bar{c}}$ states. The kernel amplitudes for the transition to the $J/ψN$ channel include both $t$-channel heavy-meson exchange and $u$-channel heavy-baryon exchange. We find that the rescattering contributions from the $\bar{D}^{(*)}Σ_c$ channels -- indispensable for the formation of the $P_{c\bar{c}}$ resonances -- are about one order of magnitude smaller than those from $\bar{D}^{(*)}Λ_c$, since $g_{\bar{D}^{(*)}NΣ_c}$ is roughly five times smaller than $g_{\bar{D}^{(*)}NΛ_c}$. Since the $P_{c\bar{c}}$ resonances couple to the $J/ψN$ channel predominantly through the $\bar{D}^{(*)}Σ_c$ intermediate states, their suppression prevents the pentaquark signal from appearing in photoproduction. With only a single parameter controlling the overall normalization, the present work describes the GlueX and CLAS12 cross sections well. These results suggest that the null result from photoproduction need not be in conflict with the pentaquark signals observed by the LHCb Collaboration.

The behavior of higher-order radiative corrections due to initial state radiation in processes of electron-positron annihilation is analyzed. Numerical results for energies of future colliders are presented. Uncertainties of the known results on these corrections are estimated. A modified scheme for simultaneous exponentiation of pure photonic and non-singlet pair corrections is presented. Matching of the exponentiated results with the existing analytic higher-order calculations is constructed. A new DIS-like subtraction scheme is discussed.

Motivated by the observed $KD^{(*)}$ molecular candidates $D_{s0}(2317)$ and $D_{s1}(2460)$, their bottom--strange counterparts, $K\bar{B}^{(*)}$ molecular states, are naturally expected, although not yet experimentally established. This discrepancy may reflect sizable heavy-quark flavor symmetry breaking, which introduces significant model uncertainties. Current studies of heavy-quark flavor symmetry breaking effects still exhibit strong parameter dependence, and further experimental input is required to constrain these effects, in particular regarding possible additional $K^{(*)}D^{(*)}$ and $K^{(*)}\bar{B}^{(*)}$ molecular states. In this work, we examine whether additional $K^{*}D^{(*)}$ molecular states can be identified among the observed $D_s$ resonances. Within the Gaussian expansion method, we solve the Schrödinger equation using $σ$, $ρ$, $ω$, $π$, and $η$ exchange potentials, systematically including $S$-wave and higher partial waves. We find that $D_{s1}(2700)$ can be interpreted as a pure $P$-wave $DK^{*}$ molecule, while $D_{s1}(2860)$ and $D_{s3}(2860)$ are well described as $D^{*}K^{*}$ molecular states dominated by the $^{1}P_{1}$ and $^{5}P_{3}$ components, respectively. We also predict additional molecular states with various $J^{P}$ quantum numbers. These results provide a new description of the charmed--strange spectrum and a useful benchmark for heavy-quark flavor symmetry breaking effects.

We explicitly compute the axial anomaly in dense matter and show that its form remains unchanged from that in vacuum, even in the massless limit. This result follows from a subtle cancellation in the anomalous Ward identity between the medium-induced contributions to the divergence of the axial current and to the pseudoscalar density. We then revisit the chiral magnetic effect in a fermionic medium coupled to an axial chemical potential under an external magnetic field. We show that the medium supports a persistent, conserved anomalous current carried by fermions. The current is determined by the axial chemical potential and suppressed by the Fermi velocity, in agreement with anomalous axial-current correlation functions. We finally discuss applications to axion physics, where axion dark matter acts as an effective axial chemical potential.

WIMP scattering from 40Ar is investigated using a self-consistent Skyrme Hartree-Fock-Bogoliubov (HFB) approach. Nuclear form factors relevant to dark matter direct detection are calculated from the resulting one-body density matrix elements and compared with shell-model predictions. Good agreement is found for the spin-independent response, while significant differences are observed for the spin-orbit response due to variations in single-particle occupancies. The effects of particle-number projection are shown to be small for 40Ar. These results demonstrate the sensitivity of certain dark matter response channels to the underlying nuclear structure model and establish a framework for extending mean-field calculations to nuclei beyond the reach of large-scale shell-model studies.

A nontopological soliton of the Q-ball type in a Chern-Simons-Higgs gauge model is studied using both analytical and numerical methods. The general non-self-dual case is considered. It is shown that the soliton solution is an extremum of the energy functional at a fixed Noether charge. A differential relation between the energy, Noether charge, and the boundary value of the gauge potential of the soliton is derived. A linear relation between the components of the soliton energy is obtained. The parametric domain of existence of the soliton solution is determined. It is established that the soliton properties depend significantly on the form of the self-interaction potential of the scalar field. In particular, the energy and charge of the soliton can take arbitrarily large values only if the self-interaction potential has two degenerate zero minima.

Using a dataset of 7.33 fb$^{-1}$ of $e^+ e^-$ annihilation data collected with the BESIII detector at center-of-mass energies from 4.128 to 4.226 GeV, we report an updated measurement of the branching fraction of $D_s^+ \to τ^+ ν_τ$ via four $τ^+$ decay modes: $τ^+ \to e^+ ν_e \barν_τ$, $μ^+ ν_μ \barν_τ$, $π^+\barν_τ$, and $π^+ π^{0} \barν_τ$. The branching fraction is determined to be $\mathcal{B}({D_s^+ \to τ^+ ν_τ}) = (5.37 \pm 0.08_{\rm stat} \pm 0.06_{\rm syst})$ %. The product of the modulus of the Cabibbo-Kobayashi-Maskawa matrix element $|V_{cs}|$ and the $D_s^+$ decay constant $f_{D_s^+}$ is measured to be $f_{D_s^+} |V_{cs}| = (248.2 \pm 1.9_{\rm stat} \pm 1.4_{\rm syst} \pm 0.6_{\rm input} \pm 0.8_{\rm EM})$ MeV. Both the branching fraction and the product $f_{D_s^+} |V_{cs}|$ are the most precise results yet obtained. Then, taking $f_{D_s^+}$ from lattice quantum chromodynamics calculations results in $|V_{cs}| = 0.993 \pm 0.008_{\rm stat} \pm 0.006_{\rm syst} \pm 0.003_{\rm input} \pm 0.003_{\rm EM}$. Conversely, one finds $f_{D_s^+} = (255.0 \pm 1.9_{\rm stat} \pm 1.4_{\rm syst} \pm 0.6_{\rm input} \pm 0.8_{\rm EM})$ MeV when taking $|V_{cs}|$ from the CKMfitter group as an input. Combining with the world average value of $D_s^+ \to μ^+ ν_μ$, the ratio of the branching fractions between $D_s^+ \to τ^+ ν_τ$ and $D_s^+ \to μ^+ ν_μ$ is estimated to be $10.04 \pm 0.29$, which is consistent with the Standard Model prediction of lepton flavor universality.

The Circular Electron Positron Collider (CEPC) distributes a reference clock distributed to 192 control nodes along its 100~km underground tunnel. The required synchronization precision is 30~ps (standard deviation). We present an enhanced White Rabbit (WR)-based clock synchronization system designed to meet this requirement. A noise-budget analysis of the standard WR slave loop identifies the analog actuation chain (DAC + VCXO + multiplier PLL) and restart-induced timing uncertainty as the dominant limitations. In our redesigned node, the DAC+VCXO chain is replaced by a Si5345A DSPLL clock generator with DCO-based phase control, removing the board-level analog tuning stage. GTX transceiver phase alignment and manual byte-alignment fixing reduce restart uncertainty from 88.8~ps to 12~ps peak-to-peak. For multi-node operation, we introduce a cascaded global-control architecture with PC-side PID auto-tuned by TD3 reinforcement learning, on-chip-temperature feed-forward calibrated to $-0.76\,\mathrm{ps}/^\circ\mathrm{C}$. The measured point-to-point synchronization precision is 3.38~ps over 1~m fiber and 3.92~ps over 50~km. In a 12-level cascade, the end-node precision reaches 6.66~ps at constant temperature and 7.30~ps under a 13$\,^\circ$C temperature swing. Synchronized-clock TIE jitter stays below 1~ps regardless of cascade depth. Restart uncertainty is 2.82~ps (std.\ dev.). A 4-level cascade operated stably for 25 hours of continuous monitoring. All measured metrics fall well within the CEPC 30~ps budget.

We investigate whether a minimal $\mathbb{Z}_4$-symmetric fermion-scalar extension of the Standard Model can simultaneously realise viable dark matter, a strong electroweak phase transition, and a stochastic gravitational-wave signal. The model contains a real scalar singlet and a Dirac fermion, allowing thermal two-component dark matter, mixed WIMP-FIMP histories, and an effectively fermionic relic abundance generated by scalar decays. We impose theoretical consistency, the correct electroweak vacuum, and dark-matter constraints from relic density, direct detection, and invisible Higgs decays before using the surviving points as input for the finite-temperature analysis. This reveals that the compatibility between dark matter and a strong first-order electroweak phase transition is highly selective. After current dark-matter constraints are imposed, the strong-transition criterion along the Higgs direction is satisfied only in two viable regimes: the thermal two-component case with $M_ψ<M_S<2M_ψ$ and the decay-driven WIMP-FIMP case with $M_S>2M_ψ$. By contrast, the thermal regime with $M_S<M_ψ$ and the stable mixed WIMP-FIMP scenario with $M_S<2M_ψ$ are largely concentrated at small portal couplings or near the Higgs-resonance region, and do not yield a strong transition in the parameter space considered. The successful transitions typically proceed through an intermediate singlet-like phase. For representative nucleating benchmark points in the viable strong-transition regions, we compute the gravitational-wave spectra from sound waves and turbulence. Some spectra enter the projected reach of future space-based interferometers, showing that detectable signals arise only in selected dark-matter-compatible regions where a sufficiently active Higgs portal appears in correlated combination with the scalar mass and the remaining dark sector parameters.

We investigate a predictive and testable framework in which dark matter (DM), heavy-neutrino dynamics, and the BAU originate from correlated interactions within a local $U(1)_{B-L}$ extension of the SM. Unlike conventional $B-L$ constructions based on the type-I seesaw, we employ an inverse-seesaw mechanism realized through a sterile fermion $S_{1}$ and a complex scalar field $φ$, whose vacuum expectation value simultaneously generates the masses of the heavy neutrinos $N_{1,2}$ and the Majorana DM fermion $χ$ via Yukawa couplings. The small lepton-number-violating parameter induced by a higher-dimensional operator leads to naturally light active neutrinos together with TeV-scale heavy neutrinos and sizable active-sterile mixing, yielding distinctive collider signatures unavailable in minimal $B-L$ models. The relic abundance of $χ$ is governed by annihilation channels mediated by the same scalar and gauge interactions, producing a direct and model-specific correlation between successful leptogenesis and the observed DM relic density. A combined parameter-space analysis incorporating neutrino oscillation data, lepton-flavor-violating processes, direct-detection limits, and collider bounds on $N_{1,2}$ and $Z^\prime$ reveals a narrow yet robust region consistent with all these constraints. Representative benchmark points in this region are examined at a future multi-TeV muon collider. Heavy-neutrino production through electroweak processes yields striking signatures in the dilepton plus missing energy ($2\ell + E\!\!\!/_T$) and single-lepton plus di-jet plus missing energy ($1\ell + 2j + E\!\!\!/_T$) final states. These channels demonstrate that next-generation muon colliders offer a powerful and complementary probe of the inverse-seesaw origin of neutrino masses, the DM relic density, and the TeV-scale leptogenesis within such an extended $U(1)_{B-L}$ framework.

The Electron-Ion Collider (EIC) will enable a series of measurements at unprecedented energies and luminosities, providing new opportunities to investigate the microscopic structure of nucleons and nuclei at small $x_B$. Exclusive processes such as Deeply Virtual Compton Scattering (DVCS) offer unique access to the three-dimensional structure of hadrons through Generalized Parton Distributions (GPDs), while polarized electron and ion beams further enable detailed studies of spin-dependent structure. A model for coherent DVCS on polarized $^3$Heis developed and applied to simulations of for $9\times166$-GeV $e^3$He collisions at the EIC. Using this framework, the statistical precision achievable is estimated for measurements of beam-spin asymmetries and for the extraction of the Compton Form Factors (CFFs) $\mathcal H_{^3\mathrm{He}}$ and $\tilde{\mathcal H}_{^3\mathrm{He}}$. Early EIC data are found to enable precise differential measurements of the unpolarized CFF $\mathcal H_{^3\mathrm{He}}$ and to provide significant constraints on its real and imaginary components. By contrast, meaningful constraints on the polarized CFF $\tilde{\mathcal H}_{^3\mathrm{He}}$ require substantially larger integrated luminosities. The kinematics of the recoil $^3$He nucleus are also examined, and the far-forward detector capabilities at the EIC required to tag the intact nucleus and perform fully exclusive measurements of coherent nuclear DVCS are discussed.

Infinitesimal deformations of $\text{AdS}_5 \times \mathbb{S}^5$ form a representation of the AdS supersymmetry algebra. The structure of this representation has not yet been completely described in the literature. Some information can be obtained just from the fact that the space of deformations is the cohomology of a nilpotent BRST operator. We can consider the bicomplex formed by the BRST operator and the Lie cohomology differential, and its two spectral sequences. Their matching imposes some constraints on the structure of representations, which we start exploring in this paper. In particular, we clarify the structure of ghost number three zero modes.

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 \texttt{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 \texttt{T4-2-i} topology within a non-holomorphic 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.

These lecture notes are intended as a coherent introduction to conformal field theory in general, and composite operators in particular, through a semiclassical framework for computing scaling dimensions, with emphasis on operators of the form $φ^n$. In doing so, they aim to fill a gap in the literature and to help decode some of the relevant concepts. The physical idea is that at large $n$ an (heavy) operator creates a highly occupied state. Through the state-operator correspondence, this state lives on the cylinder $\mathbb{R}\times S^{d-1}$, and its scaling dimension is the corresponding energy of the theory on the cylinder. The notes are organized as a self-contained route from conformal symmetry to semiclassical dynamics. Part I reviews the conformal group, primary operators, radial quantization, the state-operator correspondence, and operator mixing. Part II builds the semiclassical framework, first in the free scalar theory, where the dimension of $φ^n$ is recovered in three independent ways, and then through the double-scaling limit, the action variable, and Bohr-Sommerfeld quantization. Part III develops the general machinery of periodic saddles, Floquet theory, fluctuation determinants, the Gel'fand-Yaglom method, and the Gutzwiller trace formula. Part IV applies the framework to the $O(N)$ $φ^4$ theory in $d=4-ε$ at the Wilson-Fisher fixed point, deriving the classical elliptic solution, the Lamé fluctuation spectrum, the zero modes, and the one-loop contribution to the large-$n$ scaling dimensions. Beyond the explicit computation, the notes emphasize the role of composite operators as probes of collective sectors of quantum field theory, with extensions to gauge theories, conformal windows, and asymptotically safe field theories.

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.

This work proposes a new terrestrial probe for millicharged particles (mCPs) and demonstrates promising discovery prospects. mCPs can be copiously produced in core-collapse supernovae (SNe), and a fraction may escape, travel to Earth and yield distinct signals. The mCP mass induces a time-of-flight (ToF) delay relative to the SN neutrino burst, opening a clean search window after the neutrino signal has passed. We compute the mCP-induced electron-recoil signals at XENONnT, JUNO, DUNE, and Hyper-Kamiokande for benchmark SN scenarios, and find that for $\varepsilon = 10^{-9}$ and sub-MeV to MeV-scale masses, more than 10 events per year can be detected. This search can improve upon existing SN cooling bound on $\varepsilon$ by up to an order of magnitude.

We contribute to the systematic analysis of New Physics effects in neutrino experiments using Effective Field Theory (EFT) methods. We review and extend the quantum field-theoretical formalism for generic neutrino interactions, discussing the inclusion of matter effects and deriving the connection with the density matrix formalism. On the phenomenological side, we apply this framework for the first time to medium-baseline reactor neutrino experiments. We derive analytical expressions for the relevant oscillation observables and perform a first EFT analysis of the recent JUNO dataset, extracting bounds on the leading non-standard interaction parameters.

Multipartite entanglement is a uniquely quantum form of correlation that captures collective properties of a composite quantum state beyond those encoded in its bipartite subsystems. We investigate this phenomenon in the process $e^+e^-\to t\bar tZ$ at a future lepton collider, where the final state spins span the tripartite Hilbert space $\mathscr{H}=\mathbb{C}^{2}\otimes\mathbb{C}^{2}\otimes\mathbb{C}^{3}$. Starting from the Standard Model helicity amplitudes, we reconstruct the full $12\times 12$ spin density matrix and characterise its entanglement structure through one-to-one negativities, one-to-other negativities, and the genuine multipartite negativity, evaluated at three increasingly inclusive levels of phase space integration. Pairwise entanglement is generally suppressed relative to the collective (one-to-other) and the genuine multipartite entanglement, and all measures decrease as more kinematic information is integrated out. Assuming quantum tomography in the fully leptonic decay channel at $\sqrt{s}=1$ TeV, we find that the collective entanglement is accessible at a realistic high-luminosity polarised lepton collider, while a direct observation of genuine multipartite entanglement is challenging and would benefit from further optimisation of the event analysis and observable choice. The study establishes $e^+e^-\to t\bar tZ$ as an attractive laboratory for probing multipartite entanglement in high-energy collisions and provides a general mixed state framework that applies to any tripartite spin system.

Inspired by the $P$-wave molecular interpretation of the recently observed vector state $G(3900)$, we analyze the production of its possible pseudoscalar partner, denoted here as $G_0(3900)$, via the radiative decay $Y(4230) \to γG_0(3900)$. The $G_0(3900)$ is interpreted as a $P$-wave molecular state with quantum numbers $J^{PC}=0^{-+}$, dominated by the $D\bar{D}^{\ast}/\bar{D}D^{\ast}$ components. Although not yet experimentally established, such a structure is expected to appear near the $D\bar{D}^*$ threshold and to exhibit characteristic production patterns. The decay is assumed to proceed through a triangle mechanism. Depending on the model parameters and the binding energy of $G_0(3900)$, the resulting branching fraction lies in the range $\mathcal{B}(Y(4230) \to γG_0(3900)) = 3.8 \times 10^{-5} - 3.3 \times 10^{-4}$. Our results offer a pathway to search for signatures of $G_0(3900)$ in radiative channels and also provide a test of the consistency of loop-mediated radiative decays with a molecular description of the $Y(4230)$.

Small-scale galaxy clustering is expected to contain substantial cosmological information, but the extent to which this information constrains halo-based cosmologies independent of an assumed galaxy--halo connection remains unclear. We quantify this dependence using LRG-like mock galaxy catalogs built from 81 cosmologies in the {\tt \textsc{AbacusSummit}} suite. We analyze two-point correlation function multipoles on scales ranging from $5$--$80$ Mpc/$h$ and compare two limiting treatments, the \enquote{floor} and \enquote{ceiling}, of the standard five-parameter HOD model. In the conservative floor case, we impose only broad initial HOD bounds and profile over HOD parameters to determine the minimum constraining power available; we accomplish this with {\tt HODmin}, a two-stage global optimization algorithm written for minimizing $χ^2$ in HOD space. In the optimistic ceiling case, we assume the HOD parameters are known exactly. We find a significant difference between the floor and ceiling when comparing against the same Planck $Λ$CDM mock data vector with identical modeling assumptions: for the floor, $25\%$ of the discrete {\tt \textsc{AbacusSummit}} cosmologies tested are excluded at $3σ$, whereas for the ceiling, $\sim81\%$ are excluded. Many cosmologies agree well with data in the floor, and yet in the ceiling are excluded by multiple orders of magnitude in $χ^2$. We therefore observe the strength of small-scale clustering constraints depends heavily on the amount of prior HOD information assumed. We compare the sensitivity of this effect to various choices like scale cut, angle cut, multipole inclusion, mock phase, and mock HOD model. Our wide floor--ceiling bracket indicates that informative galaxy--halo priors are necessary for extracting strong small-scale clustering constraints.

Overview: Wind-driven bubbles are dynamic systems that can accelerate cosmic rays, depending on their physical properties, up to very high energies. We investigate how a time-dependent description of the particle transport may impact the escaping cosmic-ray flux. Model: The wind bubble system is modeled as spherically symmetric. Cosmic rays are continuously injected at the position of the termination shock and propagate through advection and diffusion until the escape at the time-dependent position of the forward shock, which is treated as a free escape boundary. Methods: The one-dimensional spherical time-dependent transport equation is solved by transforming it into the corresponding set of stochastic differential equations, and integrated with a modified version of the open source cosmic-ray propagation framework CRPropa. Results: We find that, during the wind driven phase, the downstream escaping spectra from wind bubbles can be harder than $\sim E^{-2}$, the conventional expectation from diffusive shock acceleration. Depending on the turbulence model the initial energy spectrum can be significantly suppressed at lowest energies, which could be an observable feature to distinguish between different turbulence realizations. This effect could lead to an efficient confinement of low energy particles, potentially leading to observable implication in terms of multi-messenger radiation and cosmic-ray accumulated grammage within the bubble.

Weak lensing by large-scale structure is a powerful cosmological probe. While most analyses rely on second-order correlations, these are primarily sensitive to the parameter combination $S_8 = σ_8 (Ω_m/0.3)^{0.5}$, limiting their ability to constrain $Ω_m$ and other cosmological parameters independently. Higher-order statistics capture non-Gaussian features of the density field and can therefore break parameter degeneracies and extract more cosmological information from weak lensing surveys. We present a joint analysis of second- and third-order cosmic shear in the final data release of the Kilo-Degree Survey (KiDS-Legacy). We combine COSEBIs (Complete Orthogonal Sets of E-/B-mode Integrals) at scales between 2' and 300' with third-order aperture mass moments at scales between 4' and 32' to perform a joint analysis of second- and third-order statistics. Compared to previous KiDS analyses, we implement several methodological advances: an intrinsic alignment model with redshift and mass dependence, a baryon correction model validated on multiple hydrodynamical simulations, and corrections for reduced shear and source clustering. Combining COSEBIs with third-order aperture mass statistics in KiDS-Legacy yields $Ω_m = 0.297^{+0.056}_{-0.040}$ and $S_8 = 0.806^{+0.025}_{-0.023}$, significantly tightening the $Ω_m$ constraints and more than doubling the figure of merit in the $Ω_m$--$S_8$ plane compared to the two-point analysis alone. The third-order measurements pass stringent internal consistency tests, are fully compatible with the KiDS-Legacy 2-point constraints, other 2+3-point lensing results and with Planck CMB measurements within $1σ$, providing no evidence for an $S_8$ tension and demonstrating the maturity of 3-point cosmic shear as a key probe for forthcoming surveys.

Recent Very Long Baseline Interferometry (VLBI) and Gaia astrometry reveal systematic milliarcsecond-scale offsets between the radio and optical centroids of active galactic nuclei (AGN). These "radio-optical offsets" do not alter the standard opacity-driven interpretation of radio core shifts. Instead, they indicate that the optical emission centroid is frequently displaced downstream of the radio synchrotron optical depth $τ= 1$ surface, implying that additional dissipation and particle reacceleration occur beyond the opacity radio core within relativistic jets. We perform steady-state, axisymmetric relativistic magnetohydrodynamic (RMHD) simulations of AGN jets, including baryonic mass-load from stellar winds, varying jet kinetic power, and stellar core radius. Synthetic synchrotron emission maps in radio and optical bands are generated via a radiative transfer code, and centroid offsets are extracted for comparison with observations. Parsec-scale radio-optical offsets arise only for jet powers $L_{\rm j} \sim 10^{42.5} - 10^{44}\,\rm{erg}\,\rm{s}^{-1}$. In this regime, stellar winds trigger jet deceleration at intrinsic distances of a few $10^2-10^3\,\rm{pc}$, shifting the optical centroid downstream and producing offsets of $\sim 0.1 - 4\,\rm{mas}$ (a few tens of parsecs at $z=1$). Offsets depend on stellar distribution, viewing angle, and optical jet dominance, and vanish outside this power range. We reproduce the observed redshift evolution of offset incidence, linking it to the cosmic evolution of thermally pulsing asymptotic giant branch (TP-AGB) mass loss. Although stellar mass loading is unlikely to be the sole dissipation mechanism, its unavoidable presence in galactic nuclei makes it a natural baseline for energy dissipation. Radio-optical offsets therefore offer a constraint on AGN jet power and jet-host coupling, independent of traditional lobe-based methods.

Little Red Dots (LRDs) are a heterogeneous class of objects, with several proposed scenarios for their physical nature and evolution. While these theories have been tested on individual LRDs using limited spectral features, a systematic Bayesian analysis of the LRD population incorporating the different models across a broad wavelength range is still lacking. In this study, we conduct a consistent ultraviolet (UV)-to-optical continuum fitting analysis of 66 LRDs at 2<z<6 using JWST/NIRSpec PRISM spectroscopy. Employing a modified version of Bagpipes--including blackbody (BB) emission affected by Balmer absorption, stellar and nebular emission attenuated by dust, and an active galactic nucleus (AGN) component--we assess the performance of the black hole star (BH*) model in describing the LRD population. We adopt broad priors and therefore do not impose any specific physical scenario. Our results show that only ~6% of LRDs with statistically robust solutions (52 objects in total) are best-fit by a BH* in the optical and a host galaxy in the UV. ~8% of LRDs show BB-dominated optical continua but lack a stellar component or exhibit AGN UV leakage. Most LRDs are dominated by stellar and/or AGN emission in the optical, with minor BB contribution. When we adopt a prior that disfavors a strong AGN continuum to enforce BH*-like solutions, the percentage of BH$^*$ systems increases to ~40%, highlighting the strong degeneracy between a BH* solution and alternative scenarios. Even when BH*-like solutions are enforced, many LRDs still require a stellar-dominated optical continuum. This may reveal limitations of the BH* model or point to an evolutionary sequence in which the BB contribution decreases as the host grows, leading to lower BB temperatures and higher stellar masses at lower z. In this scenario, more pronounced ''V'' shapes would correspond to later stages in LRD evolution.

Quasars can serve as standard candles through an empirical scaling relation between their ultraviolet (UV) and X-ray luminosities. As high-redshift probes, it is critical to test whether this relation evolves with redshift. In this work, we reconstruct the Hubble diagram of the Pantheon+ sample using the deep learning--based LADDER algorithm and use it as a reference to investigate the quasar scaling relation. Our results, which are consistent with those from Gaussian process regression and narrow-bin analyses, show that the potentially contaminated sample at $z<0.7$ differs significantly from the $z>0.7$ sample; thus, it should be further screened or excluded when quasars are used as cosmological probes. We find that the scaling relation exhibits a non-linear redshift dependence that cannot be accounted for by a simple linear correction, and that this behavior is a feature of the current data sample rather than a consequence of cosmological model misspecification. To use quasars as standardizable candles, further modeling of the scaling relation and intrinsic dispersion, or more advanced data processing techniques, is required.

Nonlinear structure growth generates higher-order correlations and morphological features in the cosmic density field that cannot be fully characterized by two-point statistics. Upcoming surveys will measure these features with greater precision, making it essential to develop methods capable of extracting as much cosmological information as possible from them. Field-level inference (FLI) is one such approach, in which cosmological parameters are constrained by comparing observed maps to forward-modeled maps, either directly or through learned summaries that retain map-level information. In this work, we compare FLI with power-spectrum-based inference using the same forward-modeling pipeline for generating weak lensing maps, with the goal of quantifying the gain from map-level analysis relative to two-point statistics. We perform this comparison with both implicit and explicit inference methods, using 8-million-parameter forward models based on Lagrangian perturbation theory and particle-mesh (PM) N-body evolution. The two FLI approaches yield closely consistent posteriors; this agreement, together with coverage tests confirming the calibration of the implicit analyses, gives us confidence in the recovered field-level constraints. Relative to the power-spectrum-based analyses, these results show significant gains in cosmological information, especially when small scales are included in the PM-based forward model. We then discuss the remaining challenges that must be addressed before PM-based explicit FLI can be applied to observational datasets.

Studying young stellar populations within galaxies can help refine our understanding of recent star formation in galaxies and their evolution. With this motivation, we present a catalog of ~25,000 recently formed (within 400 Myr) star-forming clumps (SFCs) in 17 morphologically diverse nearby galaxies, including 8 massive, classic spirals, 6 intermediate-mass, flocculent spirals, and 3 dwarf irregulars. We used far- and near-UV observations from the UltraViolet Imaging Telescope (UVIT), whose ~1.5" angular resolution and 28' field-of-view allow us to probe SFCs at a mean physical scale of ~54 parsec, within the full extent of our galaxies. We adopted a homogeneous SFC detection criterion, corrected for spatially varying dust attenuation (using 6" resolution A_V maps, made by combining FUV with archival infrared observations), and estimated the SFC ages by comparing the observed UV color-magnitude diagrams with Starburst99 simple stellar population models. Using our SFC catalog, we studied the age demographic of the recently formed stellar populations across different galaxy morphologies and observed age trends consistent with several well-known phenomena, such as the inside-out formation of disc galaxies, local gravitational instabilities leading to flocculent spiral arms, and the stochastic nature of star formation in dwarf galaxies. Leveraging full galaxy coverage and FUV data, our catalog complements existing optically-identified star cluster catalogs in the literature towards improving our understanding of star formation across a wide range of galaxy morphologies, masses, and environments. We make the SFC catalog and A_V maps of our 17 galaxies publicly available with this paper.

An accurate determination of the dust attenuation within galaxies is essential to derive key physical properties such as the star formation rate (SFR). We present an analysis using the JWST/NIRISS data from the GLASS-JWST ERS programme to investigate and characterise the stellar and nebular reddening of galaxies at $1.0<z<2.4$, down to the sub-kpc scale. We use a multiregion fitting method to extract high-quality H$α$ and H$β$ emission line maps for 99 individual galaxies across a stellar mass range $7.0<\log_{10}(M_*/\mathrm{M}_{\odot})<10.5$. We find no evidence for ratios of the Balmer decrement (H$α$/H$β$) below the intrinsic limit for Case B recombination, beyond the expected variation from observational uncertainties. We reproduce the local correlation between the Balmer decrement and total stellar mass, and find no measurable difference when splitting the sample by redshift, with negligible attenuation below $\log_{10}(M_*/\mathrm{M}_{\odot})\lesssim8.5$. Similarly, the best-fit relation between the nebular and continuum reddening follows the same relation as in local starburst galaxies, $E(B-V)_{\mathrm{SED}} = (0.46\pm0.02)E(B-V)_{\mathrm{neb}}$, together indicating no significant evolution in the dust geometry within galaxies out to $z\lesssim2.4$. We derive best-fit linear relations between the differential nebular-stellar reddening and the SED-derived star formation rate (SFR) and stellar mass, finding statistically significant relations for both quantities. We use our spatially-resolved measurements to derive an empirical calibration between the resolved differential reddening, and the SFR surface density. These will enable crucial dust attenuation corrections for spatially-resolved science at higher redshifts where the Balmer lines are inaccessible, such as with future Roman grism observations.

We develop a stochastic framework for the stationary spherical Jeans equation, motivated by the field-dependent nature of the gravitational coupling in scalar--tensor theories. We model unresolved spatial fluctuations of the scalar sector as an effective stochastic contribution to the gravitational coupling, $\Geff(r,ω)=\Gbar(r)+Γ_G(r)ξ(r,ω)$. This approach induces a linear Itô stochastic differential equation for the radial velocity dispersion $y(r)=σ_r^2(r)$, defining a nonautonomous radial random flow rather than a time-evolution problem. We derive the associated Fokker--Planck equation and obtain integral expressions for the mean, variance, and covariance of the radial velocity dispersion. Because the noise is additive, the deterministic Jeans solution is recovered as the mean profile, while the stochastic sector produces a probability band around it. We specialize the construction to Navarro--Frenk--White, Hernquist, and Einasto halo models and propagate the radial covariance to the projected line-of-sight velocity dispersion. This provides a semi-analytical framework for assessing how effective gravitational fluctuations can affect halo kinematic observables in the stationary Jeans regime.

The shutdown of star formation - quenching - marks a pivotal transition in the lives of massive galaxies, which dominate the present-day stellar mass density. This review synthesizes our current understanding of the mechanisms that trigger and maintain quiescence. We discuss the nuances of how quiescent systems are identified across cosmic time and summarize the evolving physical properties of the growing massive population, including their stellar populations, chemical enrichment histories, and gas and dust reservoirs, highlighting several key results: (1) Quiescent galaxies can be identified with empirical color selections, but evolving specific star formation rate thresholds offer a more robust physical distinction from star-forming systems. (2) The earliest massive quiescent stellar populations show rapid formation histories and high metallicities, with enhanced $α$-elemental abundances often distinct from local analogs. (3) Nascent studies of gas and dust in quiescent galaxies reveal diverse multiphase reservoirs and outflows, pointing to fast ejective and slow regulatory modes of galaxy quenching. (4) In situ processes establish galaxy central density, while assembly continues via (minor) mergers post-quenching, reshaping all massive galaxies and disrupting rotation in most cases. We distill observations into two broad modes by which massive galaxies form and quench: one involves a rapid, early shutdown driven by supermassive black hole outflows on short timescales; the other proceeds gradually through gas exhaustion, virial heating, or preventative feedback, each leaving distinct observational signatures. Together, these pathways offer a testable framework for modeling the formation and evolution of massive galaxies, which will be informed by future studies of their stars, gas, dust, and dynamics.

We measure the dark matter halo mass function, with backsplash halos removed, from a wide range of cosmological-box and zoom-in simulations. These include the MultiDark Planck boxes, along with a suite of zoom-in simulations of Group, Milky Way, and LMC-mass halos. The Milky Way simulations include both CDM and non-CDM initial conditions. Using these measurements, we calibrate the parameters of flexible fitting functions for the halo mass function and the window function, along with parameterized models for various systematics, including finite box size effects, halo isolation criteria, halo detection efficiency, and contamination by artificial halos (objects forming from particle noise in the initial conditions). We show that this model shows remarkable consistency with N-body simulations over a broad range of redshifts, and ten orders of magnitude in halo mass ($10^6\mathrm{M}_\odot$ to $10^{16}\mathrm{M}_\odot$). Our model typically maintains a high precision of 12% and captures complex behaviors, including small-scale cut-offs, oscillations, and enhancements. In specific mass intervals for certain power spectra, we see larger deviations of 40-50%. Furthermore, when integrated with a simple model for environmental dependence, this fitting function provides a robust description of how environmental density influences the halo mass function. This precision model captures a wide variety of dark matter paradigms (including thermal relics, axions, and models with dark-sector interactions), is accurate for halo masses down to $10^7\mathrm{M}_\odot$, and is a critical ingredient for model-independent dark-matter inference from forthcoming data.

SN 1006 is a historical Type Ia supernova remnant that exhibits non-thermal emission ranging from radio to multi-TeV $γ$-rays. Most of this emission (particularly X-rays and $γ$-rays) is concentrated in polar caps aligned with the ambient magnetic field, which makes it an ideal laboratory for studying cosmic ray (CR) acceleration at different shock obliquities and the hadronic/leptonic nature of the $γ$-ray emission. We model SN 1006's morphology, multi-wavelength spectrum, and radial profile using a self-consistent multi-zone kinetic model of particle acceleration that accounts for: CR-driven shock modification, magnetic field amplification, drift in magnetic fluctuations, and temporal dynamics including adiabatic and synchrotron losses. Our model can reproduce both the observed spectral and spatial properties, with the exception of the radio profile that we argue requires 3D hydrodynamic effects to replicate. We find that quasi-parallel regions (where the shock normal aligns with the ambient magnetic field) exhibit very prominent CR acceleration ($\sim$20% efficiency), while quasi-perpendicular regions exhibit efficiencies below 1%, consistent with the results of kinetic simulations. We also find that electrons are responsible for the majority of the $γ$-ray emission from SN 1006 (i.e., it is a leptonic source), with the exception of the northwest region due to an encounter with a dense cloud.

The population of Compton-thick (CTK) AGN represents a critical yet elusive phase in the growth of supermassive black holes. Constraining their abundance and evolution at high z is essential for understanding both SMBH growth and the origin of the cosmic X-ray background. We investigate the X-ray luminosity function (XLF) of CTK AGN at z > 3 using one of the largest available samples of X-ray-selected AGN at high z, containing 811 sources from XMM-Newton XXL-N and Chandra CCLS and CDF-S/N surveys. We first selected a subsample of ten high-probability CTK candidates, identified through x-ray spectral fitting. Their multiwavelength properties are examined through SED modelling to assess the robustness of their CTK classification. For most sources, the inferred X-ray luminosities appear overestimated when compared with their IR luminosities. After updating the NH posteriors with IR-informed priors, only three sources remain consistent with the CTK regime. To compute the XLF for the entire CTK AGN population, we used 24 microns photometry to estimate IR luminosities and update the X-ray posteriors for all the remaining sources. Incorporating IR priors systematically reduces the inferred CTK number densities, yielding a more conservative and physically consistent estimate of the XLF. We find that CTK AGN constitute 17 per cent of the total AGN population at 3 < z < 6, consistent with results at lower z. Our analysis reveals no statistically significant evolution in the CTK fraction up to z about 6, suggesting that the most heavily obscured accretion phase remains a persistent component of black hole growth throughout cosmic history. While the overall obscured AGN fraction (NH > 1e23 cm-2) increases toward higher redshifts, the stable CTK fraction supports the interpretation that, at these epochs, the interstellar medium in typical host galaxies cannot produce CTK levels of obscuration.

Our study focuses on the strong tidal disruption of a cold helium white dwarf passing a black hole. We model the white dwarf as a Bose-Fermi droplet and use quantum hydrodynamic equations to simulate the binary system's evolution. As the white dwarf passes through periastron, it loses a significant amount of mass. This mass falls onto the black hole and forms an accretion disc. Quantized vortices appear in the accretion disc, manifesting as strong electromagnetic radiation signals that exhibit characteristic flickering patterns changing on a timescale of a few seconds. Meanwhile, the white dwarf moves away from the black hole. As the white dwarf moves through space, vortices run along its surface. This elongates its geometry, causing it to rotate and emit gravitational waves.

$g$ mode resonances in eccentric binary neutron star systems are potential probes of internal stratification, superfluidity, composition gradients, and the equation of state. Although such weak dynamical tidal signatures are unlikely to be resolved with current detector sensitivities, third generation observations may make them accessible, in which case identifying the weak resonant phase shift would provide information beyond the bulk adiabatic tidal deformability. We build a four class dataset in an eccentric harmonic framework, containing point particle, adiabatic tide, resonant $g$ mode, and pure noise samples, and use Einstein Telescope (ET) and Cosmic Explorer (CE) detector data to test whether this weak resonant phase signature can be identified from noisy time domain strain. The ET, CE, and ET+CE deep learning models reach accuracies of $0.655$, $0.815$, and $0.897$, respectively. On the same simulated samples, the matched filtering method reaches lower accuracies of $0.514$, $0.677$, and $0.689$. This result arises from the fact that the resonant correction manifests as a weak phase morphology difference superimposed on the adiabatic tidal background, whereas matched filtering is sensitive only to the overall similarity. Hence, in the presence of weak phase differences, the neural classifier employed in deep learning is better able to learn these local phase and morphology features from the complete time domain strain segment. The results indicate that joint third generation observations improve the identifiability of weak internal mode phase information.

Dwarf galaxies, characterized by their low luminosities and masses, are excellent candidates for searches for intermediate-mass black holes (IMBHs), particularly when they show strong accretion and ejection activity. The dwarf galaxy SDSS J101747.09+393207.7 has recently been found to display a very high X-ray luminosity and an X-shaped optical structure, possibly caused by a dwarf--dwarf merger. To explore its potential IMBH ejection activity, we performed very long baseline interferometry (VLBI) observations at 4.9 GHz. In this work, we present the detection of a milliarcsecond-scale, compact, sub-microjansky radio component near the optical centroid. According to some existing radio sky survey data, the radio component was not detected until 2015; it displayed an optically thin steep radio spectrum and declining flux densities across 0.8--5 GHz from 2019 to 2025. Therefore, we identify it as a short-lived and rarely seen ejecta that was produced by unstable accretion onto a massive black hole and likely faded away in a few decades. These results indicate that short-lived, episodic jet activity from accreting IMBHs in dwarf galaxies might exist.

The propagation of very-high-energy (VHE; $E_γ \geq 100$ GeV) gamma rays from extragalactic sources is affected by interactions with photons of the extragalactic background light (EBL), resulting in pair production that attenuates the intrinsic gamma-ray flux. This interaction renders the Universe increasingly opaque to VHE photons at high energies and redshifts. New physics scenarios involving axion-like particles (ALPs) could modify this expected optical depth. In particular, ALPs with masses $m_a \sim 10$ eV can decay into two photons over cosmological timescales, thereby contributing to the diffuse EBL. If such ALPs constitute a significant fraction of the dark matter density, their decay would enhance the EBL intensity and consequently increase the gamma-ray optical depth. In this study, we investigate this scenario using a large sample of gamma-ray spectra observed with the High Energy Stereoscopic System (H.E.S.S.) and the Fermi Large Area Telescope. We model the contribution of decaying ALPs to the EBL and assess their impact on the spectra of blazars across redshifts. By comparing these observations with standard EBL models, we place constraints on the properties of heavy ALPs, specifically their mass and photon coupling, and evaluate their viability as a dark matter candidate capable of modifying the gamma-ray transparency of the Universe. From the combined analysis, and under the assumption that ALPs constitute the entire dark matter density, we derive 95% confidence exclusion limits on the photon-ALP coupling down to $g_{aγ} \sim 7 \times 10^{-12}$ GeV$^{-1}$ for masses $m_a\sim 15$ eV. These constraints are competitive with existing astrophysical bounds and provide complementary sensitivity to other techniques, closing a previously unconstrained region of parameter space in the $m_a \sim 2.5$-$20$ eV range.

With the aim to study the spectral properties of some X-ray sources from recently observed {\it Chandra} data, 9 galaxies which have been observed by {\it Chandra} ACIS-S during the year 2018 to 2022 have been considered for the present work. 27 sources with net source counts $ \ge$ 100 have been considered. The spectra of all the sources were fitted using two empirical models -- an absorbed powerlaw and an absorbed disk blackbody. From their estimated bolometric luminosities, the 27 X-ray sources are categorized as 6 X-ray binaries (XRBs) and 21 Ultraluminous X-ray sources (ULXs). All the six XRBs are found to be in the spectrally hard state ($Γ\sim$ 1.52-2.29) which indeed may be due to thermal comptonization. Only one ULX, CXOUJ032251.2-370950 (X-5), was found to be spectrally soft while the remaining 20 ULXs were spectrally hard. The spectral parameters of X-5 with an inner disk temperature (kT$_{in}$) $\sim $ 0.5 keV and an estimated bolometric luminosity, L$_X \sim$ 3.26 $\times$ 10$^{39}$ erg s$^{-1} $ requires a black hole of mass, M$_{BH} \sim$ 137.86$^{+66.62}_{-47.41}$ M$_\odot $ accreting at $ \sim$ 0.19 times its Eddington limit. 8 ULXs, X-4, X-8, X-9, X-10, X-11, X-12, X-20 and X-21, were found to be in the Extremely luminous X-ray sources (ELXs) regime with even their lower limit of luminosity $>$ 10$^{40}$ erg s$^{-1}$. Softening/Hardening of spectra with or without changes in the luminosity were also observed in some ULXs/ELXs. In the hard ELX, X-8, spectral softening with almost consistent luminosity was observed. While in the ULXs X-20 and X-25 spectral softening with increasing luminosity was observed. However spectral hardening with increase in luminosity were observed in the ULXs X-21 and X-26.

We investigate two massive young stellar objects (YSOs), IRAS21078+5211 and G035.02+0.35, where evidence for magnetohydrodynamic (MHD) disk winds (DWs) has been obtained at scales of 10-100 au through measurements of the 22GHz water maser velocity distribution within the Protostellar Outflows at the EarliesT Stages (POETS) survey. We employ IRAM Northern Extended Millimeter Array and archival Atacama Large Millimeter Array observations of IRAS21078+5211 and G035.02+0.35, respectively, to study kinematics and physical conditions of the corresponding protostellar winds on scales of 100-1000 au using the same molecular tracers. In IRAS21078+5211, the emissions of several molecules, particularly SO, SO2, CH3CN and CH3OH, are distributed along the axis of the radio jet, and present a LSR velocity (Vlsr) gradient transversal to the jet axis. Position-velocity (PV) plots of the SO lines show patterns consistent with Keplerian rotation. The SO2 emission comes from high velocity gas flowing close to the jet axis, while CH3CN and CH3OH present larger radial extension than the S-bearing species. In G035.02+0.35, the same molecules are instead distributed along the major axis of the rotating disk, and their Vlsr gradients consistently trace the disk rotation. The corresponding PV plots present Keplerian profiles. SO is the only molecular species whose emission extends well outside the disk. In both YSOs, the spatial and velocity distributions of SO are consistent with a rotating wind magneto-centrifugally launched from the YSO disk. The comparison with models of molecule formation and excitation in shocks indicates that the different radial extension of the molecular species observed in the protostellar wind of IRAS21078+5211, as well as the lack of molecules, except SO, in the G035.02+0.35's wind, can be explained in terms of a radially extended MHD DW, rather than a compact X-wind.

Most galaxies host supermassive black holes (SMBHs) that remain weakly accreting or dormant for much of their lifetimes. At the lowest accretion rates, these systems may represent the transition between active nuclei and dormant black holes, but whether they can still launch collimated jets remains unclear. The nuclei in our Galaxy (\sgra) and M31 are key examples of this regime, although no clear jet structure has yet been detected in either source. Here we report multi-frequency very long baseline interferometric observations of \Msixty\ (NGC~4649), a nearby elliptical galaxy hosting a nearly dormant SMBH with an Eddington ratio of $\sim10^{-8}$. We detect a compact two-sided jet with an unusually steep synchrotron spectrum, demonstrating that collimated outflows can persist even under nearly dormant accretion conditions. The apparent radio core exhibits an unprecedentedly steep frequency-dependent position shift toward the SMBH, locating the central engine only $\sim57\,μ$as, corresponding to a projected distance of $\sim10$ Schwarzschild radii, upstream of the 8.37-GHz core. The observed jet morphology and steep core-shift behaviour are reproduced by general relativistic magnetohydrodynamic and radiative-transfer simulations, indicating a magnetically dominated, non-equipartition jet-launching region that departs from the standard conical equipartition picture. These results provide direct observational evidence that jet production can survive near the dormant SMBHs and establish \Msixty\ as a unique laboratory for probing jet formation on event-horizon scales in the lowest-accretion SMBH regime.

The rotational velocity of interstellar gas in the Milky Way, and other galaxies, has been taken to represent the circular velocity of a test particle in the Galaxy gravitational field, and hence an indicator of the Galaxy mass. The derived circular velocity is found to be too high for the gas to be gravitationally bound to the galaxy given the observed Galaxy mass in stars and gas, and consequently an extra component of mass in the Milky Way and other galaxies, namely dark matter, has been postulated. However recently the observational satellite Gaia, has been carrying out ground-breaking astrometric observations to accurately measure, inter alia, the three dimensional velocities of stars in the vicinity of the Sun and beyond. This has revealed that the circular velocity derived from the stellar population is much less than that of the gas, and the rotation curve, circular velocity versus radius, is distinctly declining with radius, whereas the gas rotation curve is not declining. By combining results from multiple observations of the gas velocity, averaging the velocities in radial bins, we establish that there is a grand average rotation curve. This can be compared directly with a grand average of the published Gaia rotation curves, and the confidence level in the difference between the two estimated by statistical analysis. The difference is shown to have a high degree of confidence, and increases with galactocentric radius. The lower circular rotation curve from the stellar velocities has resulted in significantly reduced estimates of the dark matter mass fraction of the Milky Way. The higher rotation of the gas lacks an explanation, but it is unlikely to be an accurate indicator of the kinematic mass of the Galaxy. This also has significant consequences for the mass of external galaxies based on gas rotation curves.

A recent analysis of the centrifugal instability in the case of pressure-supported compressible relativistic rotation, with application to astrophysical jets, yielded a generalisation of the famous Rayleigh criterion for Newtonian flows. According to this criterion, the centrifugal instability is strongly affected by the flow Mach number, and not only in the relativistic fluid dynamics but also in its Newtonian limit. To validate the Newtonian version of this criterion, we performed axisymmetric numerical simulations of non-relativistic transonic rotating flows which are stable according to the original Rayleigh criterion but can be either stable or unstable according to the new one. The results of computer simulations are found to be in perfect agreement with the theory. The hydrodynamic stability of accretion disks is often explained by referring to the original Rayleigh criterion, even if their rotation is highly supersonic. To clarify the matter, we analysed the hydrodynamic stability of flows rotating about central compact object and derived an instability criterion that retains the explicit dependence on the flow Mach number. This criterion turns out to be equivalent to the standard Solberg-Høiland criterion, which does not involve the Mach number. The same applies to the case of pressure-supported rotation, where the role of gravity is played by the centrifugal force.

Dielectronic recombination (DR) is expected to be the dominant recombination process during the non-local thermodynamic equilibrium (non-LTE) phase of kilonovae, yet reliable DR data remain unavailable for most heavy ions. Current spectral models therefore rely on simplified recombination prescriptions, introducing significant uncertainties into predicted spectra. We present an optimization strategy for open f-shell ions using \texttt{AUTOSTRUCTURE}, targeting uranium ions U II--U IV relevant to kilonova ejecta. As a benchmark case, calculations are performed for Nd III to validate the treatment of the f-shell structure and its impact on DR. The resulting DR rate coefficients are of order $10^{-10}$--$10^{-12}$ cm$^{3}$s$^{-1}$ over temperatures relevant to kilonova plasmas. The optimized rates are intended for implementation in radiative-transfer calculations with \texttt{SUMO} to assess the sensitivity of kilonova spectra to improved recombination physics. The Nd III benchmark demonstrates that refinements to the atomic structure can produce measurable changes in spectral features, motivating similar calculations for actinide ions.

We analyze the geometry and kinematics of substructures within molecular clouds identified in an unbiased catalog from the MWISP survey. These substructures are defined as spatially connected regions enclosed by the 20% peak-integrated-intensity contour of each cloud. After applying selection criteria on voxel size and excluding structures truncated by map boundaries, we construct a sample and quantify their projected morphology using the projected scale ratio $R=Δb/(Δl\cdot\cos b)$. This ratio essentially measures $\tanθ$ where $θ$ is the plane-of-sky angle of an elongated filament relative to the Galactic plane. The resulting sample exhibits a median $R=0.96$, indicating a slight but systematic preference for elongation along Galactic longitude. This tendency becomes more pronounced at larger spatial scales. We further investigate the relative orientations among the structural major axes, velocity-gradient directions, and plane-of-sky magnetic-field orientations derived from Planck data for a subsample of well-defined structures. We find that, for cloud structures within our sample, with physical scale $\sim 0.3$ to $\sim 30$ pc, velocity gradients tend to be perpendicular to the major axes, while magnetic-field are generally aligned parallel to them. This scale range differs from those typically probed in studies of dense cores ($\sim 0.05$ pc) and GMC-scale structures ($\gtrsim$ 10 to 100 pc), which have reported scale-dependent variations in relative orientations. In addition, the alignment between velocity gradients and magnetic fields shows a gradual weakening with increasing physical scale. These results suggest that the observed anisotropy of molecular cloud substructures may arise from a combination of large-scale Galactic dynamics, anisotropic gas motions, and magnetic fields, with the relative importance of these effects varying with scale.

We perform a Bayesian inference analysis of an inflationary model based on an inverse-tangent potential, incorporating reheating dynamics in both constant and dynamical equation-of-state (DEOS) frameworks. Using Planck and ACT constraints on the scalar spectral index, we find preferred values $κ\simeq0.5-0.6$ and $N_k\simeq40-60$, leading to reheating temperatures $T_{RH}\sim10^{10}-10^{14}$ GeV and reheating durations $N_{RH}\sim3-36$ e-folds. Reheating weighted $H_0$ posteriors shift the Planck inference towards the ACT preferred region through the intrinsic $n_s-H_0$ degeneracy of the CMB likelihood. In the DEOS framework, reheating with a constant decay rate yields $N_{RH}\simeq4-8$ e-folds and $T_{RH}\simeq10^{13}$ GeV, while a dynamical decay rate produces a strong dependence on the Yukawa coupling $y$, with $N_{RH}$ varying from $\mathcal{O}(30)$ to $\mathcal{O}(1)$ e-folds and the reheating temperature spanning $\sim10^{-2}-10^{14}$ GeV. Imposing inflation-reheating consistency significantly restricts the viable parameter space to a narrow region around $n_s\simeq0.9720-0.9725$ and $r\simeq0.026-0.060$, demonstrating that reheating dynamics provide a nontrivial bridge between early-universe inflation and late-time cosmological parameter inference.

Reliable estimates of stellar effective temperature ($T_{\mathrm {eff}}$) are fundamental to stellar population studies and Galactic astrophysics. However, the majority of stars observed in modern large-scale photometric surveys lack spectroscopic measurements, making empirical colour--$T_{\mathrm {eff}}$ relations essential tools. In this work, we present updated empirical colour--$T_{\mathrm {eff}}$ calibrations based on Sloan Digital Sky Survey (SDSS) $ugriz$ photometry combined with 2MASS $JHK_{\mathrm s}$ data. Effective temperatures are determined on a homogeneous InfraRed Flux Method (IRFM) scale using a combined sample of 3902 GALAH and 2535 APOGEE stars with high-quality photometry and well-characterised atmospheric parameters. Using this dataset, we establish empirical relations between $T_{\mathrm {eff}}$ and colour indices constructed from SDSS and 2MASS combinations. We provide both colour--metallicity--$T_{\mathrm {eff}}$ and colour--$T_{\mathrm {eff}}$ relations for dwarfs and giants. The calibrations are derived using low-order polynomial models with iterative $3σ$ clipping. Their performance depends on the adopted colour index, with long-baseline colours such as $(g-K_{\mathrm s})_0$ and $(g-z)_0$ achieving internal precisions of $\sim$30--50~K. Comparisons with previous calibrations show general agreement, with differences attributable to sample selection, photometric zero-points, and functional form. The resulting relations provide a homogeneous and internally consistent framework for estimating $T_{\mathrm {eff}}$ from SDSS and 2MASS photometry alone, and are well suited for application to large photometric surveys lacking spectroscopic information.

Accurate photometric zero-points are essential for translating observed magnitudes into physical fluxes, from comparing with models to ensuring consistency across surveys. We determine the zero-points needed to place the Sloan Digital Sky Survey (SDSS) $ugriz$ system on its nominal AB definition, by exploiting the sensitivity of the Infrared Flux Method (IRFM) to broadband flux calibration. Using benchmark effective temperatures for over 6,000 FGK-type stars, we invert the method to identify the zero-point corrections required for SDSS photometry to reproduce the adopted temperature scale. The $r$ band is found to be very well standardized, while the $i$ and $z$ bands show offsets of a few hundredths of a magnitude, consistent with previous studies. We also find a small offset in the $g$ band. The largest discrepancy occurs in the $u$ band, where the derived offset depends strongly on the adopted filter transmission curves, in particular whether one uses the original definition commonly adopted in the literature or the updated measurements that account for the presence of a red leak. This effect introduces a colour-dependent zero-point offset that becomes apparent when using a sample of late-type stars. Independent comparisons with CALSPEC spectrophotometric standards and Gaia XP spectra broadly support the offsets derived from the IRFM analysis. Our results provide a revised set of SDSS zero-points anchored to the IRFM temperature scale and demonstrate that large stellar samples can be used to constrain photometric calibration. The methodology presented here offers a complementary approach to traditional spectrophotometric calibration and may prove useful for future large-scale surveys.

We investigate stationary halo-like solutions of the Gross-Pitaevskii-Poisson (GPP) system, which describes self-gravitating Bose-Einstein condensates with repulsive self-interactions, as a dark matter model. The boson mass $m_φ$, scattering length $a_s$, and total particle number $N$ are kept explicit, with $N$ treated as an independent macroscopic control parameter. Solving the stationary GPP equations over a broad parameter space, we identify ground-state, excited-state, and unbound solution branches according to their binding properties and nodal structure. The ground-state branch occupies a well-defined region of the $(m_φ,N)$ plane whose location depends strongly on the self-interaction strength, whereas the excited-state and unbound regions are largely insensitive to the initial ansatz. From the converged solutions, we derive empirical scaling relations connecting the characteristic halo radius $R_{99}$ to $m_φ$, $a_s$, and $N$. In the weakly interacting regime, the results reproduce the standard Schrodinger-Poisson mass-radius relation, while finite self-interactions reveal an intermediate regime in which gravity, quantum pressure, and repulsive interactions jointly determine the equilibrium structure. As an astrophysical application, we show that ground-state solutions can reproduce representative dwarf-galaxy rotation curves using only the solitonic component. We also examine the implications of current Lyman-$α$ forest constraints and find that, although increasing $a_s$ shifts equilibrium solutions toward larger boson masses compatible with existing bounds, the resulting configurations do not reproduce the observed dwarf-galaxy kinematics. These results provide a systematic characterization of stationary GPP halos and establish a direct connection between microscopic particle properties and observable galactic quantities.

We report on NICER and MAXI observations of a long-duration thermonuclear X-ray burst and a subsequent outburst-like X-ray flare from the neutron star low-mass X-ray binary MAXI J0911--655. Prior to the burst, the source was in a persistent low/hard state with a power-law-dominated spectrum ($Γ\sim 1.7$) and a mass accretion rate of $\sim 1\%$ of the Eddington limit. The long burst, detected by MAXI on 2020 May 22 (MJD 58991.7101), was rapidly followed up by NICER. From time-resolved spectroscopy of the cooling tail, we estimate an exponential decay time of $\approx43$ minutes, the ignition column depth of $\approx0.1\times 10^{12}~{\rm g ~cm^{-2}}$, the burst fluence of $\approx 1.1\times 10^{-4}~{\rm erg~cm^{-2}}$, and the total energy release of $\approx1.2\times10^{42}$ erg. Approximately one day after the burst onset, the 0.5-10 keV light curve unexpectedly re-brightened, initiating an outburst-like flare. During the peak of this flare, the persistent power-law flux increased from its pre-burst level of $\sim0.27\times10^{-9}~{\rm erg~cm^{-2}~s^{-1}}$ to $1.4\times10^{-9}~{\rm erg~cm^{-2}~s^{-1}}$. This flux enhancement was accompanied by significant spectral softening, with the photon index increasing to $Γ\sim 2.2$. Subsequently, the flux decayed and the source returned to its baseline low/hard state. The observed timescales and energetics suggest that intense irradiation from the long burst amplified the ongoing thermal-viscous accretion process. This heating drove an inside-out heating front that temporarily enhanced the mass accretion rate, providing compelling observational evidence of a thermonuclear burst directly modulating the accretion dynamics of its surrounding disk.

Ultraviolet spectroscopy with the Hubble Space Telescope (HST) provides the most direct and sensitive probe of the disk-circumgalactic medium (CGM) interface at radii of 20 kpc, where galaxies exchange gas, metals, and energy with their surroundings. Many of the key diagnostics of the multiphase circumgalactic medium -- including H I, O VI, C II-IV, Si II-IV, N V, Ne VIII, and other metal transitions -- lie in the ultraviolet and are inaccessible from the ground, making HST the only observatory capable of making the required observations. By measuring the physical (column density, density), chemical (metallicity, ionization structure), and kinematical properties of the gas at the disk-CGM interface, UV absorption-line spectroscopy reveals how galaxies acquire fresh fuel, recycle enriched material, and drive feedback into their halos. When combined with spectroscopic characterization of the host galaxy's stellar populations and the feedback they generate (outflow velocity, mass loading), we will establish a direct understanding of how stellar populations enable circulation of gas and metals through the galactic ecosystem. HST's ultraviolet (UV) spectroscopic capability provides the only comprehensive observational pathways for uncovering the physical drivers that regulate galaxy growth and evolution in the low-redshift Universe.

We present ALMA high-resolution ($\sim$0.25$^{\prime\prime}$/2 kpc) CO(5-4) and CO(4-3) observations of three $z\sim 3$ submillimetre-selected dusty galaxies from the ALESS survey. These data complement existing [sub]-kpc scale ALMA 870$μ$m continuum imaging and JWST NIRCam and MIRI imaging from the ALESS-JWST program, allowing us to trace the molecular gas, dust-obscured star formation, and stellar populations on similar spatial scales. We spectroscopically confirm that two of the sources lie at the same redshift and are likely interacting. We find that the molecular-gas distribution broadly follows the dusty star-forming structures seen in the 870$μ$m dust continuum imaging, but that the gas reservoirs are significantly more extended than the dust emission with a spatial extent comparable to the rest-frame near-infrared stellar emission. By modeling the kinematics for the two highest signal-to-noise sources, we find that the galaxies are well-fit by rotating disc models with high ratios of ordered to random motion ($V_{\rm{max}}/\overlineσ=5\pm1$ and $6\pm1$), although smaller-scale kinematic deviations cannot be ruled out at the current sensitivity and spatial resolution. Finally, utilizing the high-resolution 870$μ$m dust continuum and CO data, we investigate star-formation scaling relations on kpc-scales in these high-redshift galaxies. Assuming a constant CO-to-H$_{2}$ conversion factor and excitation ratio, we find that the data are offset from theoretical star-formation relation predictions that do not take turbulence into account, but consistent with gravo-turbulent models, thereby suggesting that turbulence plays a central role in regulating star formation at high redshift.

In the outer Solar System, for example in the Saturnian system, a planet's strong gravity attracts micrometeoroids and generates hypervelocity impacts on bodies such as rings and satellites. Micrometeoroids are seemingly non-icy, whereas the targets are typically icy, and both the impactor and the target may span a wide range of porosities. In this study, we perform three-dimensional iSALE simulations of hypervelocity impacts of rocky impactors onto icy targets, varying the impact angle and the porosities of the impactor and target ($φ_{\rm imp}$ and $φ_{\rm tar}$). We consider two end-member porosities (0% and 90%) for oblique ($45^\circ$) impacts. At an impact velocity of 30 km/s, characteristic of Saturn's rings, we find that the morphology of early-stage crater formation varies significantly with porosity, transitioning from deep-penetration, narrow-channel cavities ($φ_{\rm imp}=0$, $φ_{\rm tar}=90%$) to very shallow craters driven by near-surface vapor blowoff ($φ_{\rm imp}=90%$, $φ_{\rm tar}=0%$), with intermediate, more hemispherical cavity shapes when the porosities are comparable. Here, we focus on the thermodynamic fate of the impactor, which represents the exogenic material responsible for modifying the target surface. The impactor material is strongly heated and is efficiently vaporized regardless of the porosities of the impactor and target. However, the peak pressure and peak temperature experienced by the impactor vary by nearly an order of magnitude. These results imply that hypervelocity impacts occurring, for example, in Saturn's rings efficiently vaporize exogenic non-icy impactors upon impact, while the subsequent thermodynamic pathways $-$ such as condensation and chemical evolution $-$ may differ depending on the thermodynamic conditions. Our results are expected to be applicable to a variety of planetary systems.

We present the first spatially contiguous study of the physical and chemical structure of the Helix Nebula (NGC~7293, PNG 036.1-57.1) based on integral-field spectroscopy from the SDSS-V Local Volume Mapper (LVM). The wide-field observations provide nearly complete spectroscopic coverage of the nebula, enabling a spaxel-by-spaxel analysis of extinction, electron density and temperature, ionisation structure, and chemical abundances. We reconstruct calibrated datacubes from the LVM row-stacked spectra and measure 41 optical emission lines, including hydrogen, helium, and collisionally excited metal lines. The resulting maps reveal a strongly stratified nebula, with highly ionised gas traced by \heii~concentrated toward the central cavity, low-ionisation material dominating the bright shell, and neutral or transition-zone gas enhanced in the outer regions. The Helix is a low-density object, with typical electron densities of $\sim10^{2}\mathrm{cm^{-3}}$, and exhibits a non-uniform temperature structure, with variations of several thousand Kelvin across different ionisation zones. We derive a near-solar oxygen abundance, $12+\log(\mathrm{O/H})\simeq8.7$, consistent with spatially complete sampling. The central abundance pattern indicates a significant contribution from unobserved O$^{3+}$, suggesting that apparent abundance variations are primarily driven by ionisation effects rather than true chemical inhomogeneities. We also find evidence for a sulfur deficit of $\sim$1 dex, consistent with the planetary-nebula sulfur anomaly. The helium and nitrogen abundances place the Helix near the classical boundary of Type~I planetary nebulae, suggesting moderate chemical enrichment by its progenitor star.

The source of the optical/UV emission in tidal disruption events (TDEs) remains an enduring question in the field. Connecting the observed emission to the source is critical for both our understanding of these transients and for using TDEs to study the efficiency of super-Eddington accretion and black hole growth. To explore this connection, we ran time-dependent 1D radiation hydrodynamic simulations of TDE emission with the Sedona monte carlo radiative transfer code, focusing on the reprocessing paradigm. Our simulations follow a compact, evolving X-ray and EUV bright source and surrounding reprocessing outflow over multiple months, using luminosities and mass flow rates consistent with hydro simulations of tidal disruptions. We determine the efficiency of reprocessing as a function of time in this dramatically changing environment and reproduce key observables including timescales, luminosities, and color evolution. Notably, we see a strong wavelength-dependence in the emission timescale due to reprocessing effects. Early on there is an X-ray flare which quickly fades as material builds up and obscures the hot source. At the same time, the optical/UV luminosity begins to rise. Though the optical/UV light curve has a similar shape to the bolometric light curve, the optical peak is offset by $\sim$3 weeks from the bolometric peak due to the time required to build up the reprocessing layer. This implies that early time, high energy emission may be missed for TDEs discovered in optical surveys, and the initial disruption and mass return time to the black hole may occur earlier than optical light curves suggest.

The quasar Eigenvector-1/Main Sequence (E1/MS) provides a physically motivated empirical framework to organize the spectroscopic diversity of type~1 active galactic nuclei (AGN). In its optical plane, the full width at half maximum of H$β$ and the Fe\,II strength ratio $R_{\mathrm{FeII}}$ define a sequence that is primarily driven by Eddington ratio, with important secondary roles played by black-hole mass, orientation, spectral energy distribution, and chemical enrichment. The E1/MS framework is therefore well suited to identify highly accreting and possibly super-Eddington (SE) sources, usually associated with the extreme Population~A (xA) spectral types. We discuss why E1/MS is a useful tool to search for SE accretors among jetted AGN and, conversely, to place $γ$-ray-detected AGN in the broader context of quasar phenomenology. We summarize two complementary results: (1) some candidate SE accretors show radio properties such as high brightness temperature non-thermal cores or radio lobes} consistent with jet activity; and (2) a subset of low-redshift $γ$-ray narrow-line Seyfert~1 galaxies exhibit optical spectra consistent with xA or borderline-xA classification. We also expand the discussion of recent developments in E1/MS studies, including metallicity trends, the spectral energy distribution of xA quasars, and the role of highly accreting quasars as discovery tools for extreme accretion states, as probes of quasars at the reionization epoch, and as possible cosmological probes.

Following our previous work constraining interacting dark energy (IDE) models, which showed their potential to alleviate the Hubble tension, in this work we investigate the non-linear effects of the IDE scenario favoured by CMB and DESI observations. The implications of IDE for the $S_8$ tension remain unclear, since current weak-lensing and large-scale-structure analyses either exclude highly non-linear scales or model the non-linear regime using prescriptions calibrated within $Λ$CDM. We address this issue by implementing a fully self-consistent IDE pipeline. We perform N-body simulations of the IDE model with a transfer rate $Q=ξ{\cal H}ρ_x$ using a modified implementation of RAMSES. Since the dark matter Euler equation remains unchanged with respect to $Λ$CDM, the interaction can be incorporated through the modified background evolution and an effective time-dependent dark matter particle mass. We find scale-dependent deviations in the quasi-linear and non-linear regimes of the matter power spectrum, together with modifications to the density-field morphology and halo abundance. Our results show that the impact of IDE on quasi-linear and non-linear structure formation cannot be captured by standard $Λ$CDM-calibrated prescriptions, highlighting the importance of model-consistent non-linear modelling for future weak-lensing and large-scale-structure constraints on interacting dark energy cosmologies.

The interstellar medium (ISM) is a complex, multiphase medium, where disentangling the distribution of gas and magnetic field structure across different phases remains a considerable challenge. Recently, Faraday tomography enabled by broadband polarized radio observations has emerged as a promising probe of 3D ISM gas and magnetic field structures. However, the interpretation of these observations is obscured by our limited understanding of the different ISM components probed by the distinct Faraday depth features. In this work, we present a comprehensive multi-frequency ($\sim$300 MHz - 23 GHz) analysis comparing features in the Faraday-rotated, polarized synchrotron emission and HI structures over the full high-latitude (|b|>30 degrees) diffuse sky. Using measures of HI structure complexity along the line of sight (LOS), we observe enhanced depolarization across synchrotron radio frequencies in regions with high HI complexity characterized by multiple HI velocity components. We also find that the first and second moments of the Faraday depth spectra are linked to the underlying neutral gas structure. These results indicate that regions of the ISM that are dominated by neutral gas could directly contribute a significant portion of the diffuse synchrotron emission and Faraday rotation. These findings establish new observational constraints for Galactic magnetic field models that synthesize multiphase tracers into a single coherent picture.

We present gas-phase metallicities of star-forming galaxies at $z=1$-10 with deep JWST/NIRSpec spectra from the JADES full data release, Dark Horse, and OASIS programmes. We stack $\sim$1500 medium-resolution spectra, yielding detections of the [OIII]$λ$4363 auroral line down to $12+\log(\mathrm{O/H})=7.0$ to derive stack-based strong-line calibrations over the metallicity range $12+\log(\mathrm{O/H})=7.0$-8.7. At a fixed metallicity, our stacks exhibit [OIII]$λ$5007/H$β$ and [OIII]$λ$5007/[OII]$λλ$3726,3729 values generally lower than calibrations based on high-$z$ individual auroral-line emitters, suggesting an observational bias towards higher excitation introduced when requiring auroral line detections in individual spectra. Based on our new calibrations, we obtain canonical mass-metallicity relations (MZRs) at z$=$1-10, identifying a decrease in metallicities from $z\sim0$ to z$\sim$4-10, without significant change in slope. Moreover, we identify 50 promising candidates of extremely metal-poor galaxies (EMPGs) with $12+\log(\mathrm{O/H})=6.7$-7.3 (1-4\% solar metallicity) at $z=1.2$-9.1. The MZRs of EMPGs are characterised by a large scatter, with those having lower metallicities generally exhibiting lower sSFRs, opposite of what expected from the local Fundamental Metallicity Relation. These results support a stochastic star-formation history involving gas consumption/ejection and metal-poor inflow, strongly affecting metallicities of low-mass galaxies. Furthermore, we identify two Little Red Dots in our EMPG candidates, both exhibiting broad H$α$ and prominent Ly$α$, offering insights into the early black-hole growth in extremely metal-poor environments.

In this paper we present the chemical screening of the complex stellar population discovered in the Bulge Fossil Fragment Liller 1. This study is part of the Bulge Cluster Origin (BulCO) survey based on a Large Program at the ESO-VLT with the high resolution spectrograph CRIRES+. The survey is aimed at performing an unprecedented chemical screening of 17 stellar systems orbiting the Milky Way bulge, with the ultimate goal of unveiling their origin and true nature. We measured precise chemical abundances of iron, CNO, iron-peak, $α$- other light-elements, and neutron-capture elements for a sample of 30 red giant branch stars, kinematic members of Liller 1. The presented analysis provides the high-resolution spectroscopic proof of the complex chemistry of this massive stellar system, with multi-metallicity sub-populations of different ages that nicely fits into a self-enrichment scenario. We find no evidence for the Na-O anticorrelation associated with genuine globular clusters; rather the overall abundance trends are similar to those seen in the bulge field and in Terzan 5, providing definitive evidence of an in-situ formation of Liller 1 within the Galactic bulge.

After quenching galaxies keep growing in size across time, as established in the literature up to cosmic noon. In this work, we assemble one of the largest and most comprehensive multi-wavelength photometric sample of massive quenched galaxies at z > 3, counting 137 quiescent candidates within 825 sq.arcmin and redshift 3<z<7 across the well-studied extragalactic JWST fields. We model their surface brightness distribution across 5 bands mapping their UV-to-NIR rest-frame wavelength with Sersic profiles, and derive their sizes, concentrations, and ellipticities. The size-mass relation is consistent with previous studies. Size decreases with increasing redshift, in agreement with previous studies and extending them up to z_spec=4.9. The intrinsic scatter of the mass-size relation has no relevant dependence on the redshift and the filter used. Our multi-wavelength modeling reveals that size decreases with increasing observed wavelength, and the wavelength gradient decreases with increasing stellar mass. This result proves that in the NIR-bands massive elliptical galaxies appear more compact. The Sersic index does not show significant dependence on wavelength, independently of the stellar mass. Following a forward/backward Bayesian fit analysis to assess the significance of several parameters in predicting the size of our sample, we do not identify any significant secondary dependence of the size on axis ratio q, Sersic index n, UVJ-colors, and environment. The combination of stellar mass and redshift is sufficient to predict the size of quiescent galaxies at z>3, albeit with a large scatter. This suggests that the commonly used parameters of a Sersic distribution cannot explain the large intrinsic scatter around the stellar mass-size relation, suggesting that other physical quantities need to be taken into account to break the degeneracy between evolution paths across the galaxy population.

Magnetic fields permeate the multiphase interstellar medium (ISM), yet their phase-dependent structure remains poorly constrained by observations. Dust polarization and \ion{H}{1} emission together offer complementary probes of the plane-of-sky magnetic field and cold neutral medium (CNM) gas structure, respectively. Recent observational work has shown that in the diffuse ISM, the dust polarization fraction correlates positively with the CNM mass fraction ($f_{\rm CNM}$) but not with total \ion{H}{1} column density, suggesting a phase-dependent magnetic field geometry. Here, we use extremely high-resolution ($2048^3$) simulations of the turbulent, magnetized, multiphase ISM to investigate the physical origin of this trend. By constructing synthetic \ion{H}{1} and dust polarization maps, we directly compare our simulations to the observational results of \citet{Lei:2024}. We recover a positive $f_{\rm CNM}$-polarization correlation most clearly for sightlines intersecting fewer than $\sim$20 discrete CNM clouds, while the trend becomes weak or intermittent for larger cloud counts, consistent with the expectation that high-Galactic-latitude sightlines contain relatively few independent cold structures. We show that this correlation reflects genuine phase-dependent magnetic structure: CNM clouds tend to be elongated along the local magnetic field and, when normalized by column density, exhibit lower magnetic disorder than the warm neutral medium (WNM). We further demonstrate that apparent discrepancies between simulation- and observation-based measures of magnetic disorder arise from whether disorder is quantified per unit path length or per unit mass. Our results support a picture in which CNM structures host relatively ordered magnetic fields, producing higher polarization fractions along CNM-dominated sightlines in the diffuse ISM.

Event Horizon Telescope (EHT) images of black holes appear as rings with a brightness asymmetry. Here, we expand on our previous study of the asymmetry magnitude $a_1$ to study the position angle of the peak brightness asymmetry $\mathrm{PA}_1$ in general relativistic magnetohydrodynamic (GRMHD) models. For larger spin magnitudes ($a_{*}>0$ and $a_{*}\lesssim-0.5$), the mean $\mathrm{PA}_1$ falls within $1σ$ of the approaching limb of the black hole, regardless of viewing inclination, disk magnetization, or source. By comparing the $(a_1, \mathrm{PA}_1)$ distribution in M87* observations with models, we demonstrate that we can mildly disfavor low-magnitude spins and strongly disfavor all spin vectors that point toward Earth. The alignment of $\mathrm{PA}_1$ relative to the large-scale jet axis may suggest that M87*'s disk does not have a large tilt. By combining $\mathrm{PA}_1$ with the pattern speed measured in optimistic 2026 M87* video conditions, the EHT can constrain whether M87* is prograde or retrograde with $\sim 84\%$ accuracy. In Sgr A*, we show that a detection of $(a_1, \mathrm{PA}_1)$ could constrain the magnitude and direction of the galactic center spin vector. Finally, if future EHT expansions increase the sample of horizon-scale sources, a simple set of observables (ring diameter, asymmetry magnitude, and asymmetry angle) could enable robust constraints on black hole mass, spin, and inclination.

The properties of spiral galaxies in the early universe remain poorly studied and, as such, little is known about their nature and evolution. We use JWST data to measure the pitch angles of spiral galaxies across cosmic time. Our sample consists of 593 spiral galaxies with stellar masses ($M_*$) greater than $10^{10} M_\odot$ up to $z \sim 3.5$, drawn from the CEERS and JADES surveys. Spiral galaxies are identified by fine-tuning a Zoobot deep-learning model. We use SpArcFiRe to identify spiral arms and measure their pitch angles. We find no significant redshift evolution in the average pitch angle across the full sample. However, in the most massive systems (log$(M_*/M_\odot)=11-12$), spiral arms slightly wind up with time. We show that at $z>1.25$, pitch angle does not correlate with some key internal galaxy properties (stellar mass, bulge mass, disk mass, specific star formation rate [sSFR]). In contrast, at $z<1.25$, pitch angle shows a weak but statistically significant negative correlation with stellar mass, bulge mass, and disk mass, and a positive correlation with sSFR at $z<0.75$. We also find no dependence of pitch angle on the tidal strength applied by nearby companions. These results indicate a transition epoch at $z\sim1$: above this redshift, spiral structures appear to be primarily locally driven and not correlated with global galaxy properties; and below this redshift, spiral arms are regulated by global gravitational potential, consistent with the predictions of the density wave theory.

Low surface brightness tidal debris around galaxies, such as tails, streams, and shells, together with other interaction-driven morphological disturbances, serve as valuable indicators of past or ongoing galaxy mergers. With the growing data volume from surveys like the Vera C. Rubin Observatory's Legacy Survey of Space and Time (LSST), automated detection methods are essential. This paper evaluates the performance of two automated methods, a Self-Supervised Learning (SSL) model and the Concentration-Asymmetry-Smoothness (CAS) parameter method, in tracing interaction-driven disturbances and merger signatures, with visual classification used as the benchmark. Visual classification yields a high-confidence disturbance fraction of 25.1 +/- 1.5% in our sample and serves as the reference standard for assessing the completeness and precision of the automated approaches. Visual classification is affected by galaxy distance and image resolution, which limit the detectability of faint low surface brightness structures. The SSL model achieves high recall (0.86 +/- 0.04) and low contamination (0.2) by retraining only its linear classifier on a small labelled dataset, making it suitable for identifying a broad set of disturbed systems, including faint tidal debris and other interaction-driven morphological disturbances, thereby providing a more complete census of merger-related features. The CAS method, using the traditional threshold A > 0.35, shows higher precision (0.77) but lower recall (0.20), indicating a conservative approach that captures cleaner but less complete samples. Visual classification and the SSL model show a significant positive correlation between stellar mass and disturbance fraction, while the CAS method exhibits a much weaker trend.

We present cosmological constraints from the Dark Energy Survey Year 3 (DES Y3) weak lensing data using hierarchical hybrid statistics within a Bayesian simulation-based inference framework that is based on the Gower Street simulations. To maximize the precision of the inference, we have developed a new, information-theory based, data compression of the weak lensing maps to just seven highly informative summary statistics. The hybrid scheme exploits the high information content of the power spectrum, compressing both the power spectrum and neural-based summaries that are designed to extract further information. Our simulation-based approach enables principled forward modelling of all major sources of systematic uncertainty and survey properties into realistic mock observations, including the survey mask, photometric redshift uncertainties, intrinsic galaxy alignments, multiplicative shear calibration bias, source galaxy clustering, non-Gaussian shape noise, and non-linear structure formation. The summary statistics are then used in a Bayesian simulation-based inference pipeline. The inference is validated through coverage tests and checks for robustness against baryonic feedback. Assuming a $w$CDM cosmology, our analysis yields $S_8 = 0.808 \pm 0.017$, $Ω_{\rm m} = 0.325 \pm 0.024$, and $w < -0.766$ (marginalized posterior 68 per cent credible intervals). This rigorous combination of information theory, physics- and neural network-based extreme data compression, and principled Bayesian analysis improves the figure of merit for $(Ω_{\rm m}, S_8, w)$ by 60 per cent over the previous state-of-the-art, and by almost a factor of 3 over two-point analyses of the same data. They are the most precise joint constraints on $(Ω_{\rm m}, S_8, w)$ from weak gravitational lensing data alone of any survey to date. We intend to apply this analysis to the more recent DES Y6 data.

The physical processes that shape a galaxy's spectrum are strongly degenerate in observations, obscuring which processes act independently. Leveraging the pop-cosmos generative galaxy population model, we investigate how many independent degrees of freedom the rest-frame optical SED contains. We use a $β$-variational autoencoder (VAE) to compress a 16-parameter stellar population synthesis (SPS) description into a disentangled latent representation interpreted through mutual information (MI). We find that five independent dimensions suffice, corresponding to stellar mass, recent star formation, dust, and two -- not the three or four assumed by standard nebular models -- degrees of freedom in the ionization state of the gas. Stellar metallicity and stellar age are not among these primary drivers; their spectral effects are distributed across the others rather than independently encoded. By tying each dimension to specific spectral features, this decomposition breaks the star-formation--dust--metallicity degeneracies that limit broadband photometry, and recovers the physical conditions of the gas more cleanly than the line-ratio diagnostics in standard use.

Synchrotron self absorption (SSA) is seen across a variety of astrophysical sources, and observation of an SSA peak in the spectrum is a powerful tool for estimating the physical conditions and the minimum energy of the emitting region. We begin with the (re)derivation of the usual SSA parameter estimates, carefully considering dependencies and assumptions, obtaining the most accurate traditional SSA minimum energy equations currently available. Traditional methods rely on the assumption that the emitting region is quasi-spherical and homogeneous. However, many observations of SSA show that the spectral index at frequencies below the peak is less than the expected $+2.5$ (non-thermal) or $+2$ (thermal). We argue that an inhomogeneous emitting region is the most likely explanation in many cases. Power law inhomogeneous cylindrical slab and broken power law inhomogeneous sphere models are used to investigate how the presence of inhomogeneity affects parameter estimates using traditional SSA methods. We find that in some cases inhomogeneity can lead to traditional SSA methods underestimating the minimum energy and the size of the emitting region by over an order of magnitude. Quantitative correction factors are found which can be applied to traditional estimates to correct for inhomogeneity, depending on the value of the observed flattened spectral index and the range in frequency over which this value is observed. Furthermore, we derive simple correction factors for non-spherical homogeneous emitting regions. Finally, we explore the effects of inhomogeneity on measurements of polarisation around the spectral peak, and on lightcurves for expanding emitting regions.

For broadband imaging surveys, filter-substrate refraction causes light at different wavelengths to follow slightly different paths through the filter substrate before reaching the detector. This effect produces two chromatic perturbations to the point spread function (PSF): a shift in the effective focal position along the optical axis (longitudinal shift), which manifests as a defocus-like perturbation, and a wavelength-dependent displacement of the image position in the focal plane (lateral shift), which manifests as image decentering. Using image simulations, we provide the first study of these two effects independently across all eight Roman imaging bands and over the full focal plane. We compute the resulting PSF and photometric errors from images with and without the effect included, and compare the magnitude of the effect to the Roman science requirements. We find that the lateral shift is the dominant contribution, producing PSF size and ellipticity residuals in most bands of order ~0.3-0.4%. These exceed the Roman science requirements for weak lensing by roughly an order of magnitude. The effect is also strongly field dependent, increasing toward the edges of the focal plane. By contrast, flux residuals remain below one third of the 1% requirement for most bands, except in R062 and W146. We find the longitudinal shift to be subdominant and negligible in most bands, including the weak lensing bands. Finally, we implement the dominant lateral-shift effect in a framework suitable for large-scale image simulations and validate that the resulting PSF size and shape changes are accurately reproduced. Overall, we find that filter-substrate refraction is a relevant chromatic effect for Roman PSF modeling, and we provide tools to model and incorporate it in large-scale image simulations.

Parity-violating physics in the Universe can generate correlations between the Cosmic Microwave Background (CMB) $E$- and $B$-modes, but detecting such signals requires extremely accurate calibration of instruments. We describe a data-driven method to calibrate the relative polarisation angle between CMB experiments using cross-correlations of observations over a common sky region. Unlike standard self-calibration approaches, this method does not assume vanishing isotropic cosmic birefringence or primordial $EB$ correlations when estimating the relative misalignment angle, and therefore preserves sensitivity to parity-violating physics. As a proof of concept, we forecast the performance of this method using the Simons Observatory (SO) Small Aperture Telescopes (SATs) as a calibrated reference. If they can be calibrated to an uncertainty of $0.08^\circ$, as anticipated from the SO wire grid calibration system, we show that the SO Large Aperture Telescope and Planck could be calibrated to uncertainties of $0.10^\circ$ and $0.17^\circ$, respectively, at $\sim 145$ GHz. This approach relies on the availability of at least one well-calibrated instrument, and provides a complementary path to improving polarisation calibration across experiments, enabling more robust searches for parity-violating physics in the CMB, such as cosmic birefringence.

Based on its predominantly thermal X-ray emission and long spin period, the isolated neutron star eRASSU J131716.9-402647 is one of the most promising candidates for membership in the still small class of X-ray dim isolated neutron stars (XDINSs). Confirmation of this classification, however, requires a more detailed characterisation of the source's timing and spectral properties. In this work, we present new NICER observations which, together with previous X-ray follow-up, allow us to constrain the timing properties and long-term evolution of eRASSU J131716.9-402647. We obtain a coherent timing solution with a spin period of $P\sim12.8$ s and a period derivative of $\dot{P}\sim9\times 10^{-14}$ s s$^{-1}$, which best-describes the spin evolution of the source. These parameters imply a dipolar magnetic field strength of $3\times10^{13}$ G and a spin-down luminosity of order $10^{30}$ erg s$^{-1}$. Spectral modelling reveals no significant change in the spectral state over the 15 months of observational monitoring and indicates a thermal luminosity that likely exceeds the rotational energy loss. This suggests a thermal evolution that has been significantly influenced by past reheating. The energy dependence of the double-humped pulse profile closely resembles that observed in the XDINS RX J1308.6+2127, with the pulsed fraction increasing towards higher energies. Taken together, these results unambiguously confirm the XDINS nature of eRASSU J131716.9-402647, making it the first newly confirmed XDINS in more than two decades.

We present a broadband spectral analysis of the neutron star LMXB GRS 1747-312 using $\sim$ 40 ks AstroSat data. The source was observed during the decay phase of the 2017 outburst, with an absorbed 1.0-5.5 keV flux of 1.67$^{+0.04}_{-0.07}\times$10$^{-11}$ erg s$^{-1}$ cm$^{-2}$, corresponding to a luminosity of $\sim$(0.9-1.80)$\times$10$^{35}$ erg s$^{-1}$. The continuum is modeled with thermal Comptonization of blackbody emission and interstellar absorption. A mildly broad iron line at $\sim$6.4 keV is fitted with a disc reflection component. Narrow lines below 2 keV are described by a hot plasma using the XSPEC model APEC. Additionally, there is a potential detection of an emission line at 4.19$^{+0.12}_{-0.10}$ keV with width $σ$ = 0.2 $\pm$ 0.2 keV and line flux of 13$^{+10}_{-9}\times$10$^{-5}$ erg s$^{-1}$ cm$^{-2}$. Examination of several short-duration ($\sim$ few kiloseconds) Swift observations at a few times the AstroSat source flux provided upper limits to the line flux of $<$30$\times$10$^{-5}$ erg s$^{-1}$ cm$^{-2}$. The 4.2 keV line likely originates from reflection off the neutron star surface. Shifting the neutral Fe K$_α$ line from its rest energy of 6.4 keV to 4.2 keV requires a redshift of z $\sim$ 0.6, consistent with that expected from the surface of a non-spinning 1.4 M$_\odot$, 10 km radius neutron star. If confirmed, this feature provides a potential direct measurement of gravitational redshift, allowing us to place strong constraints on the neutron star's mass-to-radius ratio and gain valuable insights into the equation of state (EOS) of dense matter.

JWST discoveries of the earliest ($z \gtrsim 9$) supermassive black holes (BHs, $M_\bullet \gtrsim 10^6\,\rm{M}_\odot$) challenge the BH seeding and accretion models of most cosmological simulations. In this work, we compare early BH growth arising from three different accretion prescriptions characterized by distinct scalings between the accretion rate ($\dot{M}_{\rm \bullet}$) and the BH mass ($M_{\rm \bullet}$): the commonly used Bondi-Hoyle model ($\dot{M}_{\rm \bullet}\propto M_{\rm \bullet}^2$), and two free-fall models with shallower scalings ($\dot{M}_{\rm \bullet}\propto M_{\rm \bullet}^{1/2}$ and $M_{\rm \bullet}$). Bondi accretion tends to produce stronger runaway growth than the free-fall models when using heavy ($\sim10^5\,\rm{M}_\odot$) seeds in extreme environments owing to the steeper $M_\bullet$ scaling, but its sensitivity to the local gas sound speed makes it more susceptible to suppression from temperature increases due to AGN and stellar feedback. The free-fall models tend to produce stronger growth for lower-mass seeds ($\sim10^{3-4}\,\rm{M}_\odot$) in moderate environments as they are less dependent on the BH's mass to accrete effectively, however in this regime BH growth remains negligible for all accretion models in the presence of fiducial stellar feedback. Enhancing early BH growth via many BH-BH mergers disproportionately enhances subsequent accretion-driven growth for Bondi due to the steeper $M_{\rm \bullet}$ dependence. Our simulations can thus assemble BHs with masses of $\sim10^6-10^7~M_{\odot}$ at $z\gtrsim9$, as inferred by JWST, under two circumstances: 1) abundant heavy-seed formation that drives BH-BH mergers, or 2) Bondi accretion with weak feedback.