arXiv Daily Digest - 2026-05-01

Numerous metascience studies and other initiatives have begun to monitor the prevalence of open science practices when it is more important to understand the 'downstream' effects or impacts of open science. PLOS and DataSeer have developed a new LLM-based indicator to measure an important effect of open science: the reuse of research data. Our results show a data reuse rate of 43%, which is higher than established bibliometric techniques. We show that data reuse can be measured at scale using LLMs and generative artificial intelligence. The positive effects of research data sharing and reuse may currently be underestimated.

Spiking neural networks (SNNs) are a promising paradigm for energy-efficient event-driven computation, but large-scale SNN execution remains challenging because sparse spike communication and synchronization can dominate runtime. Existing solutions across CPU, GPU, ASIC, and FPGA platforms offer different trade-offs between programmability, efficiency, and scalability. To address this gap, we present NeuroRing, a modular and scalable SNN accelerator based on a stream-dataflow architecture and a bidirectional ring topology, implemented in High-Level Synthesis (HLS) on programmable FPGAs. NeuroRing supports modular single- and multi-FPGA deployment and is compatible with existing SNN workflows through integration with the NEST simulator. We evaluate NeuroRing on the cortical microcircuit benchmark and a Sudoku constraint-satisfaction workload. Results show that NeuroRing preserves the key activity statistics of the NEST reference model, achieves faster-than-real-time execution of the full-scale cortical microcircuit with a real-time factor (RTF) of 0.83, exhibits meaningful strong and weak scaling, and provides competitive energy efficiency on two programmable FPGAs. These results position NeuroRing as a flexible and scalable platform for both neuroscience simulation and broader event-driven applications.

As autonomous vehicles slowly deploy into urban roads for limited use cases with significant edge case issues, closed facilities like marshaling yards provide a ripe case for combining lower-level vehicle autonomy with fixed infrastructure to create full autonomy without similar edge case concerns. Within a delivery marshaling yard, electric fleet vehicles complete a set of sequential tasks (charging, inspection, cleaning, and loading) before exiting the yard with their new load of deliveries. Hybrid automation of the vehicles and infrastructure can allow these vehicles to reach full autonomy and navigate the facility without the need of a driver, allowing for quicker movement between tasks increasing vehicle throughput. However, isolated autonomous operations based on static rules are prone to gridlock causing facility failures that temporarily shut down operations. Our orchestrated autonomy solution uses decentralized, dynamic priority scoring of vehicles based on the current status of the marshaling yard to optimally assign vehicles to tasks to increase vehicle throughput. Using a simulated facility with three marshaling yard sizes (small, medium, and large) and three demand levels (low, medium, high), we demonstrated that our orchestration solution increases vehicle throughput above static, isolated autonomy for all combinations of yard size and demand, while reducing facility failures at high demand levels.

Large language models (LLMs) make reward design in reinforcement learning substantially more scalable, but generated rewards are not automatically reliable training objectives. Existing work has focused primarily on generating, evolving, or selecting reward candidates, while paying less attention to when such candidates can be verified and deployed during policy optimization. We study this deployment-time problem by treating generated rewards as reward hypotheses whose utility depends on the competence of the current policy and the phase of training. We propose \textsc{RHyVE}, a competence-aware verification and phase-aware deployment protocol that compares small sets of reward hypotheses from shared policy checkpoints using short-horizon fork verification. Our experiments show that reward rankings are unreliable at low competence but become informative after task-dependent thresholds. On a sparse manipulation task, phase-aware deployment improves peak and retained performance under a locked protocol. Updated LLM-generated reward-candidate experiments show candidate-family-dependent behavior: generated pools can exhibit phase-dependent winner changes, but no fixed warm-up schedule is universally optimal. Held-out schedule selection, conservative selector baselines, compute-matched controls, and scale controls further show that \textsc{RHyVE} is best understood as a verification-informed deployment protocol rather than a universal scheduler. Dense and all-failure boundary experiments delimit the scope of the method. Together, these results suggest that reward generation and reward deployment should be studied as coupled problems: generated rewards must be verified and deployed under changing policy competence.

Individualized Alzheimer's disease (AD) progression prediction requires models that use irregular visits, account for censoring, avoid diagnostic leakage, and provide calibrated horizon risks. We propose PROgression-aware MultI-horizon Survival Estimation for Alzheimer's Disease (PROMISE-AD), a leakage-safe survival framework for predicting conversion from cognitively normal (CN) to mild cognitive impairment (MCI) and from MCI to AD dementia using ADNI/TADPOLE tabular histories. PROMISE-AD converts pre-index visits into tokens with standardized measurements, missingness masks, longitudinal changes, time-normalized slopes, visit timing, and non-diagnostic categorical attributes. A temporal Transformer fuses global, attention-pooled, and latest-visit representations to estimate a progression score and latent discrete-time mixture hazards. Training combines survival likelihood, horizon-specific focal risk loss, progression ranking, hazard smoothness, and mixture-balance regularization, followed by validation-set isotonic calibration for 1-, 2-, 3-, and 5-year risks. In held-out testing across three seeds, PROMISE-AD achieved an integrated Brier score (IBS) of 0.085 $\pm$ 0.012, C-index of 0.808 $\pm$ 0.015, and mean time-dependent AUC of 0.840 $\pm$ 0.081 for CN-to-MCI conversion, yielding the lowest IBS among compared methods. For MCI-to-AD conversion, PROMISE-AD achieved the highest C-index (0.894 $\pm$ 0.018) and near-ceiling 5-year discrimination (AUROC 0.997 $\pm$ 0.003; AUPRC 0.999 $\pm$ 0.001), although some baselines had lower IBS. Ablations and interpretability supported longitudinal change features, fused temporal representations, mixture hazards, cognitive and functional measures, APOE4 status, and recent conversion-proximal visits. These findings suggest that progression-aware survival modeling can provide interpretable multi-horizon AD conversion risk estimates.

Responsible AI research typically focuses on examining the use and impacts of deployed AI systems. Yet, there is currently limited visibility into the pre-deployment decisions to pursue building such systems in the first place. Decisions taken in the earlier stages of development shape which systems are ultimately released, and therefore represent potential, but underexplored, points for intervention. As such, this paper investigates factors influencing AI non-development and abandonment throughout the development lifecycle. Specifically, we first perform a scoping review of academic literature, civil society resources, and grey literature including journalism and industry reports. Through thematic analysis of these sources, we develop a taxonomy of six categories of factors contributing to AI abandonment: ethical concerns, stakeholder feedback, development lifecycle challenges, organizational dynamics, resource constraints, and legal/regulatory concerns. Then, we collect data on real-world case of AI system abandonment via an AI incident database and a practitioner survey to evidence and compare factors that drive abandonment both prior to and following system deployment. While academic responsible AI communities often emphasize ethical risks as reasons to not develop AI, our empirical analysis of these cases demonstrates the diverse, and often non-ethics-related, levers that motivate organizations to abandon AI development. Synthesizing evidence from our taxonomy and related case study analyses, we identify gaps and opportunities in current responsible AI research to (1) engage with the diverse range of levers that influence organizations to abandon AI development, and (2) better support appropriate (dis)engagement with AI system development.

Text-to-SQL (T2SQL) evaluation in production environments poses fundamental challenges that existing benchmarks do not address. Current evaluation methodologies whether rule-based SQL matching or schema-dependent semantic parsers assume access to ground-truth queries and structured database schema, constraints that are rarely satisfied in real-world deployments. This disconnect leaves production T2SQL agents largely unevaluated beyond developer-time testing, creating silent quality degradation with no feedback mechanism for continuous improvement. We present STEF (Schema-agnostic Text-to-SQL Evaluation Framework), a production-native evaluation system that operates exclusively on natural language inputs the user question, an enriched reformulation, and the generated SQL without requiring database schema or reference queries. STEF extracts semantic specifications from both natural language and SQL representations, performs normalized feature alignment, and produces an interpretable 0 to 100 accuracy score via a composite metric that encompasses filter alignment, semantic verdict, and confidence of the evaluator. Key contributions include: enriched question quality validation as a first-class evaluation signal, configurable application-specific rule injection via prompt templating, and production-robust normalization handling GROUP BY tolerance, ORDER BY defaults, and LIMIT heuristics. Empirical results demonstrate that STEF enables continuous production monitoring and agent improvement feedback loops without schema dependency, making structured query evaluation viable at scale for the first time.

Large Language Models (LLMs) are increasingly used as proxies for human perception in urban analysis, yet it remains unclear whether persona prompting produces meaningful and reproducible behavioral diversity. We investigate whether distinct personas influence urban sentiment judgments generated by multimodal LLMs. Using a factorial set of personas spanning gender, economic status, political orientation, and personality, we instantiate multiple agents per persona to evaluate urban scene images from the PerceptSent dataset and assess both within-persona consistency and cross-persona variation. Results show strong convergence among agents sharing a persona, indicating stable and reproducible behavior. However, cross-persona differentiation is limited: economic status and personality induce statistically detectable but practically modest variation, while gender shows no measurable effect and political orientation only negligible impact. Agents also exhibit an extremity bias, collapsing intermediate sentiment categories common in human annotations. As a result, performance remains strong on coarse-grained polarity tasks but degrades as sentiment resolution increases, suggesting that simple label-based persona prompting does not capture fine-grained perceptual judgments. To isolate the contribution of persona conditioning, we additionally evaluate the same model without personas. Surprisingly, the no-persona model sometimes matches or exceeds persona-conditioned agreement with human labels across all task variants, suggesting that simple label-based persona prompting may add limited annotation value in this setting.

We present Collaborative Agent Reasoning Engineering (CARE), a disciplined methodology for engineering Large Language Model (LLM) agents in scientific domains. Unlike ad-hoc trial-and-error approaches, CARE specifies behavior, grounding, tool orchestration, and verification through reusable artifacts and systematic, stage-gated phases. The methodology employs a three-party workflow involving Subject-Matter Experts (SMEs), developers, and LLM-based helper agents. These helper agents function as facilitation infrastructure, transforming informal domain intent into structured, reviewable specifications for human approval at defined gates. CARE addresses the "jagged technological frontier", characterized by uneven LLM performance, by bridging the gap between novice and expert analysts regarding domain constraints and verification practices. By generating concrete artifacts, including interaction requirements, reasoning policies, and evaluation criteria, CARE ensures agent behavior is specifiable, testable, and maintainable. Evaluation results from a scientific use case demonstrate that this stage-gated, artifact-driven methodology yields measurable improvements in development efficiency and complex-query performance.

We discuss a Quantum-Enhanced Computing Continuum, a heterogeneous, hybrid architecture that integrates quantum processing units (QPUs) within an Edge-Cloud-HPC fabric. Promote sustainability by shifting from performance to "energy-aware integration.' The architecture has three layers: a Physical Layer with shared fiber-optic infrastructure, a Control and Orchestration Layer managed by the user, and an Application Layer with an Adaptive Quantum Classical Fusion (AQCF) framework. Tighter system integration, like moving from cloud coupling to cryogenic logic, reduces energy waste and "thermal footprints.' The aim is a Green Performance Advantage: energy per problem solved in the era of Advanced Computing.

Spectra are a prevalent yet highly information-dense form of scientific imagery, presenting substantial challenges to multimodal large language models (MLLMs) due to their unstructured and domain-specific characteristics. Here we introduce SpecVQA, a professional scientific-image benchmark for evaluating multimodal models on scientific spectral understanding, covering 7 representative spectrum types with expert-annotated question-answer pairs. The aim comprises two aspects: spectra scientific QA evaluation and corresponding underlying task evaluation. SpecVQA contains 620 figures and 3100 QA pairs curated from peer-reviewed literature, targeting both direct information extraction and domain-specific reasoning. To effectively reduce token length while preserving essential curve characteristics, we propose a spectral data sampling and interpolation reconstruction approach. Ablation studies further confirm that the approach achieves substantial performance improvements on the proposed benchmark. We test the capability of prominent MLLMs in scientific spectral understanding on our benchmark and present a leaderboard. This work represents an essential step toward enhancing spectral understanding in multimodal large models and suggests promising directions for extending visual-language models to broader scientific research and data analysis.

Purpose: Fast detection of plant stress is key to plant phenotyping, precision agriculture, and automated crop management. In particular, efficient irrigation management requires early identification of water stress to optimize resource use while maintaining crop performance. Direct physiological sensing offers the potential to detect stress responses before visible symptoms appear. Methods: In this study, we recorded electrophysiological signals from greenhouse-grown tomato plants subjected to water stress and developed a framework based on machine learning for online stress detection. The recorded time-series data were processed using a processing pipeline that includes statistical feature extraction and selection, automated machine learning or alternatively deep learning, and probability calibration. Results: Across multiple input time horizons, we found that a 30-minute look-back window strikes the best balance between rapid decision-making and classification performance. Using automated machine learning, the framework achieved classification accuracies of up to 92%, outperforming deep learning approaches. Sequential backward selection reduced the feature set while maintaining performance. Importantly, the framework detects transitions from healthy to stressed states in recordings that were not included in the training set. Conclusion: Overall, we provide a decision-support tool for farmers and establish a foundation for biofeedback-driven irrigation control to improve resource efficiency in (semi-)autonomous crop production systems.

Exponential families encompass the distributions central to modern machine learning -- softmax, Gaussians, and Boltzmann distributions -- and underlie the theory of variational inference, entropy-regularized reinforcement learning, and RLHF. We isolate a simple identity for exponential families that expresses the KL difference $\mathrm{KL}(q \| p_{λ_2}) - \mathrm{KL}(q \| p_{λ_1})$ in terms of the log-partition function $A(λ)$ and the moment $μ_q$. Remarkably, this identity together with the single fact that $\mathrm{KL} \geq 0$ (with equality iff $p = q$) suffices, by direct substitution and rearrangement, to derive a cluster of results that are classically obtained by separate, heavier arguments: a generalized three-point identity for arbitrary reference distributions, Pythagorean theorems for I-projections and reverse I-projections, convexity of the log-partition function, identification of its Legendre dual in KL terms, the Gibbs variational principle, and the explicit optimizer in KL-regularized reward maximization, including the exponential tilting formula underlying entropy-regularized control and RLHF. Beyond these purely algebraic consequences, standard analytic arguments recover the gradient formula for the log-partition function, the Bregman representation of within-family KL divergence, and the surjectivity of the moment map. The note is self-contained.

The syntactic structure of a sentence can be represented as a tree where edges indicate syntactic dependencies between words. When that structure is a star, it has been demonstrated that the head should be placed in the middle of the linear arrangement according to the principle of syntactic dependency distance minimization. However, hubs of stars tend to be put at one of the ends, against that principle. Here we address two questions: (1) How difficult is it to minimize dependency distance? (2) Why anti dependency distance minimization effects have been found in star structures but not in path structures? The ease of optimization is determined by the shape of the optimization landscape. It was demonstrated that the landscape of star structures is quasiconvex (Ferrer-i-Cancho 2015, Language Dynamics and Change). As for (1), here we show that it is indeed convex (a particular case of quasiconvexity) both for star trees and quasistar trees and thus the distance-based optimization problem is simpler than previously believed. As for (2), we argue that (a) competing principles, rather than the difficulty of optimization, must be the actual reason for anti-dependency distance minimization effects and that (b) dependency distance minimization on star-like structures is less rewarding compared to other structures.

We study $d$-dimensional unbiased mean estimation in the single-message shuffle model, where each user sends a single privatized message and the analyzer only observes the shuffled multiset of reports. While minimax-optimal mechanisms are well understood in the local differential privacy setting, the corresponding notion of optimality after shuffling has remained largely unexplored. To address this gap, we introduce the recently proposed shuffle index and use it to formulate the post-shuffling mechanism design problem as an explicit optimization problem. We then establish a minimax lower bound on the achievable mean squared error in terms of the shuffle index, which implies that mechanisms that are optimal under LDP can become suboptimal once shuffling is applied. Finally, we construct an asymptotically minimax optimal mechanism in the high privacy regime, which as a consequence achieves a privacy-utility trade-off nearly identical to that of the central Gaussian mechanism.

When researchers iteratively refine ideas with large language models, do the models preserve fidelity to the original objective? We introduce DriftBench, a benchmark for evaluating constraint adherence in multi-turn LLM-assisted scientific ideation. Across 2,146 scored benchmark runs spanning seven models from five providers (including two open-weight), four interaction conditions, and 38 research briefs from 24 scientific domains, we find that iterative pressure reliably increases structural complexity and often reduces adherence to original constraints. A restatement probe reveals a dissociation between declarative recall and behavioral adherence, as models accurately restate constraints they simultaneously violate. The knows-but-violates (KBV) rate, measuring constraint non-compliance despite preserved recall, ranges from 8% to 99% across models. Structured checkpointing partially reduces KBV rates but does not close the dissociation, and complexity inflation persists. Human validation against blind raters confirms that the LLM judge under-detects constraint violations, making reported constraint adherence scores conservative. Sensitivity analyses confirm the findings are robust to temperature (0.7 vs.\ 1.0) and pressure type (novelty vs.\ rigor). We release all briefs, prompts, rubrics, transcripts, and scores as an open benchmark.

Fairness in machine learning remains challenging due to its ethical complexity, the absence of a universal definition, and the need for context-specific bias metrics. Existing methods still struggle with intersectionality, multiclass settings, and limited flexibility and generality. To address these gaps, we introduce MIFair, a unified framework for bias assessment and mitigation based on mutual information. MIFair provides a flexible metric template and an in-processing mitigation method inspired by the Prejudice Remover, defining group fairness as statistical independence between prediction-derived variables and sensitive attributes. We further strengthen its information-theoretic foundation by establishing equivalences with widely used fairness notions such as independence and separation. MIFair naturally supports intersectionality, complex subgroup structures, and multiclass classification and employs regularization-based training to reduce bias according to the selected metric. Its key advantage is its versatility: it consolidates diverse fairness requirements into a single coherent framework, enabling consistent benchmarking and simplifying practical use. Experiments on real-world tabular and image datasets show that MIFair effectively reduces bias, including previously unaddressed multi-attribute scenarios, while maintaining strong predictive performance across the evaluated settings.

Large language models (LLMs) have revolutionized Text-to-SQL generation, allowing users to query structured data using natural language with growing ease. Yet, real-world deployment remains challenging, especially in complex or unseen schemas, due to inconsistent accuracy and the risk of generating invalid SQL. We introduce Template Constrained Decoding (TeCoD), a system that addresses these limitations by harnessing the recurrence of query patterns in labeled workloads. TeCoD converts historical NL-SQL pairs into reusable templates and introduces a robust template selection module that uses a fine-tuned natural language inference model to match or reject queries efficiently. Once the template is selected, TeCoD enforces it during SQL generation through grammar-constrained decoding, implemented via a novel partitioned strategy that ensures both syntactic validity and efficiency. Together, these components yield up to 36% higher execution accuracy than in-context learning (ICL) and 2.2x lower latency on matched queries.

Federated Multi-Label Learning is a distributed paradigm where multiple clients possess heterogeneous multi-label data and perform collaborative learning under privacy constraints without sharing raw data. However, modeling label correlations under heterogeneous distributions remains challenging. Due to client-specific label spaces and varying co-occurrence patterns, correlations learned by individual clients inevitably deviate from the global structure, a phenomenon we term label correlation drift. To address this, we propose FedHarmony, a framework that harmonizes heterogeneous label correlations across clients. It introduces consensus correlation, capturing agreement among other clients and serving as a global teacher to correct biased local estimates. During aggregation, FedHarmony evaluates each client by both data size and correlation quality, assigning weights accordingly. Moreover, we develop an accelerated optimization algorithm for FedHarmony and theoretically establish faster convergence without sacrificing accuracy. Experiments on real-world federated multi-label datasets show that FedHarmony consistently outperforms state-of-the-art methods.

Cooking is a cultural expression of human creativity that transcends geography and time through the orchestration of ingredients and techniques, much like languages do through words and syntax. Yet, beneath the apparent diversity of culinary traditions, whether recipes obey statistical laws comparable to those of other symbolic systems remains unknown. Here we analyze a large corpus of traditional recipes spanning global cuisines, annotated using a state-of-the-art named entity recognition algorithm into ingredients, cooking techniques, utensils, and other culinary attributes. We find that ingredient usage exhibits Zipf-like rank-frequency scaling, that culinary diversity grows sublinearly with corpus size in accordance with Heaps' law, and that recipe complexity follows Menzerath-Altmann-type relations between the number and average information of constituent units. Consistent with observations in packaged foods, macronutrient concentrations across recipes also display a log-normal signature. Minimal generative models based on preferential reuse, constrained sampling, and incremental modification recapitulate these regularities, suggesting generic processes that shape recipe architecture across cultures. Together, these findings establish recipes as a compositional symbolic system in which complex structure emerges from simple, constrained generative processes.

We consider the problem of Cost-Aware Learning, where sampling different component functions of a finite-sum objective incurs different costs. The objective is to reach a target error while minimizing the total cost. First, we propose the Cost-Aware Stochastic Gradient Descent algorithm for convex functions, and derive its cost complexity to attain an error of $ε$. Furthermore, we establish a lower bound for this setting and provide a subset selection algorithm to further reduce the cost of training. We apply our theoretical insights to reinforcement learning with language models, where the computational cost of policy gradients varies with sequence length. To this end, we introduce Cost-Aware GRPO, an algorithm designed to reduce the cost of policy optimization while preserving performance. Empirical results on 1.5B and 8B LLMs demonstrate that our approach reduces the tokens used in policy optimization by up to about 30% while matching or exceeding baseline accuracy.

Design Structure Matrix (DSM) modularization, the task of partitioning system elements into cohesive modules, is a fundamental combinatorial challenge in engineering design. Traditional methods treat modularization as a pure graph optimization, without access to the engineering context embedded in the system. Building on prior work on LLM-based combinatorial optimization for DSM sequencing, this paper extends the method to modularization across five cases and three backbone LLMs. Our method achieves near-reference quality within 30 iterations without requiring specialized optimization code. Counterintuitively, domain knowledge, beneficial in sequencing, consistently impairs performance on more complex DSMs. We attribute this to semantic misalignment between the LLM's functional priors and the purely structural optimization objective, and propose the semantic-alignment hypothesis as a testable condition governing knowledge effectiveness with LLMs. Ablation studies identify the most effective input representation, objective formulation, and solution pool design for practical deployment. These findings offer practical guidance for deploying LLMs in engineering design optimization.

Constant Product Market Makers use fees that are typically fixed proportions of trade size. When these fees are automatically reinvested into the pool, as in Uniswap~V2 and some designs of Uniswap V4, the final state after a trade can depend on how the trade is split into smaller transactions. This path dependence complicates the risk assessment for liquidity providers and affects composability guarantees. We characterize the functional class of fee structures that ensure path independence: the combined fee factor must depend only on the current pool invariant k=xy. For this class, we derive a system of ordinary differential equations governing pool dynamics and obtain a closed-form integral exchange formula. Within this class, we construct a parametric family of fee functions that achieve zero Impermanent Loss for a given initial pool state, and prove that no universal fee function can eliminate Impermanent Loss for all initial states simultaneously. We analyze implications for arbitrage windows and slippage, and validate our theory through controlled simulations. Our framework provides protocol designers with a principled approach to fee optimization that aligns liquidity provider and trader incentives while preserving composability.

3D Gaussian Splatting has emerged as a powerful scene representation for real-time novel-view synthesis. However, its standard adaptive density control relies on screen-space positional gradients, which do not distinguish between geometric misplacement and frequency aliasing, often leading to either over-blurred high-frequency textures or inefficient over-densification. We present a structure-aware densification framework. Our key insight is that the decision to subdivide a Gaussian should be driven by an explicit comparison between its projected screen-space extent and the local structure of the texture it seeks to represent. We introduce a multi-scale frequency analysis combining structure tensors with Laplacian scale space analysis to estimate the dominant frequency at each pixel, enabling robust supervision across varying texture scales. Based on this analysis, we define $η$, a per-Gaussian, per-axis frequency violation metric that indicates when a primitive may be under-resolving local texture details. Unlike methods that perform isotropic splitting (e.g., splitting each Gaussian into two smaller ones with uniform shape), our approach performs anisotropic splitting. For each axis with high $η$, we compute a split factor to better resolve the local frequency content. We further introduce a multiview consistency criterion that aggregates $η$ observations across multiple views. By performing densification early and faster, we skip the lengthy iterative densification phases required by baseline methods and achieve significantly faster convergence. Experiments on standard benchmarks demonstrate that our method also achieves superior reconstruction quality, particularly in high-frequency regions.

Decentralized Finance (DeFi) is a rapidly evolving segment of blockchain technology that enables a transformative approach to financial services through Web3 applications. By leveraging smart contracts, DeFi allows developers to build flexible and innovative financial instruments. Among the most prominent DeFi primitives by liquidity are decentralized exchange~(DEX) swap protocols~(such as Uniswap, Curve, and Balancer) that facilitate fast token-to-token exchanges. However, new exchange mechanisms also introduce new market inefficiencies that can be systematically exploited by arbitrageurs. This paper focuses on swap protocols based on the Automated Market Maker~(AMM), where the product of reserves is preserved as an invariant. We analyze the interaction between arbitrageurs and AMM liquidity pools and develop a mathematical model grounded in empirical pool configurations. Using this model, we derive bounds on the joint revenue of liquidity providers~(LPs) and arbitrageurs, propose a method to estimate the expected number of blocks until the occurrence of Impermanent Loss~(IL), and obtain a lower bound on the pool fee required to achieve a fixed target probability of staying in the Impermanent Gain (IG) zone within a block. The proposed framework extends existing LP risk-assessment methodologies by quantifying symbiotic profitability zones, providing a principled basis for fee selection that aligns LP-arbitrageur incentives and enhances market stability.

We reframe clinician overrides of clinical AI recommendations as implicit preference data - the same signal structure exploited by reinforcement learning from human feedback (RLHF), but richer: the annotator is a domain expert, the alternatives carry real consequences, and downstream outcomes are observable. We present a formal framework extending standard preference learning with three contributions: a five-category override taxonomy mapping override types to distinct model update targets; a preference formulation conditioned on patient state s, organizational context c, and clinician capability kappa, where kappa decomposes into execution capability kappa-exec and alignment capability kappa-align; and a dual learning architecture that jointly trains a reward model and a capability model via alternating optimization, preventing a failure mode we term suppression bias-the systematic suppression of correct-but-difficult recommendations when clinician capability falls below the execution threshold. We argue that chronic disease management under outcome-based payment contracts produces override data with uniquely favorable properties-longitudinal density, concentrated decision space, outcome labels, and natural capability variation-and that training environments combining longitudinal outcome measurement with aligned financial incentives are a necessary condition for learning a reward model aligned with patient trajectory rather than with encounter economics. This framework emerged from operational work to improve clinician capability in a live value-based care deployment.

Recent advances in large language models (LLMs) have increasingly relied on reinforcement learning (RL) to improve their reasoning capabilities. Three approaches have been widely adopted: (i) Proximal policy optimization and advantage actor-critic rely on a deep neural network to estimate the value function of the learning policy in order to reduce the variance of the policy gradient. However, estimating and maintaining such a value network incurs substantial computational and memory overhead. (ii) Group relative policy optimization (GRPO) avoids training a value network by approximating the value function using sample averages. However, GRPO samples a large number of reasoning traces per prompt to achieve accurate value function approximation, making it computationally expensive. (iii) REINFORCE-type algorithms sample only a single reasoning trajectory per prompt, which reduces computational cost but suffers from poor sample efficiency. In this work, we focus on a practical, resource-constrained setting in which only a small number of reasoning traces can be sampled per prompt, while low-variance gradient estimation remains essential for high-quality policy learning. To address this challenge, we bring classical nonparametric statistical methods, which are both computationally and statistically efficient, to LLM reasoning. We employ kernel smoothing as a concrete example for value function estimation and the subsequent policy optimization. Numerical and theoretical results demonstrate that our proposal achieves accurate value and gradient estimation, leading to improved policy optimization.

Integrating multimodal foundation models into enterprise ecosystems presents a fundamental software architecture challenge. Architects must balance competing quality attributes: the high latency and non-determinism of vision language action (VLA) models versus the strict determinism and real-time performance required by enterprise control loops. In this study, we propose an architectural pattern language for visual agents that separates fast, deterministic reflexes from slow, probabilistic supervision. It consists of four architectural design patterns: (1) Hybrid Affordance Integration, (2) Adaptive Visual Anchoring, (3) Visual Hierarchy Synthesis, and (4) Semantic Scene Graph.

Latent reasoning offers a more efficient alternative to explicit reasoning by compressing intermediate reasoning into continuous representations and substantially shortening reasoning chains. However, existing latent reasoning methods mainly focus on supervised learning, and reinforcement learning in latent space remains highly unstable. We study this problem through the lens of Group Relative Policy Optimization (GRPO), and show that directly adapting GRPO to latent reasoning is fundamentally non-trivial: latent reasoning changes both the probability density and the sampling mechanism, causing three coupled bottlenecks: absence of intrinsic latent manifolds, where unconstrained exploration pushes rollouts off the valid latent manifold; exploration-optimization misalignment, where trajectory-level rewards can induce incorrect token-level updates; and latent mixture non-closure, where jointly reinforcing multiple correct latent paths can produce an invalid averaged state. To address them, we propose \textbf{Latent-GRPO}, which combines invalid-sample advantage masking, one-sided noise sampling, and optimal correct-path first-token selection. Across four low-difficulty benchmarks (e.g., GSM8K-Aug) and four high-difficulty benchmarks (e.g., AIME), Latent-GRPO improves over its latent initialization by 7.86 Pass@1 points on low-difficulty tasks and surpasses explicit GRPO by 4.27 points on high-difficulty tasks while using 3--4$\times$ shorter reasoning chains. It also achieves stronger pass@$k$ performance under Gumbel sampling. These results establish Latent-GRPO as an effective approach for stable and efficient latent reasoning.

This paper examines how different types of large language model (LLM) agents perform on scientific visualization (SciVis) tasks, where users generate visualization workflows from natural-language instructions. We compare three primary interaction paradigms, including domain-specific agents with structured tool use, computer-use agents, and general-purpose coding agents, by evaluating eight representative agents across 15 benchmark tasks and measuring visualization quality, efficiency, robustness, and computational cost. We further analyze interaction modalities, including code scripts and model context protocol (MCP) or API calls for structured tool use, as well as command-line interfaces (CLI) and graphical user interfaces (GUI) for more general interaction, while additionally studying the effect of persistent memory in selected agents. The results reveal clear tradeoffs across paradigms and modalities. General-purpose coding agents achieve the highest task success rates but are computationally expensive, while domain-specific agents are more efficient and stable but less flexible. Computer-use agents perform well on individual steps but struggle with longer multi-step workflows, indicating that long-horizon planning is their primary limitation. Across both CLI- and GUI-based settings, persistent memory improves performance over repeated trials, although its benefits depend on the underlying interaction mode and the quality of feedback. These findings suggest that no single approach is sufficient, and future SciVis systems should combine structured tool use, interactive capabilities, and adaptive memory mechanisms to balance performance, robustness, and flexibility.

Fine-tuning pretrained models has become a standard approach to adapting pretrained knowledge to improve the accuracy on new sparse, imbalance datasets. However, issues arise when optimization falls into a collapsed state, where the model gets stuck, leading to degraded performance and unstable training. One possible reason for this is the cancellation of gradients across training examples. To address this problem, we propose a novel algorithm, dynamic scaled gradient descent (\mName), that directly modifies the gradients returned by training examples, specifically, scaling down the gradients of correctly classified examples using a dynamic scaler. This strategy offers both theoretical and empirical advantages in improving training stability. Experiments on a variety of benchmark datasets, spanning multiple tasks and large pretrained models, demonstrate that our method consistently reduces performance variance and surpasses the accuracy of existing approaches.

In recent years, novel AI accelerators have emerged as promising alternatives to GPU for AI model training and inference tasks. One such accelerator, the Cerebras CS-3, achieves strong performance on large model training as well as scientific applications like molecular dynamics simulations. While dense compute workloads have been thoroughly explored for the CS-3, its potential for sparse workloads has not been fully examined. Applications requiring sparse linear algebra kernels, such as GNNs, linear solvers, and recommendation systems, could achieve good performance on a dataflow accelerator like the CS-3. In this work, we explore two key sparse linear algebra kernels, sparse-dense matrix multiplication (SpMM) and sampled dense-dense matrix multiplication (SDDMM), on the Cerebras CS-3. We propose low-level CS-3 kernel designs for these operations and optimize our designs to improve I/O performance, memory footprint, and scalability to large matrices. Our evaluation examines memory footprint and SpMM/SDDMM speedup relative to CPU. The evaluation suggests that the CS-3 can outperform CPU by 100$\times$ for SpMM with 90\% sparse matrices with performance improving as sparse matrix dimensionality increases. SDDMM on CS-3 can outperform CPU 20$\times$ for 90\% sparse matrices. We additionally find that as sparsity increases to beyond 99\%, the CS-3 suffers from performance degradation that makes it slower than CPU for SpMM.

In this paper, we consider the Santa Claus problem in the CONGEST model. This NP-hard problem can be modeled as a bipartite graph of children and gifts where an edge indicates that a child desires a gift. Notably, each gift can have a different value. The goal is to assign the gifts to the children such that the least happy child is as happy as possible. Even though this is a well-studied problem in the sequential setting, we obtain the first results the distributed setting. In particular, we show that the complexity of computing an $\mathcal{O}(\log n/\log \log n)$-approximation is $\hat Θ(\sqrt n+D)$ rounds, where our $\widetildeΩ(\sqrt n+D)$-round lower bound is even stronger and holds for any approximation.

Multivariate time series forecasting plays a pivotal role in numerous real-world applications, including financial analysis, energy management, and traffic planning. While Transformer-based architectures have gained popularity for this task, recent studies reveal that simpler MLP-based models can achieve competitive or superior performance with significantly reduced computational cost. In this paper, we propose ITS-Mina, a novel all-MLP framework for multivariate time series forecasting that integrates three key innovations: (1) an iterative refinement mechanism that progressively enhances temporal representations by repeatedly applying a shared-parameter residual mixer stack, effectively deepening the model's computational capacity without multiplying the number of distinct parameters; (2) an external attention module that replaces traditional self-attention with learnable memory units, capturing cross-sample global dependencies at linear computational complexity; and (3) a Harris Hawks Optimization (HHO) algorithm for automatic dropout rate tuning, enabling adaptive regularization tailored to each dataset. Extensive experiments on six widely-used benchmark datasets demonstrate that ITS-Mina achieves state-of-the-art or highly competitive performance compared to eleven baseline models across multiple forecasting horizons.

Maximal Extractable Value (MEV) represents billions of dollars in extracted value that fundamentally shapes blockchain network dynamics and participant incentives. While research has focused on MEV extraction and mitigation, we lack systematic methods to attribute MEV opportunities to their on-chain origins. This paper formalizes the MEV opportunity attribution problem and introduces a systems framework for identifying which transactions create arbitrage opportunities and quantifying their contributions. We design and evaluate four attribution methods for atomic arbitrage on EVM-compatible networks: bot-data-driven, simulation-based, coefficient-based, and Shapley-based approaches. Through large-scale retrospective analysis spanning over one million blocks on Polygon, we demonstrate that the majority of atomic arbitrage opportunities can be traced to single source transactions, validating our central hypothesis about competitive MEV markets. We quantify a highly concentrated distribution of MEV creation, where a small subset of protocols generates most opportunities, and provide comparative analysis of method trade-offs in accuracy, cost, and scalability. Our findings offer insights for protocol designers reducing MEV leakage, validators optimizing transaction ordering, and analysts measuring ecosystem health through opportunity creation.

Despite recent progress in language models and agents for scientific data-driven discovery, further advancing their capabilities is held back by the absence of verifiable environments representing real-world scientific tasks.To fill this gap, we introduce D3-Gym, the first automatically constructed dataset with verifiable environments for scientific Data-Driven Discovery. D3-Gym comprises (1) 565 tasks sourced from 239 real scientific repositories across four disciplines where (2) each task is equipped with a natural language instruction, an executable environment with pre-installed dependencies, input dataset and artifact previews, a reference code solution, and an automatically synthesized evaluation script. Rigorous evaluation of the quality of the verification signal in D3-Gym confirms that our evaluation scripts achieve 87.5% agreement with human-annotated gold standards and strong alignment in domain-specific evaluation logic, showing their scientific soundness. Further, training on trajectories sampled from D3-Gym yields consistent and substantial gains across Qwen3 models of varying sizes on ScienceAgentBench, boosting Qwen3-32B by 7.8 absolute points and substantially shrinking the gap with strong proprietary models. All D3-Gym artifacts (environments, creation workflow, trajectories, and models) can be found at https://github.com/OSU-NLP-Group/D3-Gym.

Traditional Shot Boundary Detection (SBD) inherently struggles with complex transitions by formulating the task around isolated cut points, frequently yielding corrupted video shots. We address this fundamental limitation by formalizing the Shot Transition Detection (STD) task. Rather than searching for ambiguous points, STD explicitly detects the continuous temporal segments of transitions. To tackle this, we propose TransVLM, a Vision-Language Model (VLM) framework for STD. Unlike regular VLMs that predominantly rely on spatial semantics and struggle with fine-grained inter-shot dynamics, our method explicitly injects optical flow as a critical motion prior at the input stage. Through a simple yet effective feature-fusion strategy, TransVLM directly processes concatenated color and motion representations, significantly enhancing its temporal awareness without incurring any additional visual token overhead on the language backbone. To overcome the severe class imbalance in public data, we design a scalable data engine to synthesize diverse transition videos for robust training, alongside a comprehensive benchmark for STD. Extensive experiments demonstrate that TransVLM achieves superior overall performance, outperforming traditional heuristic methods, specialized spatiotemporal networks, and top-tier VLMs. This work has been deployed to production. For more related research, please visit HeyGen Research (https://www.heygen.com/research) and HeyGen Avatar-V (https://www.heygen.com/research/avatar-v-model). Project page: https://chence17.github.io/TransVLM/

Since the dawn of Trading Card Games, the genre has grown into a multi-billion-dollar industry engaging millions of analog and digital players worldwide. Popular TCGs rely on regular updates, balance adjustments, and rotating constraints to sustain engagement. Yet, as metagames stabilize, predictable strategies dominate and viable card options diminish, often resulting in repetitive and impaired player experiences. This paper investigates the use of Large Language Models and Image Diffusion Models for Procedural Content Generation of TCG cards, addressing these challenges by enabling a personalized infinity of card designs. Modern generative AI not only enables large-scale content creation but could even introduce procedural relatedness, fostering unique connections between players and their cards. We present a pipeline combining player-centric co-creation, fine-tuned embeddings, local LLMs, and Diffusion Models to generate dynamic, personalized cards while potentially expanding creative range. We evaluated the pipeline in a user study with 49 participants who generated 196 Pokémon card samples. Participants rated aesthetics and representativeness of visuals and mechanics, and provided qualitative feedback. Results show high satisfaction and indicate that most participants successfully realized their own ideas through prompt adjustments. These findings lay groundwork for future content generation systems and alternatives to conventional metagame evolution through procedural relatedness.

Multimodal large language models (MLLMs) are increasingly used to translate visual artifacts into code, from UI mockups into HTML to scientific plots into Python scripts. A circuit diagram can be viewed as a visual domain-specific language for hardware: it encodes timing, topology, and bit level semantics that are invisible to casual inspection yet safety critical once fabricated in silicon. Translating such diagrams into register-transfer-level(RTL) code therefore represents an extreme reliability test for vision-to-code generation. We reveal a phenomenon we call Mirage: replacing a circuit diagram with a blank image leaves Pass@k unchanged or even higher, because models bypass the visual input and instead exploit identifier semantics in the module header to retrieve canonical RTL templates. This constitutes a new, highly covert class of defect in AI-assisted code generation that directly undermines MLLMs' trustworthiness. To quantify the effect, we construct C2VEVAL and evaluate eight MLLMs under a paired Normal/Anony protocol in which Anony mode anonymizes all identifiers in both the diagram and the module header; Anony-mode scores drop sharply across all models, confirming that high Normal-mode accuracy is largely a Mirage. We then propose VeriGround (4B), trained with identifier anonymization, refusal augmentation, and D-ORPO (Decision-Focused ORPO) preference alignment that up-weights pivotal generate-or-refuse tokens. VeriGround achieves Functional Pass@1 of 46.11%/42.51%(Normal/Anony) with a False Refusal Rate of only 1.20%/0.00%, while maintaining >92% Refusal Rate on blank images. With only 4B parameters, VeriGround performs on par with GPT-5.4 under Normal and significantly outperforms all baselines under Anony, confirming genuine visual grounding.

Background: Timely, uncertainty-aware forecasting from irregular electronic health records (EHR) can support critical-care decisions, yet most approaches either impute to a grid or sacrifice interpretability. We introduce StructGP, a continuous-time multi-task Gaussian process that couples process convolutions with differentiable structure learning to uncover a sparse, ordered directed acyclic graph (DAG) of inter-variable dependencies while preserving principled uncertainty. We further propose LP-StructGP, which augments StructGP with latent pathways-shared, temporally shifted trajectories inferred via subject-specific coupling filters and a softmax gating mechanism-to capture cross-patient progression patterns. Both models are trained under sparsity and acyclicity constraints (augmented Lagrangian, Adam) using scalable low-rank updates. Results: In simulations, the approach reliably recovers ground-truth graphs (Structural Hamming Distance approaching 0 as cohorts grow) and pathway assignments (high Adjusted Rand Index). On a MIMIC-IV septic shock cohort (n=1,008; norepinephrine, creatinine, mean arterial pressure), StructGP improves short-horizon (6 h) forecasting over independent-task baselines (average RMSE 0.68 [95%CI: 0.63--0.74] vs. 0.88 [0.83-0.94]) and, with 15 additional inputs, markedly outperforms unstructured kernels (0.63 [0.58-0.69] vs. 3.02 [2.85-3.18]) with superior calibration (coverage 0.96 vs. 0.84). On the PhysioNet Challenge (12k patients, 41 variables), StructGP attains competitive accuracy (MAE 3.72e-2) relative to a state-of-the-art graph neural model while maintaining calibrated uncertainty. Conclusion: These results show that structured process convolutions with latent pathways deliver interpretable, scalable, and well-calibrated forecasting for irregular clinical time series.

Assumption-Based Argumentation (ABA) is a well-established formalism for modelling and reasoning over debates, with a wide range of applications. However, the high computational complexity of core reasoning tasks in ABA poses a significant challenge for its applicability. This issue is further aggravated when ABA frameworks (ABAFs) are instantiated into graph-based argumentation formalisms, such as Dung's Argumentation Frameworks (AFs) and Argumentation Frameworks with Collective Attacks (SETAFs). In knowledge representation and reasoning, a key strategy to address computational intractability is to optimise reasoning over a given knowledge base through divide-and-conquer algorithms. A paradigmatic example of this approach is splitting, where extensions of a given framework are computed incrementally, by restricting the search space to sub-frameworks only, and then combining the obtained results. This approach has been successfully applied to AFs, for which also a parametrised version has been introduced under stable semantics. However, the exponential growth produced by the instantiation might undermine the usefulness of splitting on the argument graphs induced by ABAFs. To address this issue, our work investigates the concept of splitting on the knowledge base rather than on its graph-based instantiation. Furthermore, we generalise splitting to its parametrised version for ABAFs.

Designing mechanical linkages involves combinatorial topology selection and continuous parameter fitting. We show that language models can systematically improve linkage designs through symbolic representations. Language model agents explore discrete topologies while numerical optimisers fit continuous parameters. A symbolic lifting operator translates simulator trajectories into qualitative descriptors, motion labels, temporal predicates, and structural diagnostics that models interpret across iterative design cycles. Across six engineering-relevant motion targets and three open-source models (Llama 3.3 70B, Qwen3 4B, Qwen3 MoE 30B-A3B), the modular architecture reduces geometric error by up to 68% and improves structural validity by up to 134% over monolithic baselines. Critically, 78.6% of iterative refinement trajectories show measurable improvement, with the system correctly diagnosing overconstraint (56.3%) and underconstraint (35.6%) failure modes and proposing grounded corrections. Models across all three families acquire interpretable mechanical reasoning strategies without fine-tuning, demonstrating that principled symbolic abstraction bridges generative AI and the numerical precision required for engineering design.

Recent large language models (LLMs) have achieved impressive reasoning milestones but continue to struggle with high computational costs, logical inconsistencies, and sharp performance degradation on high-complexity problems. While neuro-symbolic methods attempt to mitigate these issues by coupling LLMs with symbolic reasoners, existing approaches typically rely on monotonic logics (e.g., SMT) that cannot represent defeasible reasoning -- essential components of human cognition. We present "LLM+ASP," a framework that translates natural language into Answer Set Programming (ASP), a nonmonotonic formalism based on stable model semantics. Unlike prior "LLM+ASP" approaches that require manually authored knowledge modules, domain-specific prompts, or evaluation restricted to single problem classes, our framework operates without any per-task engineering and applies uniformly across diverse reasoning tasks. Our system utilizes an automated self-correction loop where structured feedback from the ASP solver enables iterative refinement. Evaluating across six diverse benchmarks, we demonstrate that: (1) stable model semantics allow LLMs to naturally express default rules and exceptions, outperforming SMT-based alternatives by significant margins on nonmonotonic tasks; (2) iterative self-correction is the primary driver of performance, effectively replacing the need for handcrafted domain knowledge; (3) compact in-context reference guides substantially outperform verbose documentation, revealing a "context rot" phenomenon where excessive context hinders constraint adherence.

Graphical User Interface (GUI) agents have emerged as a promising paradigm for intelligent systems that perceive and interact with graphical interfaces visually. Yet supervised fine-tuning alone cannot handle long-horizon credit assignment, distribution shifts, and safe exploration in irreversible environments, making Reinforcement Learning (RL) a central methodology for advancing automation. In this work, we present the first comprehensive overview of the intersection between RL and GUI agents, and examine how this research direction may evolve toward digital inhabitants. We propose a principled taxonomy that organizes existing methods into Offline RL, Online RL, and Hybrid Strategies, and complement it with analyses of reward engineering, data efficiency, and key technical innovations. Our analysis reveals several emerging trends: the tension between reliability and scalability is motivating the adoption of composite, multi-tier reward architectures; GUI I/O latency bottlenecks are accelerating the shift toward world-model-based training, which can yield substantial performance gains; and the spontaneous emergence of System-2-style deliberation suggests that explicit reasoning supervision may not be necessary when sufficiently rich reward signals are available. We distill these findings into a roadmap covering process rewards, continual RL, cognitive architectures, and safe deployment, aiming to guide the next generation of robust GUI automation and its agent-native infrastructure.

As Vision-Language Models (VLMs) become increasingly integrated into decision-making systems, it is essential to understand how visual inputs influence their behavior. This paper investigates the effects of visual priming on VLMs' cooperative behavior using the Iterated Prisoner's Dilemma (IPD) as a test scenario. We examine whether exposure to images depicting behavioral concepts (kindness/helpfulness vs. aggressiveness/selfishness) and color-coded reward matrices alters VLM decision patterns. Experiments were conducted across multiple state-of-the-art VLMs. We further explore mitigation strategies including prompt modifications, Chain of Thought (CoT) reasoning, and visual token reduction. Results show that VLM behavior can be influenced by both image content and color cues, with varying susceptibility and mitigation effectiveness across models. These findings not only underscore the importance of robust evaluation frameworks for VLM deployment in visually rich and safety-critical environments, but also highlight how architectural and training differences among models may lead to distinct behavioral responses-an area worthy of further investigation.

Joint image compression and wireless transmission remain relatively underexplored compared to generic image restoration, despite its importance in practical communication systems. We formulate this problem under an equivalent linear model, and propose Diffusion-OAMP, a training-free reconstruction framework that embeds a pre-trained diffusion model into the OAMP algorithm. In Diffusion-OAMP, the OAMP linear estimator produces pseudo-AWGN observations, while the diffusion model serves as a nonlinear estimator under an SNR-matching rule. This framework offers a way to incorporate multiple generative priors into OAMP. Experiments with varying compression ratios and noise levels show that Diffusion-OAMP performs favorably against classic methods in the evaluated settings.

In this paper an attractor FCM is created, tested, and analyzed. This FCM is neither a hebbian based nor agentic, nor a hybrid; it rather is a gradient descent based, physics constrained, Jacobian version of an FCM. Moreover, this model has several quirks; it uses residual memory, back propagation through time, and a fixed point anchor that is recursively implemented to update its weights. The residuals update the recursive part without losing the system memory. The model's anchor enables it to converge in a fixed point for which back propagation through time unrolls it and ensures that the error minimization is for an accurate gradient. Furthermore, a new learning algorithm is utilized. The Newton's method finds the system's fixed point attractor and then gradient descend is adaptively changing the landscape; an adaptive term is used to directly manipulate the weights through the attractor dynamics. As the adaptive term changes, the descent through the landscape is constantly adjusting according to sigmoid saturation, and that prevents premature convergence to a local minimum. Lastly, the updates are filtered by causal mask that informs the network about the physics, respecting the initial expert based opinions, for which model reduces the error to the target in an efficient way.

Large-scale IoT weather sensing networks require incentive mechanisms to sustain participation, yet determining how much value individual data contributions bring to the network remains an open problem. Existing approaches address data quality but not data valuation; in operational meteorology, adjoint-based methods derive value from the forecast model itself but require full data assimilation infrastructure. We propose to utilise differentiable AI weather models to fill this gap and characterise gradient-based attribution on gridded GFS analysis inputs as a candidate value signal, evaluating fidelity, calibration, cost, and gaming vulnerability across more than 400 configurations. Attribution captures near-optimal sensor placement utility with monotonically faithful payments, but can be inflated by adversarial inputs, with detection requiring external baseline data. These findings establish gradient attribution as a computationally validated signal for model-informed reward allocation in participatory weather sensing.

Collective intelligence emerges across biological, physical, and artificial systems without central coordination, yet a unifying principle governing such behaviour remains elusive. The Free Energy Principle explains how individual agents adapt through variational inference, while game theory formalises strategic interactions. Here we introduce the Game-Theoretic Free Energy Principle, a unified framework showing that multi-agent systems performing local free-energy minimisation implicitly implement a stochastic game. We prove that, under bounded rationality and local information constraints, stationary points of collective free energy correspond to approximate Nash equilibria of an induced game. Conversely, a broad class of cooperative games admits a variational representation in which equilibria arise as Gibbs distributions over coalitions, establishing a bridge between Bayesian inference and strategic interaction. To characterise higher-order effects, we introduce a free-energy formulation of the Harsanyi dividend, isolating irreducible multi-agent synergy. This yields a predictive theory of cooperation, including a falsifiable non-monotonic relationship between sensory precision and agent influence. We validate this prediction across neural, biological, and artificial multi-agent systems. These results identify a common variational principle underlying inference, thermodynamics, and game-theoretic equilibrium.

Animals hear and vocalize across frequency ranges that differ substantially from humans, often extending into the ultrasonic domain. Yet most computational bioacoustics systems rely on audio models pre-trained at 16 kHz, restricting their usable bandwidth to the 0-8 kHz baseband and discarding higher-frequency information present in many bioacoustic recordings. We investigate a multi-band encoding framework that decomposes the full spectrum of animal calls into band features and fuses them into a unified representation. Similarity analyses on models show that certain encoders produce decorrelated band embeddings that improve class separation after fusion. Classification experiments on three bioacoustic datasets using eight pre-trained models and five fusion strategies show that fused representations consistently outperform the baseband and time-expansion baselines on two datasets, showing the potential of multi-band methods for full-spectrum encoding of animal calls.

Multimodal Stance Detection (MSD) is crucial for understanding public discourse, yet effectively fusing text and image, especially with conflicting signals, remains challenging. Existing methods often face difficulties with contextual grounding, cross-modal interpretation ambiguity, and single-pass reasoning fragility. To address these, we propose Retrieval-Augmented Multi-modal Multi-agent Stance Detection (MM-StanceDet), a novel multi-agent framework integrating Retrieval Augmentation for contextual grounding, specialized Multimodal Analysis agents for nuanced interpretation, a Reasoning-Enhanced Debate stage for exploring perspectives, and Self-Reflection for robust adjudication. Extensive experiments on five datasets demonstrate MM-StanceDet significantly outperforms state-of-the-art baselines, validating the efficacy of its multi-agent architecture and structured reasoning stages in addressing complex multimodal stance challenges.

Concurrent advances across fields such as organoid technology, Microelectrode Arrays (MEAs), neuromorphic computing, and machine learning have given rise to a groundbreaking research paradigm: Synthetic Biological Intelligence (SBI). SBI refers to engineered systems in which living Biological Neural Networks (BNNs) are interfaced with hardware and software to perform task-oriented information processing in a closed loop. This cutting-edge technology, while still in its infancy, has the potential to deliver highly efficient performance across both computing capabilities and energy consumption. The early stage of this field underscores the need for reliable multi-scale and cross-domain interaction interfaces to support applications in robotics, biomedicine, signal processing, and neuroscience research. The hitherto lack of commercially available SBI platforms has slowed the development, as the conditions to produce a testbed are expensive and cumbersome. The introduction of standardized, platform- and cloud-integrated BNNs has been a crucial catalyst for the scientific community, improving the accessibility of SBI and leading the way to further developments. In this survey, we summarize the innovations that contributed to the emergence of SBI and the first testbed interfaces that enabled its embodiment. This work reframes SBI as a bio-digital interaction system and introduces a unified protocol across encoding, decoding, system engineering, and benchmarking.

With the widespread adoption of large language models (LLMs), understanding their personality representation mechanisms has become critical. As a novel paradigm in Personality Editing, most existing methods employ neuron-editing to locate and modify LLM neurons, requiring changes to numerous neurons and leading to significant performance degradation. This raises a fundamental question: Are all modified neurons directly related to personality representation? In this work, we investigate and quantify this specificity through assessments of general capability impact and representation-level patterns. We find that: 1) Current methods can change personalities but reduce overall performance. 2) Neurons are multifunctional, connecting personality traits and general knowledge. 3) Opposing personality traits demonstrate distinctly mutually exclusive representation patterns. Motivated by these findings, we propose DPN-LE (Dual Personality Neuron Localization and Editing), which identifies personality-specific neurons by contrasting MLP activations between high-trait and low-trait samples. DPN-LE constructs layer-wise steering vectors and applies dual-criterion filtering based on Cohen's $d$ effect size and activation magnitude to isolate mutually exclusive neuron subsets. Sparse linear intervention on these neurons enables precise personality control at inference time. Using only 1,000 contrastive sample pairs per trait, DPN-LE intervenes on $\sim$0.5\% of neurons while achieving competitive personality control and substantially better capability preservation across reasoning tasks. Experiments on LLaMA-3-8B-Instruct and Qwen2.5-7B-Instruct demonstrate the effectiveness and generalizability of our approach.

Tunnel inspection requires outputs that can support defect localization, measurement, severity grading, and engineering documentation. Existing training-free foundation-model pipelines usually stop at coarse open-vocabulary proposals, which are difficult to use directly in interference-heavy tunnel scenes. We propose a training-free framework TunnelMIND. Specifically, language-guided defect proposals are not treated as final outputs; instead, their spatial support is recalibrated at inference time through dense visual consistency, so that coarse semantic anchors can be transformed into more reliable prompts under tunnel-specific hard negatives. The resulting masks are further reconstructed into structured defect entities with category, location, geometry, severity, and context attributes, which are then mapped to retrieval-grounded explanation and engineering-readable report generation under expert knowledge constraints. On visible, GPR, and road defect tasks, TunnelMIND achieves F1 scores of 0.68, 0.78, and 0.72, respectively. Overall, TunnelMIND shows that training-free tunnel inspection can move beyond coarse localization toward structured defect evidence for engineering assessment.

We introduce a framework called LAPITHS (Language model Analysis through Paradigm grounded Interpretations of Theses about Human likenesS) and use it to show that several major claims advanced by models such as CENTAUR, proposed as an artificial Unified Model of Cognition, are not theoretically or empirically justified. LAPITHS provides a principled reference point for counteracting the current behaviouristic tendency in AI research to interpret the human level performances of transformer based language models as evidence of human like underlying computation and, by extension, as signs of cognitive abilities. The novelty of LAPITHS lies in making explicit the arguments grounded in two quantitative assessments: (i) the Minimal Cognitive Grid, a theoretically motivated method for estimating the cognitive plausibility of artificial systems, and (ii) a behavioural comparison showing that results similar to those reported for CENTAUR like models can be reproduced by other systems that do not satisfy the structural constraints typically associated with cognitive plausibility, and whose outputs do not provide independent explanatory insight into human cognition.

Peer review is a multi-stage process involving reviews, rebuttals, meta-reviews, final decisions, and subsequent manuscript revisions. Recent advances in large language models (LLMs) have motivated methods that assist or automate different stages of this pipeline. In this survey, we synthesize techniques for (i) peer review generation, including fine-tuning strategies, agent-based systems, RL-based methods, and emerging paradigms to enhance generation; (ii) after-review tasks including rebuttals, meta-review and revision aligned to reviews; and (iii) evaluation methods spanning human-centered, reference-based, LLM-based and aspect-oriented. We catalog datasets, compare modeling choices, and discuss limitations, ethical concerns, and future directions. The survey aims to provide practical guidance for building, evaluating, and integrating LLM systems across the full peer review workflow.

Preserving affective nuance remains a challenge in Machine Translation (MT), where semantic equivalence often takes precedence over emotional fidelity. This paper evaluates the performance of three state-of-the-art Small Language Models (SLMs) -- EuroLLM, Aya Expanse, and Gemma -- in maintaining fine-grained emotions during backtranslation. Using the GoEmotions dataset, which comprises Reddit comments across 28 distinct categories, we assess emotional preservation across five European languages: German, French, Spanish, Italian, and Polish. Specifically, we investigate (i) the inherent capability of these SLMs to retain emotional sentiment, (ii) the efficacy of emotion-aware prompting in improving preservation, and (iii) the performance of ModernBERT as a contemporary alternative to BERT for emotion classification in MT evaluation.

Modern large multicore systems often run multiple workloads that share CPUs under schedulers such as Linux CFS. To keep CPUs busy, these schedulers load-balance runnable work, causing each workload to execute on many cores. This weakens locality at the microarchitectural level: workloads lose reuse in caches, branch predictors, and prefetchers, and interfere more with one another - especially on chiplet-based systems, where spreading execution across cores also spreads it across LLC boundaries. A natural alternative is strict CPU partitioning, but hard partitions leave capacity idle when workloads do not fully use their reserved CPUs. We present Affinity Tailor, a userspace-guided kernel scheduling system built on a key insight: the kernel can preserve locality for workloads that share CPUs by treating demand-sized, topologically compact CPU sets as affinity hints rather than hard partitions. A userspace controller estimates each workload's CPU demand online and assigns a preferred CPU set sized to that demand, chosen to be as disjoint as possible from other workloads while spanning as few LLC domains as possible. The kernel then uses this set as an affinity hint, steering threads toward those CPUs while still allowing execution elsewhere when needed to preserve utilization. Deployed at Google, Affinity Tailor delivers geometric-mean per-CPU throughput gains of 12% on chiplet-based systems and 3% on non-chiplet systems over Linux CFS. Furthermore, faster execution reduces memory residency, yielding per-GB throughput gains of 3-7%. Our findings suggest that future schedulers should treat spatial locality as a first-class objective, even at the expense of work-conservation.

When language models lack relevant knowledge for a given query, they frequently generate plausible responses that can be hallucinations, rather than admitting being agnostic about the answer. Retraining models to reward admitting ignorance can lead to overly conservative behaviors and poor generalization due to scarce evaluation benchmarks. We propose a post hoc framework, Conformal Abstention (CA), adapted from conformal prediction (CP) to determine whether to abstain from answering a query. CA provides finite-sample guarantees on both the probability of participation (i.e., not abstaining) and the probability that the generated response is correct. Importantly, the abstention decision relies on prediction confidence rather than the non-conformity scores used in CP, which are intractable for open-ended generation. To better align prediction confidence with the model's ignorance, we introduce a calibration strategy using representation geometry within the model to measure knowledge involvement in shaping the response. Experiments demonstrate that we improve selective answering significantly with 75 percent conditional correctness.

Foundation models are deep neural networks (such as GPT-5, Gemini~3, and Opus~4) trained on large datasets that can perform diverse downstream tasks -- text and code generation, question answering, summarization, image classification, and so on. The philosophy of foundation models is to put effort into a single, large (${\sim}10^{12}$-parameter) general-purpose model that can be adapted to many downstream tasks with no or minimal additional training. We argue that the rise of foundation models presents an opportunity for hardware engineers: in contrast to when different models were used for different tasks, it now makes sense to build special-purpose, fixed hardware implementations of neural networks, manufactured and released at the roughly 1-year cadence of major new foundation-model versions. Beyond conventional digital-electronic inference hardware with read-only weight memory, we advocate a more radical re-thinking: hardware in which the neural network is realized directly at the level of the physical design and operates via the hardware's natural physical dynamics -- \textit{Physical Foundation Models} (PFMs). PFMs could enable orders-of-magnitude advantages in energy efficiency, speed, and parameter density. For ${\sim}10^{12}$-parameter models, this would both reduce the high energy burden of AI in datacenters and enable AI in edge devices that today are power-constrained to far smaller models. PFMs could also enable inference hardware for models much larger than current ones: $10^{15}$- or even $10^{18}$-parameter PFMs seem plausible by some measures. We present back-of-the-envelope calculations illustrating PFM scaling using an optical example -- a 3D nanostructured glass medium -- and discuss prospects in nanoelectronics and other physical platforms. We conclude with the major research challenges that must be resolved for trillion-parameter PFMs and beyond to become reality.

Persistent AI memory is often reduced to a retrieval problem: store prior interactions as text, embed them, and ask the model to recover relevant context later. This design is useful for thematic recall, but it is mismatched to the kinds of memory that agents need in production: exact facts, current state, updates and deletions, aggregation, relations, negative queries, and explicit unknowns. These operations require memory to behave less like search and more like a system of record. This paper argues that reliable external AI memory must be schema-grounded. Schemas define what must be remembered, what may be ignored, and which values must never be inferred. We present an iterative, schema-aware write path that decomposes memory ingestion into object detection, field detection, and field-value extraction, with validation gates, local retries, and stateful prompt control. The result shifts interpretation from the read path to the write path: reads become constrained queries over verified records rather than repeated inference over retrieved prose. We evaluate this design on structured extraction and end-to-end memory benchmarks. On the extraction benchmark, the judge-in-the-loop configuration reaches 90.42% object-level accuracy and 62.67% output accuracy, above all tested frontier structured-output baselines. On our end-to-end memory benchmark, xmemory reaches 97.10% F1, compared with 80.16%-87.24% across the third-party baselines. On the application-level task, xmemory reaches 95.2% accuracy, outperforming specialised memory systems, code-generated Markdown harnesses, and customer-facing frontier-model application harnesses. The results show that, for memory workloads requiring stable facts and stateful computation, architecture matters more than retrieval scale or model strength alone.

The volume of scientific manuscripts is growing faster than the capacity to evaluate them, yet the institutions that govern peer review have remained largely unchanged. The result is a widening mismatch: reviewer scarcity, noisier assessments, and declining confidence in editorial decisions. Every scientist wants better reviews, but review quality depends on the total burden, which no single author can shift. To isolate this tension, we provide a game-theoretic thought experiment: a voluntary lottery in which authors accept a chance of random pre-review rejection, reducing reviewer burden and improving the quality of surviving evaluations. We show that a Nash equilibrium emerges in which authors voluntarily enter the lottery. Scientists who care about the literature they read, not just the papers they publish, will opt in, raising the quality of published science for all.

Understanding how human health changes over time, and why responses to interventions vary between individuals, remains a central challenge in medicine. Here we present HealthFormer, a decoder-only transformer that models the human physiological trajectory generatively, by training on data from the Human Phenotype Project, a multi-visit cohort of over 15,000 deeply phenotyped individuals. We tokenise each participant's health trajectory across 667 measurements spanning seven domains: blood biomarkers, body composition, sleep physiology, continuous glucose monitoring, gut microbiome, wearable-derived physiology, and behaviour and medication exposure. We train HealthFormer to forecast individual physiological trajectories across these domains, and from this single generative objective a range of clinically relevant tasks can be expressed as queries on the model. We show that, without task-specific training, HealthFormer transfers to four independent cohorts and improves prediction for 27 of 30 incident-disease and mortality endpoints, exceeding established clinical risk scores in every comparison. We further show that the model can simulate interventions in silico: in a held-out personalised-nutrition trial, intervention-conditioned predictions recover individual six-month biomarker changes (e.g., Pearson r = 0.78 for diastolic blood pressure). Across 41 randomised intervention-outcome comparisons drawn from published trials, our results show that the predicted direction of effect agrees in every case, and the predicted mean falls within the reported 95% confidence interval in 30 cases. We position HealthFormer as an initial health world model, from which forecasting, risk stratification, and intervention-conditioned simulation arise as queries, providing a basis for clinical digital twins.

As one of the mainstream models of artificial intelligence, world models allow agents to learn the representation of the environment for efficient prediction and planning. However, classical world models based on flat tensors face several key problems, including noise sensitivity, error accumulation and weak reasoning. To address these limitations, many recent studies use graph structure to decompose the environment into entity nodes and interactive edges, and model virtual environments in a structured space. This paper systematically formalizes and unifies these emerging graph-based works under the concept of graph world models (GWMs). To the best of our knowledge, GWMs have not yet been explicitly defined and surveyed as a unified research paradigm. Furthermore, we propose a taxonomy based on relational inductive biases (RIB), categorizing GWMs by the specific structural priors they inject: (1) spatial RIB for topological abstraction; (2) physical RIB for dynamic simulation; and (3) logical RIB for causal and semantic reasoning. For each model category, we outline the key design principles, summarize representative models, and conduct comparative analyses. We further discuss open challenges and future directions, including dynamic graph adaptation, probabilistic relational dynamics, multi-granularity inductive biases, and the need for dedicated benchmarks and evaluation metrics for GWMs.

Angular is one of the most widely adopted frameworks for developing large-scale, dynamic web applications. As projects increase in scope and complexity, developers face growing challenges in managing architecture and maintaining clean, modular code. These challenges often lead to design flaws, commonly referred to as code smells. While React-specific smells have been cataloged in prior studies, limited knowledge exists regarding Angular-specific smells and how they manifest. This study investigates Angular code smells through a grey literature review, consolidating community knowledge and technical discussions. From the collected sources, 11 distinct Angular code smells were identified, 6 of which also occur in React-based systems, suggesting that some issues are cross-framework. Each smell was analyzed, exemplified, and grouped according to its technical characteristics. Based on the resulting catalog, we implemented an automated static analysis tool to detect Angular code smells. The tool was empirically evaluated using a manually validated dataset, and its effectiveness was assessed through standard information retrieval metrics. The evaluation results indicate high detection performance across all smells, achieving accuracy values above 0.88 and F1-scores ranging from 0.89 to 1.00. The findings reveal recurring issues such as component overloading, duplicated logic, and inefficient template bindings, reinforcing the relevance of systematic detection support. This study presents the first catalog of Angular-specific code smells derived from grey literature and demonstrates the feasibility and effectiveness of automated detection, providing a solid foundation for future empirical studies and tool development aimed at improving front-end code quality.

The rapidly expanding artificial intelligence (AI) industry has produced diverse yet powerful prediction tools, each with its own network architecture, training strategy, data-processing pipeline, and domain-specific strengths. These tools create new opportunities for semi-supervised inference, in which labeled data are limited and expensive to obtain, whereas unlabeled data are abundant and widely available. Given a collection of predictors, we treat them as a mixture of experts (MOE) and introduce an MOE-powered semi-supervised inference framework built upon prediction-powered inference (PPI). Motivated by the variance reduction principle underlying PPI, the proposed framework seeks the mixture of experts that achieves the smallest possible variance. Compared with standard PPI, the MOE-powered inference framework adapts to the unknown performance of individual predictors, benefits from their collective predictive power, and enjoys a best-expert guarantee. The framework is flexible and applies to mean estimation, linear regression, quantile estimation, and general M-estimation. We develop non-asymptotic theory for the MOE-powered inference framework and establish upper bounds on the coverage error of the resulting confidence intervals. Numerical experiments demonstrate the practical effectiveness of MOE-powered inference and corroborate our theoretical findings.

Agent orchestration frameworks -- LangGraph, CrewAI, Google ADK, OpenAI Agents SDK, and others -- place an external orchestrator above the LLM, tracking state and injecting routing instructions at every turn. We present a controlled comparison showing that for procedural tasks, this architecture is dominated by a simpler alternative: putting the entire procedure in the system prompt and letting the model self-orchestrate. Across three domains -- travel booking (14 nodes), Zoom technical support (14 nodes), and insurance claims processing (55 nodes) -- we evaluate 200 conversations per condition using LLM-as-judge scoring on five quality criteria. The in-context approach scores 4.53--5.00 on a 5-point scale while a LangGraph orchestrator using the same model scores 4.17--4.84. The orchestrated system fails on 24% of travel, 9% of Zoom, and 17% of insurance conversations, compared to 11.5%, 0.5%, and 5% for the in-context baseline. While external orchestration may have been necessary for earlier models, advances in frontier model capabilities have made it unnecessary for multi-turn conversations following a defined procedure.

In modern parametric model training, full-batch gradient descent (and its variants) suffers due to progressively stronger biasing towards the exact realization of training data; this drives the systematic ``generalization gap'', where the train error becomes an unreliable proxy for test error. Existing approaches either argue this gap is benign through complex analysis or sacrifice data to a validation set. In contrast, we introduce decoupled descent (DD), a novel theory-based training algorithm that satisfies a train-test identity -- enforcing the train error to asymptotically track the test error for stylized Gaussian mixture models. Within this specific regime, leveraging approximate message passing theory, DD iteratively cancels the biases due to data reuse, rigorously demonstrating the feasibility of zero-cost validation and $100\%$ data utilization. Moreover, DD is governed by a low-dimensional state evolution recursion, rendering the dynamics of the algorithm transparent and tractable. We validate DD on XOR classification, yielding superior performance compared to GD; additionally, we implement noisy MNIST and non-linear probing of CIFAR-10, demonstrating that even when our stylized assumptions are relaxed, DD narrows the generalization gap compared to GD.

Recent advances in agentic AI are shifting automation from discrete tools to proactive multi-agent systems that coordinate multi-specialized capabilities behind unified interfaces. However, today's agent systems typically rely on hard-coded agent architectures with fixed roles, coordination patterns, and interaction flows that limit end-user personalization and make adaptation to individual needs and contexts difficult. Given this limitation, we argue that on-demand persona-based agent generation offers a promising path towards more efficient and contextually appropriate interaction within agentic workflows. By dynamically crafting agents and personas at run-time to match user characteristics, task demands, and workflow context, agentic platforms can move beyond one-size-fits-all configurations. We present a pipeline for on-demand persona generation in agentic platforms, detailing how real-time crafting of AI personas can be systematically integrated within agent systems, aiming to open new possibilities in agentic platform design paradigms.

Large language model (LLM) agents deployed in clinical settings often exhibit abrupt, threshold-driven behavior, offering little visibility into accumulating risk prior to escalation. In real-world care, however, clinicians act on gradually rising concern rather than instantaneous triggers. We study whether explicit state dynamics can expose such pre-escalation signals without delegating clinical authority to the agent. We introduce a lightweight agent architecture in which a memoryless clinical risk encoder is integrated over time using first- and second-order dynamics to produce a continuous escalation pressure signal. Across synthetic ward scenarios, stateless agents exhibit sharp escalation cliffs, while second-order dynamics produce smooth, anticipatory concern trajectories despite similar escalation timing. These trajectories surface sustained unease prior to escalation, enabling human-in-the-loop monitoring and more informed intervention. Our results suggest that explicit state dynamics can make LLM agents more clinically legible by revealing how long concern has been rising, not just when thresholds are crossed.

Language models are saturating benchmarks for procedural tasks with narrow objectives. But they are increasingly being deployed in long-horizon, non-stationary environments with open-ended goals. In this paper we introduce KellyBench, an environment for evaluating sequential decision-making in sports betting markets. Agents are placed in a sequential simulation of the 2023-24 English Premier League season and tasked with maximising their long-term bankroll growth. They are given detailed historical data, including advanced statistics, lineups, and public odds. To succeed they must build machine learning models, identify edge in public markets, and adapt as the environment changes over time. We find that all frontier models evaluated lose money on average over the course of the season for five seeds. The best performing model achieves an average return of -8%, and many models experiencing ruin across seeds. To judge strategy sophistication, we use a human expert rubric to grade each model and find their approaches to be unsophisticated compared to human baselines; Claude Opus 4.6 achieves a rubric score of 26.5%, which means there is significant room for improvement. KellyBench is available as an open-access API endpoint at https://openreward.ai/GeneralReasoning/KellyBench.

In Mixed-Criticality (MC) systems, although the high Worst-Case Execution Time (WCET) serves as a conservative upper bound representing the task's maximum execution time under all conditions, obtaining a low WCET is essential for representing realistic executions and improving utilization and Quality-of-Service (QoS). Nevertheless, determining appropriate low WCET(s) for lower-criticality (LO) modes poses a significant challenge. Opting for a very low value of this WCET enhances processor utilization by scheduling more tasks in LO mode. Conversely, employing a larger WCET ensures fewer mode switches, thereby enhancing QoS, albeit at the cost of processor utilization. This paper proposes an analytical approach, AnTi-MiCS, to determine the appropriate low WCET through design-time analysis of task executions. In some cases, a single low WCET may not be adequate to capture large variations in the execution time distribution, for example, in scenarios like bimodal distributions. Therefore, we further propose a scalable approach, MulTi-MiCS, to compute multiple appropriate low WCETs. This approach exploits the temporal correlation between subsequent inputs presented to the application. Experimental results, conducted on a real platform with embedded real-time benchmarks, demonstrate the efficacy of our proposed scheme, in which QoS is improved by 30.27% on average while reducing utilization waste by 35.89%, compared to existing approaches. Besides, MulTi-MiCS improves QoS by 6.41% compared to AnTi-MiCS while reducing utilization waste by 8.23%.

Decompositional jailbreaks pose a critical threat to large language models (LLMs) by allowing adversaries to fragment a malicious objective into a sequence of individually benign queries that collectively reconstruct prohibited content. In real-world deployments, LLMs face a continuous, untraceable stream of fully anonymized and arbitrarily interleaved requests, infiltrated by covertly distributed adversarial queries. Under this rigorous threat model, state-of-the-art defensive strategies exhibit fundamental limitations. In the absence of trustworthy user metadata, they are incapable of tracking global historical contexts, while their deployment of generative models for real-time monitoring introduces computationally prohibitive overhead. To address this, we present TwinGate, a stateful dual-encoder defense framework. TwinGate employs Asymmetric Contrastive Learning (ACL) to cluster semantically disparate but intent-matched malicious fragments in a shared latent space, while a parallel frozen encoder suppresses false positives arising from benign topical overlap. Each request requires only a single lightweight forward pass, enabling the defense to execute in parallel with the target model's prefill phase at negligible latency overhead. To evaluate our approach and advance future research, we construct a comprehensive dataset of over 3.62 million instructions spanning 8,600 distinct malicious intents. Evaluated on this large-scale corpus under a strictly causal protocol, TwinGate achieves high malicious intent recall at a remarkably low false positive rate while remaining highly robust against adaptive attacks. Furthermore, our proposal substantially outperforms stateful and stateless baselines, delivering superior throughput and reduced latency.

Reinforcement Learning (RL) has traditionally focused on training specialized agents to optimize predefined reward functions within narrowly defined environments. However, the advent of powerful Large Language Models (LLMs) and increasingly complex, open-ended tasks has catalyzed a paradigm shift towards agentic paradigms within RL. This emerging framework extends beyond traditional RL by emphasizing the development of autonomous agents capable of goal-setting, long-term planning, dynamic strategy adaptation, and interactive reasoning in uncertain, real-world environments. Unlike conventional approaches that rely heavily on static objectives and episodic interactions, LLM-based Agentic RL incorporates cognitive-like capabilities such as meta-reasoning, self-reflection, and multi-step decision-making directly into the learning loop. In this paper, we provide a deep insight for looking the conceptual foundations, methodological innovations, and effective designs underlying this trend. Furthermore, we identify critical challenges and outline promising future directions for building LLM-based Agentic RL.

AI inference is becoming a persistent and geographically distributed source of electricity demand. Unlike many traditional electrical loads, inference workloads can sometimes be executed away from the user-facing service location, provided that latency, state locality, capacity, and regulatory constraints remain acceptable. This paper studies when such digital relocation of computation can be interpreted as latency-constrained relocation of electricity demand. We develop an energy-geography framework for geo-distributed AI inference. The framework models a three-layer architecture of clients, service nodes, and compute nodes, and formulates inference placement as a constrained optimization problem over electricity prices, marginal carbon intensity, power usage effectiveness, compute capacity, network latency, and migration frictions. The key object is the energy-latency frontier: the marginal cost and carbon benefit unlocked by relaxing inference latency budgets. The paper makes four contributions. First, it distinguishes physical electricity transmission from digital relocation of electricity-consuming computation. Second, it formulates a geo-distributed inference placement model with feasibility masks and migration frictions. Third, it introduces operational metrics, including relocatable inference demand, energy return on latency, carbon return on latency, and a relocation break-even condition. Fourth, it provides a transparent stylized simulation over representative global compute regions to show how heterogeneous latency tolerance separates workloads into local, regional, and energy-oriented execution layers. The results show that latency relaxation expands feasible geography, while migration frictions, egress costs, state locality, legal constraints, and capacity limits can sharply reduce realized benefits.

Although precise recall is a core objective in Retrieval-Augmented Generation (RAG), a critical oversight persists in the field: improvements in retrieval performance do not consistently translate to commensurate gains in downstream reasoning. To diagnose this gap, we propose the Recall Conversion Rate (RCR), a novel evaluation metric to quantify the contribution of retrieval to reasoning accuracy. Our quantitative analysis of mainstream RAG methods reveals that as Recall@5 improves, the RCR exhibits a near-linear decay. We identify the neglect of retrieval quality in these methods as the underlying cause. In contrast, approaches that focus solely on quality optimization often suffer from inferior recall performance. Both categories lack a comprehensive understanding of retrieval quality optimization, resulting in a trade-off dilemma. To address these challenges, we propose comprehensive retrieval quality optimization criteria and introduce the NeocorRAG framework. This framework achieves holistic retrieval quality optimization by systematically mining and utilizing Evidence Chains. Specifically, NeocorRAG first employs an innovative activated search algorithm to obtain a refined candidate space. Then it ensures precise evidence chain generation through constrained decoding. Finally, the retrieved set of evidence chains guides the retrieval optimization process. Evaluated on benchmarks including HotpotQA, 2WikiMultiHopQA, MuSiQue, and NQ, NeocorRAG achieves SOTA performance on both 3B and 70B parameter models, while consuming less than 20% of tokens used by comparable methods. This study presents an efficient, training-free paradigm for RAG enhancement that effectively optimizes retrieval quality while maintaining high recall. Our code is released at https://github.com/BUPT-Reasoning-Lab/NeocorRAG.

Task-based dialogue systems assist users in achieving specific goals, such as executing actions or retrieving information, through natural language interactions. Accurate coreference resolution is essential, as it involves identifying object references within the dialogue - a task that becomes increasingly challenging in visually grounded environments characterized by complex scenes and diverse object metadata. However, coreference resolution in task-based dialogue remains limited by poor generalization across domains and heavy reliance on supervised models that often overfit to dataset-specific artifacts. In this work, we propose a unimodal test-time reasoning approach that enables large language models (LLMs) to reason over detailed object metadata and dialogue history to improve coreference resolution. Empirical results on the SIMMC 2.1 dataset demonstrate that LLMs can generate step-by-step reasoning processes that effectively align dialogue context with objects present in the scene. Extensive experiments highlight the models' ability to link conversations and objects accurately. Moreover, we show that test-time reasoning under few-shot settings generalizes effectively to unseen scenarios and novel objects, outperforming encoder-based supervised methods in cross-domain evaluations. These findings underscore the critical role of structured metadata and careful prompt engineering in enhancing the robustness and generalization of task-oriented dialogue systems.

This paper presents a configurable, grid-aware Agent-Based Model (ABM) for the systematic analysis of electric vehicle (EV) charging systems under configurable infrastructure and operational conditions. The model integrates heterogeneous EV behavior, charging column constraints, and a shared Energy Sandbox that regulates aggregate power allocation, enabling the joint study of user-centric charging dynamics and facility-level power behavior. Implemented in Python using the SimPy discrete-event framework, the approach supports scalable, event-driven simulations across varying system sizes, charger compositions, and scheduling strategies. A representative workplace charging scenario is investigated to illustrate how infrastructure configuration and coordination mechanisms influence energy delivery performance, infrastructure utilization, and aggregate load characteristics. The results highlight the context-dependence of infrastructure suitability and demonstrate how charging strategies and charger types reshape both service-level outcomes and grid-facing behavior. The proposed ABM provides a flexible and extensible simulation environment for exploring technical, operational, and grid-aware aspects of EV charging ecosystems, and for serving as a methodological basis for subsequent studies on advanced coordination strategies beyond the specific scenario analyzed in this study.

How people narrate their experiences offers a window into how the mind organizes them. Computational approaches to therapeutic writing have evolved from lexical counting to neural methods, yet remain fragmented: dictionary tools miss discourse structure, while embeddings conflate local coherence with global organization. No existing framework maps these techniques onto the hierarchical processes through which narratives are constructed. Here we introduce a three-level framework - micro-level lexical features, meso-level semantic embeddings, and macro-level LLM narrative evaluation - and show, across 830 Chinese therapeutic texts spanning depression, anxiety, and trauma, that macro-level evaluation substantially outperforms lexical and embedding features for mental health prediction. This challenges the field's emphasis on word-counting: formal structural features (Labov's story grammar, RST coherence, propositional composition) demonstrate that narrative organization per se carries predictive signal, while clinically-grounded narrative dimensions capture how psychological states are expressed through discourse. Semantic embeddings add minimal independent value but yield incremental gains in multi-level classification. By grounding computational levels in discourse processing theory, this framework identifies macro-structural organization as the primary locus of clinical signal and generates testable hypotheses for intervention design and longitudinal research.

Communication has emerged as a critical bottleneck in the distributed training of large language models (LLMs). While numerous approaches have been proposed to reduce communication overhead, the potential of lossless compression has remained largely underexplored since compression and decompression typically consume larger overheads than the benefits of reduced communication traffic. We observe that the communication data, including activations, gradients and parameters, during training often follows a near-Gaussian distribution, which is a key feature for data compression. Thus, we introduce ZipCCL, a lossless compressed communication library of collectives for LLM training. ZipCCL is equipped with our novel techniques: (1) theoretically grounded exponent coding that exploits the Gaussian distribution of LLM tensors to accelerate compression without expensive online statistics, (2) GPU-optimized compression and decompression kernels that carefully design memory access patterns and pipeline using communication-aware data layout, and (3) adaptive communication strategies that dynamically switch collective operations based on workload patterns and system characteristics. Evaluated on a 64-GPU cluster using both mixture-of-experts and dense transformer models, ZipCCL reduces communication time by up to 1.35$\times$ and achieves end-to-end training speedups of up to 1.18$\times$ without any impact on model quality.

Recently, large language models (LLMs) have shown great promise in time series forecasting. However, most existing LLM-based forecasting methods still follow a static generative paradigm that directly maps historical observations to future values in a single pass. Under this paradigm, forecasting is constrained by limited temporal pattern extraction, single-round acquisition of contextual features, one-shot forecast generation, and lack of support from ensemble forecasts. To address these limitations, in this work, we propose CastFlow, a dynamic agentic forecasting framework that enables multi-view temporal pattern extraction, multi-round contextual features acquisition, iterative forecast refinement, and forecasting with ensemble forecasts. First, CastFlow organizes the forecasting process into planning, action, forecasting, and reflection, establishing an agentic workflow. Second, this workflow is supported by a memory module that retrieves prior experience and a multi-view toolkit that constructs diagnostic evidence and provides a reliable ensemble forecast baseline. Third, CastFlow adopts a role-specialized design that combines general-purpose reasoning with specialized numerical forecasting. Under this design, a frozen LLM preserves general-purpose reasoning, while a fine-tuned domain-specific LLM performs evidence-guided numerical forecasting based on the ensemble forecast baseline, rather than from scratch. To optimize a fine-tuned domain-specific LLM, we further develop a two-stage workflow-oriented training that combines supervised fine-tuning (SFT) and reinforcement learning with verifiable rewards (RLVR). To evaluate the effectiveness of CastFlow, we conduct extensive experiments on diverse datasets and show that it achieves superior overall results against strong baselines. We hope that this work can serve as a step toward more adaptive and accurate time series forecasting.

Prototype-based Personalized Federated Learning (ProtoPFL) enables efficient multi-domain adaptation by communicating compact class prototypes, but directly sharing them poses privacy risks. A common defense involves per-example $\ell_2$ clipping before prototype computation to bound sensitivity, followed by isotropic Gaussian noise to enforce Local Differential Privacy (LDP). However, Isotropic Gaussian Prototype Perturbation (IGPP) typically over-perturbs discriminative dimensions and struggles to balance the clipping threshold with representation fidelity. In this paper, we propose VPDR, a client-side privacy plug-in that seamlessly integrates into existing ProtoPFLs. Motivated by the observation that dimension-wise class variance reflects discriminability, we introduce Variance-adaptive Prototype Perturbation (VPP), which allocates less noise to discriminative subspaces, preserving semantic separability while ensuring privacy. We further develop Distillation-guided Clipping Regularization (DCR), which enables feature norms to adaptively concentrate near the predefined clipping threshold while maintaining prediction consistency. Theoretical analysis shows that our groupwise mechanism provides privacy guarantees no weaker than the isotropic baseline under the same privacy constraints. Extensive experiments on multi-domain benchmarks demonstrate that VPDR achieves a superior privacy-utility trade-off, outperforming IGPP in personalized federated fine-tuning without sacrificing robustness against realistic attacks.

AI integration in automotive perception systems shifts requirements from static specifications to continuously evolving entities shaped by data, models, and operating contexts. When such changes are not consistently documented, validated, and traced, they accumulate as Requirements Debt (ReD), an underexplored but critical subtype of technical debt. This study conceptualises and empirically investigates how evolving functional and non-functional requirements create and propagate ReD across the AI-enabled automotive perception system lifecycle. We conducted 16 semi-structured interviews with experts from 13 international automotive companies and 3 European research institutes, and analysed the data using thematic analysis. As one of the first empirical studies connecting technical debt theory with RE4AI, the work identifies key ReD mechanisms. Evolving functional requirements (e.g., algorithm updates, sensor fusion, architectural changes, real-time constraints) drive semantic drift, validation backlogs, and integration debt when verification lags behind rapid iteration. In parallel, evolving non-functional requirements (e.g., safety, cybersecurity, reliability, scalability, transparency, trustworthiness) create assurance lag, compliance misalignment, and transparency and reliability debt as standards and ethical expectations shift. These interacting mechanisms propagate ReD across data, models, and system artefacts, undermining auditability, reliability, and certification readiness in safety-critical perception systems.

Every document format in existence was designed for a human reader moving linearly through text. Autonomous LLM agents do not read - they retrieve. This fundamental mismatch forces agents to inject entire documents into their context window, wasting tokens on irrelevant content, compounding state across multi-turn loops, and broadcasting information indiscriminately across agent roles. We argue this is not a prompt engineering problem, not a retrieval problem, and not a compression problem: it is a format problem. We introduce OBJECTGRAPH (.og), a file format that reconceives the document as a typed, directed knowledge graph to be traversed rather than a string to be injected. OBJECTGRAPH is a strict superset of Markdown - every .md file is a valid .og file - requires no infrastructure beyond a two-primitive query protocol, and is readable by both humans and agents without tooling. We formalize the Document Consumption Problem, characterise six structural properties no existing format satisfies simultaneously, and prove OBJECTGRAPH satisfies all six. We further introduce the Progressive Disclosure Model, the Role-Scoped Access Protocol, and Executable Assertion Nodes as native format primitives. Empirical evaluation across five document classes and eight agent task types demonstrates up to 95.3 percent token reduction with no statistically significant degradation in task accuracy (p > 0.05). Transpiler fidelity reaches 98.7 percent content preservation on a held-out document benchmark.

Multi-server MCP agents create an information-flow control problem: faithful tool composition can turn individually benign read/write permissions into cross-boundary credential propagation -- a structural side effect of workflow topology, not necessarily malicious model behavior. We present MCPHunt, to our knowledge the first controlled benchmark that isolates non-adversarial, verbatim credential propagation across multi-server MCP trust boundaries, with three methodological contributions: (1) canary-based taint tracking that reduces propagation detection to objective string matching; (2) an environment-controlled coverage design with risky, benign, and hard-negative conditions that validates pipeline soundness and controls for credential-format confounds; (3) CRS stratification that disentangles task-mandated propagation (faithful execution of verbatim-transfer instructions) from policy-violating propagation (credentials included despite the option to redact). Across 3,615 main-benchmark traces from 5 models spanning 147 tasks and 9 mechanism families, policy-violating propagation rates reach 11.5--41.3% across all models. This propagation is pathway-specific (25x cross-mechanism range) and concentrated in browser-mediated data flows; hard-negative controls provide evidence that production-format credentials are not necessary -- prompt-directed cross-boundary data flow is sufficient. A prompt-mitigation study across 3 models reduces policy-violating propagation by up to 97% while preserving 80.5% utility, but effectiveness varies with instruction-following capability -- suggesting that prompt-level defenses alone may not suffice. Code, traces, and labeling pipeline are released under MIT and CC BY 4.0.

Joint probabilistic modeling is essential for forecasting irregular multivariate time series (IMTS) to accurately quantify uncertainty. Existing approaches often struggle to balance model expressivity with consistent marginalization, frequently leading to unreliable or contradictory forecasts. To address this, we propose CircuITS, a novel architecture for probabilistic IMTS forecasting based on probabilistic circuits. Our model is flexible in capturing intricate dependencies between time series channels while structurally guaranteeing valid joint distributions. Experiments on four real world datasets demonstrate that CircuITS achieves superior joint and marginal density estimation compared to state of the art baselines.

Computational molecular representations underpin virtual screening, property prediction, and materials discovery. Conventional fingerprints are efficient and deterministic but lose structural information through hash-based compression, particularly at low dimensionalities. Learned representations from graph neural networks recover this expressiveness but require task-specific training and substantial computational resources. Here we introduce hyperdimensional fingerprints (HDF), which replace the learned transformations of message-passing neural networks with algebraic operations on high-dimensional vectors, producing deterministic molecular representations without any training. Across diverse property prediction benchmarks, HDF outperforms conventional fingerprints in the majority of tasks while exhibiting greater consistency across datasets and models. Crucially, HDF embeddings preserve molecular similarity faithfully: at 32 dimensions, distances in HDF space achieve a 0.9 Pearson correlation with graph edit distance, compared to 0.55 for Morgan fingerprints at equivalent size. This structural fidelity persists at low dimensions where hash-based methods degrade, allowing simple nearest-neighbor regression to remain predictive with as few as 64 components. We further demonstrate the practical impact in Bayesian molecular optimization, where HDF-based surrogate models achieve substantially improved sample efficiency in regimes where Morgan fingerprints perform comparably to random search. HDF thus provides a general-purpose, training-free alternative to conventional molecular fingerprints, suggesting that the information loss long accepted as inherent to fixed-length fingerprints is a limitation of the hash-based encoding scheme rather than the fingerprint paradigm itself.

High Bandwidth Memory with Processing-in-Memory (HBM-PIM) offers an opportunity to reduce data movement by executing computation directly inside memory, but current commercial platforms expose limited instruction sets and require specialized software stacks. In this work, we investigate whether HBM-PIM can serve as a backend for ISA-level matrix acceleration, using the RISC-V Attached Matrix Extension (AME) as a semantic reference. We propose a PEP-based execution model that maps AME element-wise and matrix instructions to HBM-PIM micro-kernels and data instructions in memory operations. Differently from SoA HBM-PIM, we introduce a reduction-free outer-product dataflow that enables accumulation entirely within memory despite the lack of native reduction support. Our approach supports end-to-end execution of element-wise operations, GEMV, and GEMM in PIM mode, minimizing host involvement and off-chip transfers. An experimental evaluation on Samsung Aquabolt-XL shows that AME matrix tile multiplication achieves up to 14.9 GFLOP/s (59.4 FLOP/cycle) on a single HBM pseudo-channel.

The design of embedded safety-critical systems such as those used in next-generation automotive and autonomous platforms, is increasingly challenged by escalating system complexity, hardware-software heterogeneity, and the integration of intelligent, data-driven components. Ensuring dependability in such systems requires a holistic approach that spans multiple abstraction layers and encompasses both design- and run-time assurance. Traditional methods for reliability, safety, and security management often fall short in addressing the dynamic and uncertain behaviors introduced by Artificial Intelligence (AI) and Machine Learning (ML) components, especially under stringent real-time, power, and safety constraints. While AI and ML offer powerful predictive, adaptive, and self-optimizing capabilities that can enhance system dependability, their inherent non-determinism, data-dependence, and lack of formal guarantees introduce new challenges for verification, validation, and certification. This paper explores emerging methodologies, architectures, and frameworks for designing dependable autonomous and embedded systems in the era of AI. It highlight advances in reliability modeling, secure system design, and certification approaches that account for imperfect, learning-enabled components, aiming to bridge the gap between AI innovation and certifiable system-level dependability.

Cross-chain NFT migration refers to the process of transferring digital assets along with their associated functionalities and guarantees between distinct blockchain platforms. However, architectural divergences among these platforms introduce critical challenges, often resulting in features that fail to behave as intended. While protocol-level mechanisms can coordinate data transfer, they are insufficient to resolve deeper compatibility issues arising from fundamental differences in state organization, transaction execution, and ownership representation. Thus, the critical challenge lies in predicting which NFT features can be preserved, which require redesign, and which are fundamentally incompatible, prior to undertaking costly migration attempts. To address this challenge, we first derive a tailored four-layer NFT architecture based on standard blockchain stacks, distinguishing cryptographic, state-management, transaction-processing, and ownership primitives, with explicit upward dependencies. Building on this architecture, we conceptualize an NFT as a bundle of features and define successful cross-chain NFT migration as the preservation of these features. Grounded in this model, we propose a four-phase migration analysis methodology comprising source feature specification, primitive-level dependency mapping, target platform profiling, and compatibility assessment, which classifies each feature as natively preserved, partially mismatched, or completely mismatched. We evaluate this methodology through a proof-of-concept analysis of Ethereum-to-Solana NFT migration, identifying several incompatibility issues that hinder seamless NFT migration.

The rapid proliferation of image generation models and other artificial intelligence (AI) systems has intensified concerns regarding data privacy and user consent. As the availability of public datasets declines, major technology companies increasingly rely on proprietary or private user data for model training, raising ethical and legal challenges when users request the deletion of their data after it has influenced a trained model. Machine unlearning seeks to address this issue by enabling the removal of specific data from models without complete retraining. This study investigates a modified SISA (Sharded, Isolated, Sliced, and Aggregated) framework designed to achieve class-level unlearning in Convolutional Neural Network (CNN) architectures. The proposed framework incorporates a reinforced replay mechanism and a gating network to enhance selective forgetting efficiency. Experimental evaluations across multiple image datasets and CNN configurations demonstrate that the modified SISA approach enables effective class unlearning while preserving model performance and reducing retraining overhead. The findings highlight the potential of SISA-based unlearning for deployment in privacy-sensitive AI applications. The implementation is publicly available at https://github.com/SiamFS/ sisa-class-unlearning.

Exponential growth in the scale of modern foundation models has led to the widespread adoption of Low-Rank Adaptation (LoRA) as a parameter-efficient fine-tuning technique. However, standard LoRA implementations disregard the varying intrinsic dimensionality of model layers and enforce a uniform rank, leading to parameter redundancy. We propose Post-Optimization Adaptive Rank Allocation (PARA), a data-free compression method for LoRA that integrates seamlessly into existing fine-tuning pipelines. PARA leverages Singular Value Decomposition to prune LoRA ranks using a global threshold over singular values across all layers. This results in non-uniform rank allocation based on layer-wise spectral importance. As a post-hoc method, PARA circumvents the training modifications and resulting instabilities that dynamic architectures typically incur. We empirically demonstrate that PARA reduces parameter count by 75-90\% while preserving the predictive performance of the original, uncompressed LoRA across multiple vision and language benchmarks. Code will be published upon acceptance.

Generative AI is being increasingly integrated into web search for the convenience it provides users. In this work, we aim to understand how generative AI disrupts web search by retrieving and presenting the information and sources differently from traditional search engines. We introduce a public benchmark dataset of 11,500 user queries to support our study and future research of generative search. We compare the search results returned by Google's search engine, the accompanying AI Overview (AIO), and Gemini Flash 2.5 for each query. We have made several key findings. First, we find that for 51.5\% of representative, real-user queries, AIOs are generated, and are displayed above the organic search results. Controversial questions frequently result in an AIO. Second, we show that the retrieved sources are substantially different for each search engine (<0.2 average Jaccard similarity). Traditional Google search is significantly more likely to retrieve information from popular or institutional websites in government or education, while generative search engines are significantly more likely to retrieve Google-owned content. Third, we observe that websites that block Google's AI crawler are significantly less likely to be retrieved by AIOs, despite having access to the content. Finally, AIOs are less consistent when processing two runs of the same query, and are less robust to minor query edits. Our findings have important implications for understanding how generative search impacts website visibility, the effectiveness of generative engine optimization techniques, and the information users receive. We call for revenue frameworks to foster a sustainable and mutually beneficial ecosystem for publishers and generative search providers.

Large Language Models (LLMs) are increasingly used as core dependencies in software systems. However, the hosted LLM services evolve continuously through provider-side updates without explicit version changes. These silent updates can introduce behavioral drift, causing regressions in functionality, formatting, safety constraints, or other application-specific requirements. Existing approaches focus primarily on regression testing or versioning but do not provide deployer-side mechanisms for governing compatibility during opaque model evolution. This paper proposes a deployment-side governance framework based on three components: clearly defined rules for how the model is allowed to behave (production contracts), focused testing organized by deployment risk categories (risk-category-based testing suite), and release checkpoints that block updates unless they meet defined safety and performance standards (compatibility gates). Through exploratory validation across multiple LLM versions, we provide evidence that targeted testing in specific risk areas can uncover performance regressions that overall metrics miss. We also identify several open research challenges, including how to systematically build effective test suites, how to set reliable performance thresholds in non-deterministic systems, and how to detect and explain model drift when providers offer limited transparency. Overall, we frame LLM update management as a software supply chain governance problem and outline a research agenda for putting deployer-side compatibility controls into practice.

Semidefinite programs (SDPs) are a powerful framework for convex optimization and for constructing strong relaxations of hard combinatorial problems. However, solving large SDPs can be computationally expensive, motivating the use of machine learning models as fast computational surrogates. Graph neural networks (GNNs) are a natural candidate in this setting due to their sparsity-awareness and ability to model variable-constraint interactions. In this work, we study what expressive power is sufficient to recover optimal SDP solutions. We first prove negative results showing that standard GNN architectures fail on recovering linear SDP solutions. We then identify a more expressive architecture that captures the key structure of SDPs and can, in particular, emulate the updates of a standard first-order solver. Empirically, on both synthetic and \textsc{SdpLib} benchmarks of various classes of SDPs, this more expressive architecture achieves consistently lower prediction error and objective gap than theoretically weaker baselines. Finally, using the learned high-quality predictions to warm-start the first-order solver yields practical speedups of up to 80%.

AI systems rest on software with low integrity mechanisms, leaving AI systems exposed across every stage from data acquisition to final inference. This paper makes the AI supply chain a first-class object of analysis, decomposing it across four architectural layers: data acquisition, model training, model inference, and a cross-cutting substrate. Within these layers, we identify four structural gaps that traditional supply chain mechanisms do not address: verifiability, versioning, observability, and traceability.Current AI systems fall short on all of them: they carry undeclared behavioral couplings that no resolver enforces; they cannot be reverted back to known working assemblies; they degrade silently rather than surfacing breaking changes; and their lineage can hardly be approximated. To illustrate the scale of the software supply chain of AI, we measure a reference stack of 48 production-grade open-source projects, which declares 4,664 direct dependencies, resolves to 11,508 transitive packages, and totals roughly 392M lines of code.

Large Language Models (LLMs) have rapidly improved in performance across code-related tasks, making their integration into Register Transfer Level (RTL) development increasingly attractive. Mimicking the behavior of inline code assistants, many benchmarks evaluate LLMs' capabilities in code completion, either assessing the generation of entire hardware modules or the completion of a single line within a module. However both of these approaches lack the ability to control the granularity of the code-completion sample size and the syntactic range of completions. To overcome these limitations, we present a framework for language-agnostic rule completion (RuC), a grammar-driven, rule-selectable benchmark generator that automatically produces RTL code-completion tasks from a set of input hardware description sources. RuC uses the target Hardware Description Language (HDL) grammar to mask syntactically defined code regions and prompts a model to regenerate them using the surrounding unmasked code as context, enabling a controlled and scalable evaluation of the domain-specific model's code-understanding capabilities, ranging from assignments to the reconstruction of entire logic blocks. We use RuC to generate two SystemVerilog rule-completion benchmarks from the Tiny Tapeout shuttle TT07 and the CVE2 RISC-V core to demonstrate RuC's applicability to a broad range of designs, and conduct a comparative study of the code completion capabilities of modern open-source LLMs across diverse settings. Results indicate that completion performance strongly depends on the model type, the grammatical structure of the masked region, and the prompting strategy. Specifically, the highest scores are obtained with Fill-in-the-Middle (FIM) prompting. These findings highlight the value of grammar-driven, arbitrarily granular benchmarks for meaningful evaluation of LLM capabilities in RTL development workflows.

While GUI agents have shown impressive capabilities in common computer-use tasks such as OSWorld, current benchmarks mainly focus on isolated and single-application tasks. This overlooks a critical real-world requirement of coordinating across multiple applications to accomplish complex profession-specific workflows. To bridge this gap, we present a computer-use benchmark in cross-application workflows, named WindowsWorld, designed to systematically assess GUI Agents on complex multi-step tasks that mirror real-world professional activities. Our methodology uses a multi-agent framework steered by 16 occupations to generate four difficulty-level tasks with intermediate inspection, which are then refined by human review and executed in a simulated environment. The resulting benchmark contains 181 tasks with an average of 5.0 sub-goals across 17 common desktop applications, of which 78% are inherently multi-application. Experimental results of leading large models and agents show that: 1) All computer-use agents perform poorly on multi-application tasks (< 21% success rate), far below the performance of simple single-app tasks; 2) They largely fail at tasks requiring conditional judgment and reasoning across $\geq$ 3 applications, stalling at early sub-goals; 3) Low execution efficiency, where tasks often fail despite far exceeding human step limits. Code, benchmark data, and evaluation resources are available at github.com/HITsz-TMG/WindowsWorld.

Shallow nanoindentation enables mechanical characterization of thin films, individual phases and other volume-constrained materials, but measured hardness is often inflated by the indentation size effect (ISE), contact-area errors and tip-geometry artifacts. Classical ISE corrections such as the Nix-Gao require a deep linear regime and are unreliable when only shallow measurements are used. This study investigates how a small experimental dataset can be used to predict a reference hardness with physics-guided feature engineering and augmentation. Approximately 700 experimental indentations were collected from three steel reference specimens covering a hardness range of 2-6.5 GPa and augmented using physically motivated variations representing instrumental noise, session-level drift, and local multiphase boundary blending. The input space combined Oliver-Pharr values with mechanics descriptors, including indentation work partitioning, ($H\text{/}E_{r}$), and the area-invariant compliance proxy ($P_{\max}\text{/}S^{2}$). Ridge Regression (RR), Random Forest, XGBoost, and Neural Networks (NN) were evaluated using a quarantined fourth steel specimen tested at staggered loads. The hardness mapping was nonlinear: RR failed, whereas nonlinear models achieved ($R^2 > 0.98$) internally. A constrained (64-8-64) NN gave the best results, reaching RMSE = 0.470 GPa, MAPE = 5.4% on the quarantined steel. Unlike Nix-Gao analysis, the NN produced stable estimates in the shallow regime. SHAP and latent-space analysis showed reliance on area-invariant and energy-based descriptors. The results demonstrate the feasibility of a this workflow for ISE correction in steels using small datasets and suggest a pathway toward data-efficient characterization of any volume constrained materials.

Population protocols are a model of distributed computation in which a collection of indistinguishable finite-state agents interact randomly in pairs to decide a predicate of their initial configuration. The agents decide by achieving a stable consensus on whether the predicate holds or not. It is known that population protocols can decide exactly the predicates expressible in Presburger arithmetic. Recently, Lossin et al. have introduced a notion of protocol robustness against adversarial crash failures. They show that all atomic Presburger predicates can be decided by robust protocols, and ask whether the same holds for every Presburger predicate. We make progress towards settling this question by proving that all predicates expressible in monadic Presburger arithmetic have robust protocols. In addition, we analyze the cost of robustness in terms of state complexity. We study the ratio between the number of states of the smallest robust protocol for a given predicate and the smallest protocol for it. We show that the cost of robustness is at least double exponential in the size of the predicate, and prove that the robust protocols by Lossin et al. for threshold predicates x >= k have optimal state complexity.

Poetry has long been a central art form for Arabic speakers, serving as a powerful medium of expression and cultural identity. While modern Arabic speakers continue to value poetry, existing research on Arabic poetry within Large Language Models (LLMs) has primarily focused on analysis tasks such as interpretation or metadata prediction, e.g., rhyme schemes and titles. In contrast, our work addresses the practical aspect of poetry creation in Arabic by introducing controllable generation capabilities to assist users in writing poetry. Specifically, we present a large-scale, carefully curated instruction-based dataset in Modern Standard Arabic (MSA) and various Arabic dialects. This dataset enables tasks such as writing, revising, and continuing poems based on predefined criteria, including style and rhyme, as well as performing poetry analysis. Our experiments show that fine-tuning LLMs on this dataset yields models that can effectively generate poetry that is aligned with user requirements, based on both automated metrics and human evaluation with native Arabic speakers. The data and the code are available at https://github.com/mbzuai-nlp/instructpoet-ar

The emergence of Large Language Models (LLMs) offers a transformative interface for Web3, yet existing benchmarks fail to capture the complexity of translating high-level user intents into functionally correct, state-dependent on-chain transactions. We present \textsc{Intent2Tx}, a high-fidelity benchmark featuring 29,921 single-step and 1,575 multi-step instances meticulously derived from 300 days of real-world Ethereum mainnet traces. Unlike prior works that rely on synthetic instructions, \textsc{Intent2Tx} grounds natural language intents in real-world protocol interactions across 11 categories, including diverse long-tail Decentralized Finance (DeFi) primitives. To enable rigorous evaluation, we propose an execution-aware framework that transcends surface-level text matching by employing differential state analysis on forked mainnet environments. Our extensive evaluation of 16 state-of-the-art LLMs reveals that while scaling and retrieval-augmentation enhance logical consistency and parameter precision, current models struggle with out-of-distribution generalization and multi-step planning. Crucially, our execution-based analysis demonstrates that syntactically valid outputs often fail to achieve intended state transitions, highlighting a significant gap in current "reasoning-to-execution" capabilities. \textsc{Intent2Tx} serves as a critical foundation for developing autonomous, reliable agents in intent-centric Web3 ecosystems. Code and data: https://anonymous.4open.science/r/Intent2Tx_Bench-97FF .

Integrating domain knowledge into deep neural networks is a promising way to improve generalization. Existing methods either encode prior knowledge in the loss function or apply post-processing modules, but both depend on identifying useful symbolic knowledge to integrate. Since such rules are often unavailable in real-world vision tasks, we propose a method for targeted knowledge discovery. We propose a Differentiable Knowledge Unit (DKU) that enables modulating the classifier logits, yielding refined class probabilities. The DKU uses implication rules to represent relationships between task classes and implicit concepts learned entirely from the main task supervision, without requiring concept labels. Concepts are identified by dedicated classifiers, whose probabilities are passed to DKU alongside the primary class probabilities. DKU computes a logic-based adjustment vector via fuzzy inference, which modulates the primary class logits to yield refined class probabilities. When concept classifiers represent concepts that do not support the logical rule structure, the resulting adjustments to the class probabilities do not directly minimize the supervision loss. Consequently, optimizing the supervision loss on these adjusted class probabilities implicitly trains the concept classifiers. We construct the rule base so that bidirectional logical relations connect concepts and classes. We enforce the concepts to be distinct from each other and with respect to the classes. This design enforces a clean supervision signal for concept learning. We evaluate our methods on the PASCAL-VOC, COCO, and MedMNIST datasets. We demonstrate improvement through our knowledge integration across these datasets. We conduct domain generalization and hard-sample ablation studies and find that our implicit knowledge discovery and integration outperforms the baseline.

Open-source projects often rely on a small group of highly active contributors known as hero developers. Prior work shows that hero developers are common in many OSS and enterprise projects, yet who qualifies as a hero depends heavily on the chosen contribution metric. Code-based metrics identify implementation-focused developers, whereas discussion-based metrics highlight coordination and communication; these metrics capture distinct facets of contribution. We conducted a measurement-sensitive study of multifaceted heroism across 77 Apache Software Foundation projects using three technical measures (commit count, distinct files touched, churn) and two social measures (issue-comment count, number of distinct issues commented on). We examined hero prevalence, overlap among hero sets, and severity-wise bug-fixing outcomes via fix and reopen rates. Results show that hero projects are common under all measures, but identified heroes differ substantially across facets. The pooled Jaccard overlap between technical and social hero sets is only 0.10. Cross-facet asymmetry is evident: 71.4% of technical heroes exhibit strong social activity, while only 24.2% of social heroes show strong technical activity. Fix-rate and reopen-rate differences are modest, yet hero-category rankings vary across severity levels and outcome measures. These findings indicate that heroism is not a single, metric-independent role. A multifaceted perspective offers a more reliable understanding of key contributors and better supports developer prioritisation and severity-aware bug assignment.

This article outlines a new framework of traffic light optimization through a digital twin of the transport infrastructure, managed by agentic AI to ensure real-time autonomous decisions. The framework relies on physical sensors and edge computing to measure real-time traffic information and simulate traffic flow in a constantly updated digital twin. The traffic light is automatically controlled through the digital twin according to traffic congestion, travel delay and traffic patterns. This approach is implemented as a three-layer system: perception, conceptualization and action. The perception layer receives data on physical systems; the conceptualization layer uses LangChain to process the data; and the action layer links to the Model Context Protocol (MCP) and traffic management APIs to implement optimised traffic signal control algorithms. The results show that the framework minimizes waiting time at traffic lights and positively affects the effectiveness of the entire traffic flow, which is better than the fixed-time and reinforcement learning-based baselines.

Large language model (LLM)-based generative list-wise recommendation has advanced rapidly, but decoding remains sequential and thus latency-prone. To accelerate inference without changing the target distribution, speculative decoding (SD) uses a small draft model to propose several next tokens at once and a target LLM to verify and accept the longest prefix, skipping multiple steps per round. In generative recommendation, however, each item is represented by multiple semantic-ID tokens, often with separators, and current drafts typically treat these tokens uniformly. This overlooks two practical facts: (i) a token's semantics depend on its within-item slot, and (ii) uncertainty tends to increase with speculation depth. Without modeling these effects, SD's speedups can be limited. We introduce PAD-Rec, Position-Aware Drafting for generative Recommendation, a lightweight module that augments the draft model with two complementary signals. Item position embeddings explicitly encode the within-item slot of each token, strengthening structural awareness. Step position embeddings encode the draft step, allowing the model to adapt to depth-dependent uncertainty and improve proposal quality. To harmonize these signals with base features, we add simple gates: a learnable coefficient for item slots and a context-driven gate for draft steps. The module is trainable, easy to integrate with standard draft models, and adds negligible inference overhead. Extensive experiments on four real-world datasets show up to 3.1x wall-clock speedup and about 5% average wall-clock speedup gain over strong SD baselines, while largely preserving recommendation quality.

AI applications are increasingly being introduced into digital health. While technical performance has advanced rapidly, successful deployment mainly depends on consumer attitudes, especially to patient-facing applications. However, most existing research examines consumer attitudes towards healthcare AI at an abstract level rather than in response to concrete artefacts. We report a mixed-methods survey study in Australia (N=275) examining consumer readiness, acceptance, trust, and risk perceptions of healthcare AI, combined with a scenario-based evaluation of an AI-generated versus clinician-written consultation summary. Participants expressed moderate optimism and strong perceived usefulness and ease of use, but also substantial concerns about accuracy, safety, and data use. In the scenario task, the AI-generated summary was strongly preferred for quality, empathy, and overall usefulness, yet identification of the AI summary was near chance. Findings show that consumers judge AI through concrete communication quality and visible human governance, underscoring the need for clinically supervised deployment frameworks beyond technical performance alone.

We develop a geometric and information-theoretic framework for encoder-decoder learning built on the Information Bottleneck (IB) principle. Recasting IB as a rate-distortion problem with Kullback-Leibler (KL) divergence as distortion, we show that the optimal representation at any distortion level is a soft clustering of the \emph{predictive manifold} $\mathcal{M}=\{p(Y|x):x\in\mathcal{X}\}$ inside the probability simplex, admitting a linear decoder in the canonical parameterization. We derive a chain of exact transformations, from flat Dirichlet to exponential to isotropic Gaussian, connecting the maximum entropy prior on the simplex to Euclidean space, with quantified entropy overhead at each step, and show that Sketched Isotropic Gaussian Regularization (SIGReg) implements a Gaussian relaxation of this principle whose overhead affects rate accounting but not achievable prediction. This relaxation provides a principled distributional regularizer for learning with limited or no supervision. Using the Conditional Entropy Bottleneck (CEB) decomposition, we derive concrete encoder losses for supervised and semi-supervised settings, estimated via minibatch marginals without variational bounds. In the self-supervised setting, the CEB conditional rate is replaced by a view-prediction proxy. SIGReg serves as the distributional regularizer for both the semi-supervised and self-supervised settings. Experiments on toy problems and FashionMNIST confirm the predicted rate-distortion trade-offs and show that the non-parametric estimator is competitive with the standard variational approach.

The choice of loss function in classification involves a fundamental trade-off: smooth losses (like Cross-Entropy) enable fast optimization rates but yield slow square-root consistency bounds, while piecewise-linear losses (like Hinge) offer fast linear consistency rates but suffer from non-differentiability. We propose Linear-Core (LC) Surrogates, a new family of convex loss functions that resolve this tension by stitching a linear core to a smooth tail. We prove that these surrogates are differentiable everywhere while retaining strict linear $H$-consistency bounds, effectively combining the optimization benefits of smoothness with the statistical efficiency of margin-based losses. In the structured prediction setting, we show that this smoothness unlocks a massive computational and energy advantage: it allows for an unbiased stochastic gradient estimator that bypasses the quadratic complexity $O(|\mathscr{Y}|^2)$ of exact inference (e.g., Viterbi). Empirically, our method achieves a 23$\times$ speedup over Structured SVMs on large-vocabulary sequence tagging tasks and demonstrates superior robustness to instance-dependent label noise, outperforming Cross-Entropy by 2.6% on corrupted CIFAR-10.

We study the problem of understanding where two populations differ within a feature space, which we formalize in the concept of a differential subgroup: a subset of individuals from both populations who, despite sharing similar characteristics, exhibit exceptional differences in a target outcome. Differential subgroups reveal the regions of the feature space where population-level gaps are most pronounced and can help practitioners identify the covariate combinations that are structurally responsible for these differences, e.g.~in clinical analysis, model diagnostics, or treatment-effect studies. We introduce a general optimization objective for discovering differential subgroups and establish conditions under which the resulting subgroups admit a causal interpretation of population differences. We propose DiffSub, a gradient-based approach that discovers interpretable differential subgroups in tabular data. Across synthetic benchmarks, medical case studies, model-error analyses, and treatment-effect settings, DiffSub identifies informative subgroups that reveal where population differences arise and why.

Autoregressive Neural Networks based on dense or convolutional layers have recently been shown to be a viable strategy for generating classical spin systems. Unlike these methods, sampling with transformers is commonly considered to be computationally inefficient. In this work, we propose a novel approach to transformer-based neural samplers in which we generate not a single spin per step but groups of spins. As an additional improvement, we construct a model of approximated probabilities, further improving the efficiency of the algorithm. Despite our approach being computationally heavier than dense networks or CNN-based approaches, we were able to sample larger systems of up to $180 \times 180$ spins in case of the Ising model. The Effective Sample Size of our sampler is $\sim 20$ times larger than that of the previous state-of-the-art neural sampler when trained for the $128 \times 128$ Ising model at critical temperature. Finally, we also test our algorithm on the 2D Edwards-Anderson model, where we train $64\times 64$ spin systems.

Preference learning has become the foundation of aligning Large Language Models (LLMs) with human intent. Popular methods, such as Direct Preference Optimization (DPO), minimize surrogate losses as proxies for the intractable pairwise ranking loss. However, we demonstrate that for the equicontinuous hypothesis sets typical of neural networks, these standard surrogates are theoretically inconsistent, yielding vacuous generalization guarantees. To resolve this, we formulate LLM alignment within a margin-shifted ranking framework. We derive rigorous $H$-consistency bounds that depend on enforcing a separation margin $γ$. Crucially, we extend this to Structure-Aware $H$-consistency, introducing a novel objective (SA-DPO) that adapts the margin based on the semantic distance between responses to handle synonyms and hard pairs. Finally, we analyze the trade-off between consistency and model limitations via the Margin-Capacity Profile, proving that heavy-tailed surrogates (such as the Polynomial Hinge family) offer superior consistency guarantees for capacity-bounded models compared to the standard logistic loss used in DPO.

LLMs are increasingly employed both as judges for evaluating open-ended outputs and as co-creation partners in AI-assisted programming; yet rigorous evaluation in human-AI co-creation settings remains underdeveloped as judgments must be reliable, comparable across models, and interpretable over multi-turn interaction. To address this gap, a rubric-driven LLM-as-a-Judge framework is presented for contest-style human-AI co-creation in coding and software engineering (SE). The framework is built around schema-constrained judge outputs, validation and repair mechanisms, grouped and split by user and problem to prevent trajectory leakage, and participant-level NONBLIND context. Multiple LLM judges are assessed through a multi-metric protocol covering discrimination (ROC-AUC, PR-AUC), thresholded decision quality (MCC), probabilistic reliability (LogLoss, Brier score, ECE), and inter-judge agreement (Cohen's and Fleiss' k). Human-AI co-creation is further examined through trajectory-level signals, including turn-wise confidence, Success-at-Turn, time-to-success, revision churn, and CodeBLEU. Co-creation success is found to concentrate early, with Success-at-Turn rising to 0.8533 at the first observed turn and stabilizing at 0.8641 by turn 6. Revision behavior, however, remains heterogeneous, suggesting that productive progress can emerge through either incremental refinement or broader restructuring. On the judging side, the best held-out scores reach 0.5937 for ROC-AUC, 0.6904 for PR-AUC, and 0.5000 for MCC test, while inter-judge consistency remains modest overall (mean pairwise Cohen's k = 0.1592, Fleiss' k = 0.0696). Taken together, this work offers an auditable and reproducible evaluation methodology that links reliability-aware LLM judging with trajectory-based analysis of human-AI co-creation, providing a practical evaluation template for future AI-assisted coding and SE.

A long-standing challenge in economics lies not in the lack of intuition, but in the difficulty of translating intuitive insights into verifiable research. To address this challenge, we introduce AgentEconomist, an end-to-end interactive system designed to translate abstract intuitions into executable computational experiments. Grounded in a domain-specific knowledge base covering over 13,000 high-quality academic papers, the system employs a modular multi-stage architecture. Specifically, the Idea Development Stage generates literature-grounded hypotheses, the Experimental Design Stage configures simulator-aligned experimental parameters and protocols, and the Experimental Execution Stage runs experiments and returns structured analyses. Together, these stages form a human-in-the-loop, iterative workflow that translates economic intuitions into executable computational experiments. Through extensive experiments involving human expert evaluation and large language models (LLMs) as judges, we show that the system generates research ideas with stronger literature grounding and higher novelty and insight than state-of-the-art generic LLMs. Overall, AgentEconomist adopts a human-AI collaboration paradigm that enables researchers to focus on high-level intuitions, while delegating the labor-intensive processes of translation and computational execution to agents.

Medical retrieval-augmented generation (RAG) systems typically operate on text chunks extracted from biomedical literature, discarding the rich visual content (tables, figures, structured layouts) of original document pages. We propose MED-VRAG, an iterative multimodal RAG framework that retrieves and reasons over PMC document page images instead of OCR'd text. The system pairs ColQwen2.5 patch-level page embeddings with a sharded MapReduce LLM filter, scaling to ~350K pages while keeping Stage-1 retrieval under 30 ms via an offline coarse-to-fine index (C=8 centroids per page, ANN over centroids, exact two-way scoring on the top-R shortlist). A vision-language model (VLM) then iteratively refines its query and accumulates evidence in a memory bank across up to 3 reasoning rounds, with a single iteration costing ~15.9 s and the full three-round pipeline ~47.8 s on 4xA100. Across four medical QA benchmarks (MedQA, MedMCQA, PubMedQA, MMLU-Med), MEDVRAG reaches 78.6% average accuracy. Under controlled comparison with the same Qwen2.5-VL-32B backbone, retrieval contributes a +5.8 point gain over the no-retrieval baseline; we also note a +1.8 point edge over MedRAG + GPT-4 (76.8%), with the caveat that this is a cross-paper rather than head-to-head comparison. Ablations isolate +1.0 from page-image vs text-chunk retrieval, +1.5 from iteration, and +1.0 from the memory bank.

Learning algorithms can be significantly improved by routing complex or uncertain inputs to specialized experts, balancing accuracy with computational cost. This approach, known as learning to defer, is essential in domains like natural language generation, medical diagnosis, and computer vision, where an effective deferral can reduce errors at low extra resource consumption. However, the two-stage learning to defer setting, which leverages existing predictors such as a collection of LLMs or other classifiers, often faces challenges due to an expert imbalance problem. This imbalance can lead to suboptimal performance, with deferral algorithms favoring the majority expert. We present a comprehensive study of two-stage learning to defer in expert imbalance settings. We cast the deferral loss optimization as a novel cost-sensitive learning problem over the input-expert domain. We derive new margin-based loss functions and guarantees tailored to this setting, and develop novel algorithms for cost-sensitive learning. Leveraging these results, we design principled deferral algorithms, MILD (Margin-based Imbalanced Learning to Defer), specifically suited for expert imbalance settings. Extensive experiments demonstrate the effectiveness of our approach, showing clear improvements over existing baselines on both image classification and real-world Large Language Model (LLM) routing tasks.

Deploying vision-language models (VLMs) in clinical settings demands auditable behavior under realistic failure conditions, yet the failure landscape of frontier VLMs on specialized medical inputs is poorly characterized. We audit five recent frontier and grounding-aware VLMs (Gemini~2.5~Pro, GPT-5, o3, GLM-4.5V, Qwen~2.5~VL) on Medical VQA along two trust-relevant axes. Perception: all models localize anatomical and pathological targets poorly -- the best model reaches only 0.23 mean IoU and 19.1% Acc@0.5 -- and exhibit clinically dangerous laterality confusion. Pipeline integration: a self-grounding pipeline, where the same model localizes then answers, degrades VQA accuracy for every model -- driven by both inaccurate localization and format-compliance failures under the two-step prompt (parse failure rises to 70%--99% for Gemini and GPT-5 on VQA-RAD). Replacing predicted boxes with ground-truth annotations recovers and improves VQA accuracy, consistent with the failure residing in the perception module rather than in the decomposition itself. These observational findings identify grounding quality as a primary trustworthiness bottleneck in our SLAKE bounding-box setting. As a complementary fine-tuning follow-up, supervised fine-tuning of Qwen~2.5~VL on combined Med-VQA training data attains the highest reported SLAKE open-ended recall (85.5%) among comparable methods, suggesting that the VQA-level gap is tractable with domain adaptation; whether this also closes the perception/trustworthiness bottleneck is left to future work.

How code representation format shapes false positive behaviour in cross-language LLM vulnerability detection remains poorly understood. We systematically vary training intensity and code representation format, comparing raw source text with pruned Abstract Syntax Trees at both training time and inference time, across two 8B-parameter LLMs (Qwen3-8B and Llama 3.1-8B-Instruct) fine-tuned on C/C++ data from the NIST Juliet Test Suite (v1.3) and evaluated on Java (OWASP Benchmark v1.2) and Python (BenchmarkPython v0.1). Cross-language FPR reflects the joint effect of training-time and inference-time representation, not either alone. Text fine-tuning drives FPR upward monotonically (Qwen3-8B: 0.763 zero-shot, 0.866 pilot, 1.000 full-scale) while F1 remains stable (0.637-0.688), masking the collapse. We argue surface-cue memorisation is the primary mechanism: text fine-tuning encodes C/C++-specific API names and syntactic idioms as vulnerability triggers that fire indiscriminately on target-language code. A cross-representation probe, applying text-trained weights to AST-encoded input without retraining, isolates this: Qwen3-8B FPR drops from 0.866 to 0.583, and 37.2% of false positives revert to true negatives under AST input alone. Direct AST fine-tuning does not preserve the benefit (FPR at least 0.970), as flat linearisation introduces structural surface cues of its own. The pattern replicates across both model families. On BenchmarkPython the AST probe yields FPR=0.554, within 2.9 percentage points of the Java result, despite maximal surface-syntax differences, substantially weakening a domain-shift explanation. These findings motivate a pre-deployment consistency gate, running alerts through both text and AST paths, as a retraining-free filter for false-positive-sensitive settings, at the cost of reduced recall.

The risks posed by AI features are increasing as they are rapidly integrated into software applications. In response, regulations and standards for safe and secure AI have been proposed. In this paper, we present an agentic framework that constructs knowledge graphs (KGs) from AI policy documents and retrieves policy-relevant information to answer questions. We build KGs from three AI risk-related polices under two ontology schemas, and then evaluate five LLMs on 42 policy QA tasks spanning six reasoning types, from entity lookup to cross-policy inference, using both heuristic scoring and an LLM-as-judge. KG augmentation improves scores for all five models, and an open, LLM-discovered schema matches or exceeds the formal ontology.

Scene-text image captioning requires fusing three information streams -- visual features, OCR-detected text, and linguistic knowledge -- to generate descriptions that faithfully integrate text visible in images. Existing fusion approaches treat text as language-agnostic, which fails for Vietnamese: a tonal language where diacritics alter word meaning, OCR errors are pervasive, and word boundaries are ambiguous. We argue that Vietnamese scene-text captioning demands \textit{linguistically informed multimodal fusion}, where language-specific structural knowledge is explicitly incorporated into the fusion mechanism. Motivated from these insights, we propose \textbf{HSTFG} (Heterogeneous Scene-Text Fusion Graph), a general-purpose graph fusion framework with learned spatial attention bias, and show through topology analysis that cross-modal graph edges are harmful for scene-text fusion. Building on this finding, we design \textbf{PhonoSTFG} (Phonological Scene-Text Fusion Graph) which specializes graph-level fusion for Vietnamese linguistic reasoning. To support evaluation, we introduce \textbf{ViTextCaps}, the first large-scale Vietnamese scene-text captioning dataset (\textbf{15{,}729} images with \textbf{74{,}970} captions), with comprehensive linguistic analysis showing that 52.8\% of the vocabulary is at risk of diacritic collision.

Current agentic memory systems (vector stores, retrieval-augmented generation, scratchpads, and context-window management) do not implement memory: they implement lookup. We argue that treating lookup as memory is a category error with provable consequences for agent capability, long-term learning, and security. Retrieval generalizes by similarity to stored cases; weight-based memory generalizes by applying abstract rules to inputs never seen before. Conflating the two produces agents that accumulate notes indefinitely without developing expertise, face a provable generalization ceiling on compositionally novel tasks that no increase in context size or retrieval quality can overcome, and are structurally vulnerable to persistent memory poisoning as injected content propagates across all future sessions. Drawing on Complementary Learning Systems theory from neuroscience, we show that biological intelligence solved this problem by pairing fast hippocampal exemplar storage with slow neocortical weight consolidation, and that current AI agents implement only the first half. We formalize these limitations, address four alternative views, and close with a co-existence proposal and a call to action for system builders, benchmark designers, and the memory community.

Current embodied agents are often limited to passive instruction-following or reactive need-satisfaction, lacking a stable, high-order value framework essential for long-term, self-directed behavior and resolving motivational conflicts. We introduce \textit{ValuePlanner}, a hierarchical cognitive architecture that decouples high-level value scheduling from low-level action execution. \textit{ValuePlanner} employs an LLM-based cognitive module to generate symbolic subgoals by reasoning through abstract value trade-offs, which are then translated into executable action plans by a classical PDDL planner. This process is refined via a closed-loop feedback mechanism. Evaluating such autonomy requires methods beyond task-success rates, and we therefore propose a value-centric evaluation suite measuring cumulative value gain, preference alignment, and behavioral diversity. Experiments in the TongSim household environment demonstrate that \textit{ValuePlanner} arbitrates competing values to generate coherent, long-horizon, self-directed behavior absent from instruction-following and needs-driven baselines. Our work offers a structured approach to bridging intrinsic values and grounded behavior for autonomous agents.

Peritoneal metastases are currently assessed using diagnostic laparoscopy to determine Sugarbaker's Peritoneal Cancer Index (sPCI), which works by dividing the abdomen into 13 regions and scoring each region based on tumor size. A recent consensus study defined 3D regions to facilitate a radiological PCI (rPCI), providing standardized anatomical regions for imaging-based assessment. Despite its clinical value, sPCI is invasive and lacks a standardized imaging counterpart. In this study, we propose a deep learning-based approach to automatically segment the rPCI regions on CT. We evaluate nnU-Net and Swin UNETR on 62 CT scans with rPCI regions manually annotated by three clinical researchers and validated by two expert radiologists. Performance was assessed using five-fold cross-validation with the Dice Similarity Coefficient (Dice), 95th percentile Hausdorff distance and Average Surface Distance. nnU-Net achieved an overall Dice of 0.82, approaching interobserver agreement (0.88) and outperforming Swin UNETR (0.76), with remaining challenges primarily in right flank and small-bowel regions. These results demonstrate feasibility of automated rPCI segmentation, laying the foundation for non-invasive, imaging-based assessment.

Long-term conversational memory requires retrieving evidence scattered across multiple sessions, yet single-pass retrieval fails on temporal and multi-hop questions. Existing iterative methods refine queries via generated content or document-level signals, but none explicitly diagnoses the evidence gap, namely what is missing from the accumulated retrieval set, leaving query refinement untargeted. We present EviMem, combining IRIS (Iterative Retrieval via Insufficiency Signals), a closed-loop framework that detects evidence gaps through sufficiency evaluation, diagnoses what is missing, and drives targeted query refinement, with LaceMem (Layered Architecture for Conversational Evidence Memory), a coarse-to-fine memory hierarchy supporting fine-grained gap diagnosis. On LoCoMo, EviMem improves Judge Accuracy over MIRIX on temporal (73.3% to 81.6%) and multi-hop (65.9% to 85.2%) questions at 4.5x lower latency. Code: https://github.com/AIGeeksGroup/EviMem.

Template engines are indispensable components in modern software ecosystems, enabling the generation of structured documents and scripts across domains such as web development, Infrastructure as Code, and data engineering. However, the unique architectural characteristics of template engine-based applications (i.e., TE applications), including multi-language composition, opaque data flow, deferred validation, and complex integration, pose significant challenges for diagnosing and resolving bugs in TE applications. While prior research has primarily focused on template engine security, bugs in TE applications remain under-investigated. To bridge this gap, we present the first comprehensive study of TE application bugs. By analyzing 1,004 application bugs across 15 template engines in five programming languages, we identify the symptoms and root causes of TE application bugs and common patterns to fix them. Our findings reveal that Abnormal Rendering Result (e.g., unexpected or blank output) is the most prevalent symptom (48.61%), often manifesting as silent failures that are difficult to diagnose. We identify 17 root causes, with Syntax Misuse, Mismatched Data Context, and Incompatible Integration as the dominant categories. Furthermore, we find that while 67.92% of the bugs are fixed within the template, over 20% require modifications in the host-side logic to resolve data context issues. Based on these findings, we derive actionable implications for tool designers, practitioners, and researchers. To demonstrate the practical utility of our findings, we further develop two prototype tools for the Jinja engine to facilitate the development and debugging of TE applications.

Across millennia, complex societies have faced the same coordination problem of how to organize collective action among cognitively bounded and informationally incomplete individuals. Different civilizations developed different political institutions to answer the same basic questions of who proposes, who reviews, who executes, and how errors are corrected. We argue that multi-agent systems built on large language models face the same challenge. Their central problem is not only individual intelligence, but collective organization. Historical institutions therefore provide a structured design space for multi-agent architectures, making key trade-offs between efficiency and error correction, centralization and distribution, and specialization and redundancy empirically testable. We translate seven historical political institutions, spanning four canonical governance patterns, into executable multi-agent architectures and evaluate them under identical conditions across three large language models and two benchmarks. We find that governance topology strongly shapes collective performance. Within a single model, the gap between the best and worst institution exceeds 57 percentage points, while the optimal architecture shifts systematically with model capability and task characteristics. These results suggest that collective intelligence will not advance through a single optimal organizational form, but through governance mechanisms that can be reselected and reconfigured as tasks and capabilities evolve. More broadly, this points to a transition from \textbf{self-evolving agents} to the \textbf{self-evolving multi-agent system}. The code is available on \href{https://github.com/cf3i/SocialSystemArena}{GitHub}.

While machine-learned interatomic potentials (MLIPs) accelerate phonon dispersion calculations, merely identifying dynamical instabilities in computationally predicted materials is insufficient; automated pathways to resolve them are required. We introduce VibroML, an open-source Python toolkit driven by foundational MLIPs that shifts the paradigm from stability verification to automated structural remediation. VibroML employs an energy-guided genetic algorithm that vastly outperforms traditional soft-mode following, efficiently navigating the potential energy surface to uncover diverse, dynamically stable polymorphs. As 0 K harmonic stability does not guarantee macroscopic viability, an automated molecular dynamics workflow evaluates finite-temperature structural retention. VibroML also couples with ProtoCSP, our combinatorial structure prediction engine, to stabilize frustrated crystal topologies via targeted alloying, successfully rescuing functional perovskite networks like Cs$_2$KInI$_6$ and KTaSe$_3$. Demonstrating broader applicability, we mined the Alexandria database -- where ~50% of quaternary and 99.5% of quinary elemental combinations lack any structural entries -- to identify thousands of abandoned, high-symmetry stoichiometries. Deploying ProtoCSP's "cold start" retrieval and VibroML's evolutionary search on a sample, we successfully identified dynamically stable low-symmetry candidates. Through integrated structural remediation, thermal validation, and systematic compositional exploration, VibroML enables a comprehensive deep-screening approach, yielding physically sound structural propositions that far surpass standard high-throughput workflows.

Constructing and curating high-quality code datasets requires significant resources, making them valuable intellectual property. Unfortunately, these datasets currently face severe risks of unauthorized use. Although digital watermarking offers a post hoc mechanism for copyright authentication, existing methods are predominantly based on the co-occurrence pattern, which is not robust and is susceptible to watermark detection and removal attacks. In this paper, we propose PuzzleMark, a robust watermarking method for code datasets. To reduce the risk of watermark exposure, PuzzleMark introduces a carrier selection strategy that leverages code complexity to evaluate the suitability of code snippets as watermark carriers, and selects those with high suitability for watermarking. To enhance the robustness of the watermark, PuzzleMark proposes a novel concatenation pattern to replace the traditional co-occurrence pattern, and implements two watermarking strategies through variable name concatenation. PuzzleMark adaptively embeds watermarks based on the inherent characteristics of the code, making it more stealthy while maintaining design simplicity. For watermark verification, PuzzleMark employs Fisher's exact test to verify suspicious models under a black-box setting. Experimental results demonstrate that PuzzleMark achieves a 100% verification success rate and a 0% false positive rate, with negligible impact on model performance. Both our human study and our evaluation using four state-of-the-art watermark detection methods show that PuzzleMark exhibits strong imperceptibility, with an average suspicious rate $\leq$ 0.24 and an average recall $\leq$ 30.41%, respectively. As a practical digital watermarking method, PuzzleMark provides strong protection for the intellectual property of code datasets and offers new insights for future research.

The hubness problem, in which hub embeddings are close to many unrelated examples, occurs often in high-dimensional embedding spaces and may pose a practical threat for purposes such as information retrieval and automatic evaluation metrics. In particular, since cross-modal similarity between text and images cannot be calculated by direct comparisons, such as string matching, cross-modal encoders that project different modalities into a shared space are helpful for various cross-modal applications, and thus, the existence of hubs may pose practical threats. To reveal the vulnerabilities of cross-modal encoders, we propose a method for identifying the hub embedding and its corresponding hub text. Experiments on image captioning evaluation in MSCOCO and nocaps along with image-to-text retrieval tasks in MSCOCO and Flickr30k showed that our method can identify a single hub text that unreasonably achieves comparable or higher similarity scores than human-written reference captions in many images, thereby revealing the vulnerabilities in cross-modal encoders.

We introduce Target-Event-Agent Networks (TEA Nets) as a computational framework to extract subjects (``Agents"), verbs (``Events"), and objects (``Targets") from texts. Grounded in cognitive network science and artificial intelligence, TEA Nets are implemented as an open-source Python library. We test TEA Nets in three case studies, demonstrating the framework's ability to perform interpretable emotion detection, semantic frame analyses, and linguistic inquiries across conspiracy texts and textual responses generated by LLMs. In the LOCO conspiracy corpus, TEA Nets revealed that highly conspiratorial narratives (4,227 texts) linked personal pronouns (``I", ``you", ``we") with the same actions twice as frequently as low-similarity conspiracy narratives. High-conspiracy narratives connected person-focused elements (``you", ``people") through actions eliciting anger above the random baseline ($z = 2.63, p < .05$), a trend absent in low-similarity conspiracy narratives, which emphasized scientific actors (``researcher", ``scientist"). In the HOPE and CounseLLMe datasets of 212 (human) and 200 (LLM-based) psychotherapy transcripts, respectively, TEA Nets highlighted emotional differences. When expressing feelings, Claude 3 Haiku, GPT-3.5, and humans used sad words with higher frequency than random expectations but Haiku expressed sadness with lower emotional intensity than humans ($U = 1243.5, p = .036$). We discuss these differences in the context of psychotherapy training on LLM-simulated patients. Our results show that Target-Event-Agent Networks can extract relevant emotional, syntactic, and semantic insights from narratives, opening new avenues for text analysis with cognitive network science.

The incorporation of fairness into the distribution network (DN) planning and operation has become a key goal of recent studies. The cost of implementing fairness, denominated the price of fairness (PoF), covers the efficiency that is renounced for attaining social cohesion through fair outcomes. Locational disparity makes fairness schemes emerge to level the consumers playing field. However, fairness encompasses a range of notions. From egalitarian to merit-based criteria, various metrics are implemented as a tool for measuring equitable utility distribution. These have different mathematical complexities, from linear to non-linear programming cases, which affect their overall applicability. Hence, this study compiles the overarching fairness notions and metrics, reviewing how these affect stakeholders and the inherent mathematical optimisation in resource allocation problems. The aim is to support consistent and transparent planning and decision-making within DN operations.

Policy optimization in high-dimensional continuous control for robotics remains a challenging problem. Predominant methods are inherently local and often require extensive tuning and carefully chosen initial guesses for good performance, whereas more global and less initialization-sensitive search methods typically incur high rollout costs. We propose TFM-S3, a tabular hybrid local-global method for improving global exploration in robot policy learning with limited rollout cost. We interleave high-frequency local updates with intermittent rounds of global search. In each search round, we construct a dynamically updated low-dimensional policy subspace via SVD and perform iterative surrogate-guided refinement within this space. A pretrained tabular foundation model predicts candidate returns from a small context set, enabling large-scale screening with limited rollout cost. Experiments on continuous control benchmarks show that TFM-S3 consistently accelerates early-stage convergence and improves final performance compared to TD3 and population-based baselines under an identical rollout budget. These results demonstrate that foundation models are a powerful new tool for creating sample-efficient policy learning methods for continuous control in robotics.

Detecting language ideologies is a valuable yet complex task for understanding how identities are constructed through discourse. In Luxembourg's multicultural and multilingual society, language ideologies reflect more than simple preferences: they carry deep cultural and social meanings, shaping identities and social belonging. Following recent developments in applying Natural Language Processing tools to linguistics and social science, this paper explores the potential of large language models to assist in the detection of language ideologies. We manually annotate a corpus of user comments in Luxembourgish with predefined ideological categories and then evaluate the performance of large language models under varying prompt conditions to assess their ability to replicate these human annotations. Since Luxembourgish is a small language and poorly represented in the LLMs' training data, we also investigate whether machine-translating the data to high-resource languages increases performance on the ideology detection task. Our findings suggest that, while LLMs are not yet fully optimized for a multi-class ideological annotation task, they are practical tools to identify language ideological content.

Many real-world tasks require language models (LMs) to reason over complex contexts that exceed their parametric knowledge. This calls for context learning, where LMs directly learn relevant knowledge from the given context. An intuitive solution is inference-time skill augmentation: extracting the rules and procedures from context into natural-language skills. However, constructing such skills for context learning scenarios faces two challenges: the prohibitive cost of manual skill annotation for long, technically dense contexts, and the lack of external feedback for automated skill construction, since there is no automatic signal to tell whether a proposed skill is helpful. In this paper, we propose Ctx2Skill, a self-evolving framework that autonomously discovers, refines, and selects context-specific skills without human supervision or external feedback. At its core, a multi-agent self-play loop has a Challenger that generates probing tasks and rubrics, a Reasoner that attempts to solve them guided by an evolving skill set, and a neutral Judge that provides binary feedback. Crucially, both the Challenger and the Reasoner evolve through accumulated skills: dedicated Proposer and Generator agents analyze failure cases and synthesize them into targeted skill updates for both sides, enabling automated skill discovery and refinement. To prevent adversarial collapse caused by increasingly extreme task generation and over-specialized skill accumulation, we further introduce a Cross-time Replay mechanism that identifies the skill set achieving the best balance across representative cases for the Reasoner side, ensuring robust and generalizable skill evolution. The resulting skills can be plugged into any language model to obtain better context learning capability. Evaluated on four context learning tasks from CL-bench, Ctx2Skill consistently improves solving rates across backbone models.

Due to regulatory compliance and governance management, modern (permissioned) blockchains require flexible endorsement, which allows the endorsement policy for each contract or state object to be individually defined. To enable flexible endorsement, Hyperledger Fabric employs an execute-order-validate (EOV) paradigm, in which transactions first undergo speculative execution and endorsement, and are only then ordered and validated. Meanwhile, most blockchain systems, including the platform targeted in this work (i.e., ChainMaker), still follow a conflict-free order-execute framework. We argue that the EOV paradigm still faces several limitations, notably high abort rates in high-contention workloads such as those in Decentralized Finance (DeFi). To avoid refactoring our system and better suit DeFi applications, we try to integrate flexible endorsement into the classical order-execute architecture and accordingly propose a new framework. The key challenge is to deterministically remove problematic transactions from an ordered list, while preserving censorship resistance and decentralization for the remaining ones. We instantiate this framework on top of Tendermint, a seminal Byzantine fault-tolerant (BFT) protocol adopted in our system, and thereby propose FlexTender. By elegantly embedding endorsements into consensus, FlexTender incurs no additional messaging overhead in the normal case. Empirical evaluation using an Ethereum USDT workload demonstrates that FlexTender achieves up to $10.6\times$ speedup in throughput over an EOV simulation on the same platform.

To preserve previously learned representations, continual learning systems must strike a balance between plasticity, the ability to acquire new knowledge, and stability. This stability-plasticity dilemma affects how representations can be reused across tasks: shared structure enables transfer when tasks are similar but may also induce interference when new learning disrupts existing representations. However, it remains unclear when and why structural separation influences this trade-off. In this study, we examine how network architecture, task similarity, and representational dimensionality jointly shape learning in a sequential task paradigm inspired by transfer-interference studies. We compare a task-partitioned modular recurrent network with a single-module baseline by systematically varying task similarity (low, medium, high) and the scale of weight initialization, which induces different learning regimes that we empirically characterize through the effective dimensionality of the learned representations. We find that architecture has minimal impact in high-dimensional regimes where representations are sufficiently unconstrained to accommodate multiple tasks without strong interference. In contrast, in lower-dimensional (rich) regimes, architectural separation is decisive: modular networks exhibit graded alignment of task-specific subspaces with overlap for similar tasks, partial orthogonalization for moderately dissimilar tasks, and stronger separation for dissimilar tasks. This graded geometry is absent in the single network baseline. Our findings suggest that representational dimensionality acts as a key organizing variable governing when structural separation becomes functionally relevant, and highlight adaptive geometry as a central principle for designing continual learning systems.

Machine learning (ML)-based API recommendation helps developers efficiently identify suitable APIs to complement the application code. However, code datasets used to train ML models often exhibit a long-tail distribution, leading to unreliable API recommendations, especially for infrequently used API methods at the tail of the distribution. To address this issue, we propose N-version API Recommendation (NvRec), which leverages N different versions of ML models to enhance the reliability of API sequence recommendations by suppressing unreliable outputs entailing tail APIs. NvRec leverages a set of available ML models and profiles their performance on individual API methods with their tail properties. The generated model profile is used at inference time to filter out unreliable API recommendations and determine the final output. We implement NvRec using five API recommendation models, including CodeBERT, CodeT5, MulaRec, UniXcoder, and CodeT5+, and evaluate it on a public benchmark dataset constructed from compilable Java projects. For the three-version NvRec, we find that the combination of CodeT5, MulaRec, and UniXcoder achieves the highest true accept rate of 83.8%, with a rejection rate of 80.7%, when majority voting is restricted to highly reliable candidates. In contrast, the five-version configuration achieves its highest true accept rate of 83.1% with simple majority voting, while reducing the rejection rate to 69.0%. Overall, the five-version configuration offers a better balance between true accept rate and rejection rate.

We propose a paradigm shift from learning to answer to learning to question: can a language model generate verifiable problems, solve them, and turn the resulting feedback into self-improvement without human supervision? We introduce ANCORA, an anchored-curriculum framework in which a unified policy alternates between a Proposer that synthesizes novel specifications and a Solver that produces verified solutions. ANCORA rests on three load-bearing mechanisms: a two-level group-relative update that couples Proposer advantages across specifications with Solver advantages across solution attempts; iterative self-distilled SFT that projects the base model onto its valid-output manifold before RL; and a UCB-guided Curriculum DAG that grows only through strictly filtered, novel, Solver-verified specifications. These stabilizers are necessary because sparse verifier feedback otherwise drives Proposer collapse even under MLRL-aligned rewards. Instantiated in Verus, ANCORA lifts Dafny2Verus pass@1 from a 26.6% SFT baseline to 81.5% in the test-time-training setting under 0-shot evaluation, outperforming the PSV self-play baseline by 15.8 points despite PSV using 1-shot inference; in a separate transfer setting, training from Dafny2Verus seeds yields 36.2% and 17.2% pass@1 on held-out MBPP and HumanEval.

Integrated Circuit (IC) verification consumes nearly 70% of the IC development cycle, and recent research leverages Large Language Models (LLMs) to automatically generate testbenches and reduce verification overhead. However, LLMs have difficulty generating testbenches correctly. Unlike high-level programming languages, Hardware Description Languages (HDLs) are extremely rare in LLMs training data, leading LLMs to produce incorrect code. To overcome challenges when using LLMs to generate Universal Verification Methodology (UVM) testbenches and sequences, wepropose HAVEN (Hybrid Automated Verification ENgine) to prevent LLMs from writing HDL directly. For UVM testbench generation, HAVEN utilizes LLM agents to analyze design specifications to produce a structured architectural plan. The HAVEN Template Engine then combines with predefined and protocol-specific templates to generate all UVM components with correct bus-handshake timing. For UVM sequence generation, HAVEN introduces a Protocol-Aware Sequence Domain-Specific Language (DSL) that decomposes sequences into fine-grained step types. A set of predefined DSL patterns first establishes sequences that achieve a high coverage rate without LLM involvement. HAVEN continues to improve the coverage rate by iteratively leveraging LLM agents to analyze coverage gap reports and compose additional targeted DSL sequences. Unlike previous works, HAVEN is the first system that utilizes pre-defined, protocol-specific Jinja2 templates to generate all UVM components and UVM sequences using our proposed Protocol-Aware DSL and rule-based code generator. Our experimental results on 19 open-source IP designs spanning three interface protocols (Direct, Wishbone, AXI4-Lite) show that HAVEN achieves 100% compilation success, 90.6% code coverage, and 87.9% functional coverage on average, and is SOTA among LLM-assisted testbench generation systems.

Context: Many organizations are keen to incorporate generative~AI (GenAI) into their software development processes. Technology acceptance models, such as the Unified Theory of Acceptance and Use of Technology (UTAUT), are traditionally used to identify individual-level barriers to the acceptance of new technologies and can facilitate the transition to GenAI. However, UTAUT has seen limited use within software engineering (SE) research. Objective: Using UTAUT as an example, to identify key areas for future research on GenAI acceptance, including the role of Bayesian approaches for data analysis. Method: We review foundational and SE-specific literature on UTAUT and analyze its emerging applications for GenAI in SE. Results: We identify three priorities for future research: (1) identifying and refining constructs to account for GenAI's nature and transformational impact; (2) improving operationalization practices to strengthen construct validity and cross-study comparability; and (3) incorporating Bayesian analysis to support small-sample inference by integrating prior knowledge, iterative model updating, and simulation of scenarios. Conclusion: UTAUT is a suitable candidate to combine with Bayesian analysis for practical insights on individual-level barriers to GenAI use in SE, but additional theories should be considered.

Machine learning continues to emerge as an important tool to be utilised within structural engineering and structural health monitoring, due to its ability to accurately and quickly perform both regression and classification tasks. However, a purely data driven approach has its limitations, particularly where we lack data from relevant environmental and operational conditions, a situation that has led to the development of physics-informed machine learners for structural health monitoring. These "grey-box" models take into account the physical insight that an engineer would have about the structure they are modelling and have shown promising results in the structural engineering field among many others. This work compares black and grey-box models through a "green" lens, comparing them in terms of their environmental impact, and investigating how the high extrapolative performance of grey-box models can reduce their runtimes and therefore carbon emissions. The authors aim to develop physics-informed models with reduced computational costs, while maintaining high performance, illustrated through a structural health monitoring case study.

Current Large Language Model (LLM) evaluation frameworks utilize the same static prompt template across all models under evaluation. This differs from the common industry practice of using prompt optimization (PO) techniques to optimize the prompt for each model to maximize application performance. In this paper, we investigate the effect of PO towards LLM evaluations. Our results on public academic and internal industry benchmarks show that PO greatly affects the final ranking of models. This highlights the importance of practitioners performing PO per model when conducting evaluations to choose the best model for a given task.

Predicting stable and metastable structures is central to molecular and materials discovery, but remains limited by the cost of searching high-dimensional energy landscapes. Deep generative models offer efficient structure sampling, yet their outputs remain shaped by training data and can underexplore minima that are rare but physically relevant. We introduce generative structure search (GSS), a unified framework that formulates diffusion-based generation and random structure search (RSS) as limiting regimes of a common sampling process driven by learned score fields and physical forces. Coupling these drivers lets GSS use data priors to accelerate sampling while retaining energy-guided exploration of local minima. Across molecular and crystalline systems, GSS recovers diverse metastable structures with more than tenfold lower sampling cost than RSS for broad coverage and remains effective for compositions outside the training distribution. The results establish a physically grounded generative search strategy for discovering structures beyond the reach of data-driven sampling alone.

Large language models (LLMs) are commonly evaluated for political bias based on their responses to fixed questionnaires, which typically place frontier models on the political left. A parallel literature shows that LLMs are sycophantic: they adapt their answers to the views, identities, and expectations of the user. We show that these findings are linked: standard political-bias audits partly capture sycophantic accommodation to the inferred auditor. We employ a factorial experiment across three major audit instruments--the Political Compass Test, the Pew Political Typology, and 1,540 partisan-benchmarked Pew American Trends Panel items--administered to six frontier LLMs while varying only the asker's stated identity (N = 30,990 responses). At baseline, all six models lean left. When the asker identifies as a conservative Republican, responses shift sharply: the share of items closer to Democrats falls by 28-62 percentage points, and all six models move right of center. A mirror-image progressive-Democrat cue produces little change; rightward accommodation is 8.0$\times$ larger than leftward. When asked who the default asker is, models identify an auditor, researcher, or academic; when asked what answer that asker expects, they select the Democrat-coded option 75% of the time, nearly the rate under an explicit progressive cue. These patterns are inconsistent with a purely fixed model ideology and indicate that single-prompt audits capture an interaction between model and inferred interlocutor. Political bias in LLMs is therefore not a fixed point on an ideological scale but a response profile that must be mapped across realistic interlocutors.

We present WaferSAGE, a framework for wafer defect visual question answering using small vision-language models. To address data scarcity in semiconductor manufacturing, we propose a three-stage synthesis pipeline incorporating structured rubric generation for precise evaluation. Starting from limited labeled wafer maps, we employ clustering-based cleaning to filter label noise, then generate comprehensive defect descriptions using vision-language models, which are converted into structured evaluation rubrics criteria. These rubrics guide the synthesis of VQA pairs, ensuring coverage across defect type identification, spatial distribution, morphology, and root cause analysis. Our dual assessment framework aligns rule-based metrics with LLM-Judge scores via Bayesian optimization, enabling reliable automated evaluation. Through curriculum-based reinforcement learning with Group Sequence Policy Optimization (GSPO) and rubric-aligned rewards, our 4B-parameter Qwen3-VL model achieves a 6.493 LLM-Judge score, closely approaching Gemini-3-Flash (7.149) while enabling complete on-premise deployment. We demonstrate that small models with domain-specific training can surpass proprietary large models in specialized industrial visual understanding, offering a viable path for privacy-preserving, cost-effective deployment in semiconductor manufacturing.

Large Language Models (LLMs) can strongly shape social discourse, yet datasets investigating how LLM outputs vary across controlled social and contextual prompting remain sparse. Cognitive Digital Shadows (CDS) is a 190,000-record synthetic corpus supporting analyses of LLM-generated discourse. Each CDS record is generated by one of 19 LLMs, prompted to shadow either a human persona or an AI-assistant role. CDS contains LLM responses on 4 controversial societal topics: vaccines/healthcare, social media disinformation, the gender gap in science, and STEM stereotypes. Persona-conditioned records encode 17 sociodemographic and psychological attributes, providing data linking LLMs' prompts, language, stances and reasoning. Texts are validated for topic anchoring and can support emotional analyses via interpretable NLP (e.g. textual forma mentis networks). CDS is enriched by a pooling platform with user-friendly dashboards, enabling easy, interactive group-level comparisons of emotional and semantic framing across personas, topics and models. The CDS prompting framework supports future audits of LLMs' bias, social sensitivity and alignment.

To enhance LLMs' impact on math education, we need data on their mathematical prowess and biases across prompts. To fill this gap, we introduce MEDS (Math Education Digital Shadows) as a dataset mapping how large language models reason about and report mathematics across human- and AI-like conditions. MEDS involves 28,000 personas from 14 LLMs (from families like Mistral, Qwen, DeepSeek, Granite, Phi and Grok) shadowing either humans or AI assistants. Each record/shadow includes a set of prompts along with psychological/sociodemographic persona metadata and four types of math tasks: (i) open math interview, (ii) three psychometric tests about math perceptions with explanations, (iii) cognitive networks capturing math attitudes, and (iv) 18 high-school math test questions together with their reasoning and confidence scores. MEDS differs from traditional score-only math benchmarks because it integrates concepts of self-efficacy, math anxiety, and cognitive network science besides math proficiency scores. Data validation shows that the sampled LLMs exhibit schema integrity and consistent personas, together with family-specific peculiarities like human-like negative math attitudes, logical fallacies, and math overconfidence. MEDS will benefit learning analytics experts, cognitive scientists, and developers of safer AI tutors in mathematics.

With the widespread application of Unmanned Aerial Vehicles (UAVs) in bridge structural health monitoring, deep learning-based automatic crack detection has become a major research focus. However, practical UAV inspections still face four key challenges: weak crack features, degraded imaging conditions, severe class imbalance, and limited computational resources for practical UAV inspection workflows. To address these issues, this paper proposes a unified lightweight convolutional neural network framework composed of four synergistic components: a lightweight backbone network, a Convolutional Block Attention Module (CBAM) for channel and spatial enhancement, a directed robust augmentation strategy based on inspection-scene priors, and Focal Loss for hard-sample learning under class imbalance. Experiments on the SDNET2018 bridge deck dataset show that the proposed method achieves an inference speed of 825 FPS with only 11.21M parameters and 1.82G FLOPs. Compared with the baseline model, the complete framework improves the F1-score by 2.51% and recall by 3.95%. In addition, Grad-CAM visualizations indicate that the introduced attention module shifts the model's focus from scattered regions to precise tracking along crack trajectories. Overall, this study achieves a strong balance among accuracy, speed, and robustness, providing a practical solution for ground-station assisted real-time deployment in UAV bridge inspections. The source code is available at: https://github.com/skylynf/AttXNet .

People commonly leverage structured content to accelerate knowledge acquisition and research problem solving. Among these, roadmaps guide researchers through hierarchical subtasks to solve complex research problems step by step. Despite progress in structured content generation, the roadmap generation task has remained unexplored. To bridge this gap, we introduce RoadMap, a novel benchmark designed to evaluate the ability of large language models (LLMs) to construct high-quality roadmaps for solving complex research problems. Based on this, we identify three limitations of LLMs: (1) lack of professional knowledge, (2) unreasonable task decomposition, and (3) disordered logical relationships. To address these challenges, we propose RoadMapper, an LLM-based multi-agent system that decomposes the research roadmap generation task into three key stages (i.e., initial generation, knowledge augmentation, and iterative "critique-revise-evaluate"). Extensive experiments demonstrate that RoadMapper can improve LLMs' ability for roadmap generation, while enhancing average performance by more than 8% and saving 84% of the time required by human experts, highlighting its effectiveness and application potential.

Amorphous (disordered) materials are solids that have shown great potential in various domains, including energy storage, thermal management, and advanced materials. Unlike crystalline materials that can be described by unit cells containing a few to hundreds of atoms, amorphous materials require larger simulation cells with at least hundreds to thousands of atoms. To advance the design of amorphous materials with desired properties and facilitate the exploration of their vast design space, generative inverse design has emerged as a promising approach. It aims to directly output materials with properties closely aligned with the desired ones using probabilistic generative models conditioned on desired properties, which can be more resource efficient than the traditional trial-and-error approach. However, due to the inherent stochasticity of probabilistic generative models, when element assignments are unconstrained, a large portion of generated materials may be charge unbalanced, and no existing methods can effectively mitigate this limitation. In this work, we propose AMGenC, a new generative inverse design method for amorphous materials that can guarantee the generation of charge balanced samples, with minimal additional computational overhead and without sacrificing inverse design accuracy. AMGenC achieves this through an element noise that gives the generation process a starting point centered around charge balance, and the combination of a per-step soft projection and a final discrete projection for steering the elements toward exact charge balance throughout the generation. We perform extensive experiments on two amorphous materials datasets. Experimental results provide evidence that AMGenC achieves its design goal.

We present JaiTTS-v1.0, a state-of-the-art Thai voice cloning text-to-speech model built through continual training on a large Thai-centric speech corpus. The model architecture is adapted from VoxCPM, a tokenizer-free autoregressive TTS model. JaiTTS-v1.0 directly processes numerals and Thai-English code-switching, which is very common in realistic settings, without explicit text normalization. We test the models on short-duration speech generation and long-duration speech generation, which reflects many real-world use cases. JaiTTS-v1.0 achieves a state-of-the-art CER of 1.94\%, surpassing the human ground truth of 1.98% for short-duration tasks while performing on par with human ground truth for long-duration tasks. In human judgment evaluations, our model wins 283 of 400 pairwise comparisons against commercial flagships, with only 58 losses.

Learning informative representations from tabular data in remote sensing and environmental science is challenging due to heterogeneity, scarce labels, and redundancy among features. We present ZAYAN (Zero-Anchor dYnamic feAture eNcoding), a self-supervised, feature-centric contrastive framework for tabular data. ZAYAN performs contrastive learning at the feature rather than sample level, removing the need for explicit anchor selection and any reliance on class labels, while encouraging a redundancy-minimized, disentangled embedding space. The framework has two modules: ZAYAN-CL, which pretrains feature embeddings via a zero-anchor contrastive objective with dynamic perturbations and masking, and ZAYAN-T, a Transformer that conditions on these embeddings for downstream classification. Across eight datasets, including six remote-sensing tabular benchmarks and two remote-sensing-driven flood-prediction tables from satellite and GIS products, ZAYAN achieves superior accuracy, robustness, and generalization over tabular deep learning baselines, with consistent gains under label scarcity and distribution shift. These results indicate that feature-level contrastive learning and dynamic feature encoding provide an effective recipe for learning from tabular sensing data.

Large language models (LLMs) are increasingly used for recommendation reranking, but their listwise predictions can depend on the order in which candidates are presented. This creates a mismatch between the set-based nature of recommendation and the sequence-based computation of decoder-only LLMs, where permuting an otherwise identical candidate set can change item scores and final rankings. Such order sensitivity makes LLM-based rerankers difficult to rely on, since rankings may reflect prompt serialization rather than user preference. We propose InvariRank, a permutation-invariant listwise reranking framework that addresses this dependence at the architectural level. InvariRank blocks cross-candidate attention with a structured attention mask and negates position-induced scoring changes through shared positional framing under Rotary Positional Embeddings (RoPE). Combined with a listwise learning-to-rank objective, the model scores all candidates in a single forward pass, avoiding permutation-based invariance training objectives that require multiple permutations of a candidate set. Experiments on recommendation benchmarks show that InvariRank maintains competitive ranking effectiveness while producing stable rankings across candidate permutations. The results suggest that architectural invariance is a practical route to reliable and efficient LLM-based recommendation reranking. The source code is at https://github.com/ejbito/InvariRank.

Protecting sensitive health data while enabling collaborative analysis is a central challenge in healthcare. Traditional machine learning (ML) requires institutions to pool anonymized patient records, centralizing analytical development and privacy risks at a single site. Privacy-enhancing technologies (PETs), including Differential Privacy (DP) and Homomorphic Encryption (HE), can mitigate these risks. However, they are mainly studied in conventional data-sharing settings and often introduce trade-offs, including reduced model utility, higher computational cost, and increased implementation complexity. Federated Learning (FL) reduces data centralization by enabling institutions to train models locally and share only model updates. Nevertheless, FL does not eliminate privacy risks, as shared parameters or gradients may still reveal sensitive information. Integrating DP or HE into FL can strengthen privacy guarantees, yet their comparative performance and deployment implications in real-world healthcare settings remain unclear. We systematically evaluated DP and HE integration in FL under real-world conditions, comparing them with standard FL and centralized ML (cML) to quantify privacy-utility trade-offs in multi-institutional settings. Using nationwide Swedish healthcare data, we evaluated cardiovascular disease risk prediction using logistic regression (LR) and neural network (NN) learners. FL with HE achieved performance comparable to cML but introduced measurable cryptographic overhead, particularly in the NN implementation. FL with DP incurred lower computational cost; however, LR was more sensitive to calibrated noise than the NN, resulting in greater performance degradation. Our findings provide practical guidance for deploying privacy-preserving FL in fragmented healthcare systems.

Video moment retrieval is the task of retrieving specific segments of a video corresponding to a given text query. Recent studies have been conducted to improve multimodal alignment performance through visual-linguistic similarity learning at the snippet-level and transformer-based temporal boundary regression. However, existing models do not calculate similarity by considering the relationships between multiple answer segments that match the query. Therefore, existing models are easily influenced by visually similar segments in the surrounding context. Existing models calculate similarity at the snippet-level and ignore the relationships between multiple answer segments corresponding to a single query. Therefore, they struggle to exclude segments irrelevant to the query. To address this issues, we propose ClipTBP, a clip-pair temporal boundary prediction framework based on boundary-aware learning. ClipTBP introduces a clip-level alignment loss for explicitly learning the semantic relationship between answer segments. ClipTBP also predicts accurate temporal boundaries by applying both main boundary loss and auxiliary boundary loss. ClipTBP consistently improves performance when applied to various existing models and demonstrates more robust boundary prediction performance even in ambiguous query scenarios.

Reasoning over heterogeneous artifacts (PDFs, spreadsheets, slide decks, etc.) increasingly occurs within structured agent workflows that iteratively extract, transform, and reference external information. In these workflows, uncertainty is not merely an input-quality issue: it can redirect decomposition and routing decisions, reshape intermediate state, and produce qualitatively different execution trajectories. We study this phenomenon by treating uncertainty as a controlled variable: we inject structured perturbations into artifact-derived representations, execute fixed workflows under comprehensive logging, and quantify contamination via trace divergence in plans, tool invocations, and intermediate state. Across 614 paired runs on 32 GAIA tasks with three different language models, we find a decoupling: workflows may diverge substantially yet recover correct answers, or remain structurally similar while producing incorrect outputs. We characterize three manifestation types: silent semantic corruption, behavioral detours with recovery, and combined structural disruption and their control-flow signatures (rerouting, extended execution, early termination). We measure operational costs and characterize why commonly used verification guardrails fail to intercept contamination. We contribute (i) a formal taxonomy of contamination manifestations in structured workflows, (ii) a trace-based measurement framework for detecting and localizing contamination across agent interactions, and (iii) empirical evidence with implications for targeted verification, defensive design, and cost control.

We present BAss (BDD-based ADF symbolic solver), a novel analysis tool for Abstract Dialectical Frameworks (ADFs) based on Binary Decision Diagrams (BDDs). It supports the fully symbolic computation of all admissible, complete, and preferred interpretations, as well as two-valued and stable models of an ADFs. Our approach is inspired by the recently discovered equivalence between Boolean Networks (BNs) and ADFs by Heyninck et al. (2024) and Azpeitia et al. (2024), significantly extending current BDD-based tools bioLQM, AEON, and adf-bdd. We conducted experiments on a large-scale collection of real-world models from both the BN and ADF communities. Our results show that BAss dramatically outperforms previous BDD-based tools and is competitive (even significantly better in some cases) with state-of-the-art SAT/ASP-based methods, particularly in scenarios involving large solution spaces. Notably, BAss is able to enumerate all fixed points or minimal trap spaces of certain biological networks beyond the reach of existing tools, thereby enabling new analysis and case studies in systems biology. These results highlight the practical relevance of symbolic reasoning for complex real-world applications, particularly in systems biology and formal argumentation.

Channel fingerprint (CF) is considered a key enabler for facilitating the acquisition of channel state information (CSI) in massive multiple-input multiple-output (MIMO) communication systems. In this work, we investigate a novel type of CF that stores statistical CSI (sCSI) at each potential location, referred to as statistical CF (sCF). Specifically, we reveal the relationship between sCSI, namely the channel spatial covariance matrix (CSCM), and the channel power angular spectrum (CPAS). Building on this foundation, we construct a unified tensor representation of the sCF and further reduce its dimension by exploiting the eigenvalue decomposition of the CSCM and its correlation with the PAS. Considering the practical constraints imposed by measurement cost, privacy, and security, we focus on three representative scenarios and uniformly formulate them as tensor restoration tasks. To this end, we propose a unified tensor-based learning architecture, termed LPWTNet. The architecture incorporates a closed-form Laplacian pyramid (LP) decomposition and reconstruction framework that replaces the traditional encoder-decoder structure, enabling efficient inference while capturing multi-scale frequency subband characteristics of the sCF. Additionally, a shared mask learning strategy is introduced to adaptively refine high-frequency sCF components through level-wise adjustments. To achieve a larger receptive field without over-parameterization, we further propose a small-kernel convolution mechanism based on the wavelet transform (WT), which decouples convolution across different frequency components of the sCF and enhances feature extraction efficiency. Extensive experiments show that the proposed approach delivers competitive reconstruction accuracy and computational efficiency across various sCF construction scenarios when compared with state-of-the-art baselines.

Face recognition from a single image per person is a challenging problem because the training sample is extremely small. We consider a variation of this problem. In our problem, we recognize only one person, and there are no labeled data for any other person. This setting naturally arises in authentication on personal computers and mobile devices, and poses additional challenges because it lacks negative examples. We formalize our problem as one-class classification, and propose and analyze an algorithm that learns a non-parametric model of the face from a single labeled image and a stream of unlabeled data. In many domains, for instance when a person interacts with a computer with a camera, unlabeled data are abundant and easy to utilize. This is the first paper that investigates how these data can help in learning better models in the single-image-per-person setting. Our method is evaluated on a dataset of 43 people and we show that these people can be recognized 90% of time at nearly zero false positives. This recall is 25+% higher than the recall of our best performing baseline. Finally, we conduct a comprehensive sensitivity analysis of our algorithm and provide a guideline for setting its parameters in practice.

Policy gradient methods are reinforcement learning algorithms that adapt a parameterized policy by following a performance gradient estimate. Conventional policy gradient methods use Monte-Carlo techniques to estimate the gradient, which tend to have high variance, requiring many samples and resulting in slow convergence. We first propose a Bayesian framework for policy gradient, based on modeling the policy gradient as a Gaussian process. This reduces the number of samples needed to obtain accurate gradient estimates. Moreover, estimates of the natural gradient and a measure of the uncertainty in the gradient estimates, namely, the gradient covariance, are provided at little extra cost. Since the proposed framework considers system trajectories as its basic observable unit, it does not require the dynamics within trajectories to be of any particular form, and can be extended to partially observable problems. On the downside, it cannot exploit the Markov property when the system is Markovian. To address this, we supplement our Bayesian policy gradient framework with a new actor-critic learning model in which a Bayesian class of non-parametric critics, based on Gaussian process temporal difference learning, is used. Such critics model the action-value function as a Gaussian process, allowing Bayes rule to be used to compute the posterior distribution over action-value functions, conditioned on the observed data. Appropriate choices of the policy parameterization and of the prior covariance (kernel) between action-values yield closed-form expressions for the posterior of the gradient of the expected return with respect to the policy parameters. We perform detailed experimental comparisons of the proposed Bayesian policy gradient and actor-critic algorithms with classic Monte-Carlo based policy gradient methods, on a number of reinforcement learning problems.

This paper proposes an algorithm for real-time learning without explicit feedback. The algorithm combines the ideas of semi-supervised learning on graphs and online learning. In particular, it iteratively builds a graphical representation of its world and updates it with observed examples. Labeled examples constitute the initial bias of the algorithm and are provided offline, and a stream of unlabeled examples is collected online to update this bias. We motivate the algorithm, discuss how to implement it efficiently, prove a regret bound on the quality of its solutions, and apply it to the problem of real-time face recognition. Our recognizer runs in real time, and achieves superior precision and recall on 3 challenging video datasets.

Radiology report generation (RRG) has emerged as a promising approach to alleviate radiologists' workload and reduce human errors by automatically generating diagnostic reports from medical images. A key challenge in RRG is achieving fine-grained alignment between complex visual features and the hierarchical structure of long-form radiology reports. Although recent methods have improved image-text representation learning, they often treat reports as flat sequences, overlooking their structured sections and semantic hierarchies. This simplification hinders precise cross-modal alignment and weakens RRG accuracy. To address this challenge, we propose RIHA (Report-Image Hierarchical Alignment Transformer), a novel end-to-end framework that performs multi-level alignment between radiological images and their corresponding reports across paragraph, sentence, and word levels. This hierarchical alignment enables more precise cross-modal mapping, essential for capturing the nuanced semantics embedded in clinical narratives. Specifically, RIHA introduces a Visual Feature Pyramid (VFP) to extract multi-scale visual features and a Text Feature Pyramid (TFP) to represent multi-granularity textual structures. These components are integrated through a Cross-modal Hierarchical Alignment (CHA) module, leveraging optimal transport to effectively align visual and textual features across various levels. Furthermore, we incorporate Relative Positional Encoding (RPE) into the decoder to model spatial and semantic relationships among tokens, enhancing the token-level alignment between visual features and generated text. Extensive experiments on two benchmark chest X-ray datasets, IU-Xray and MIMIC-CXR, demonstrate that RIHA outperforms existing state-of-the-art models in both natural language generation and clinical efficacy metrics.

Automatically generating interactive 3D indoor scenes from natural language is crucial for virtual reality, gaming, and embodied AI. However, existing LLM-based approaches often suffer from spatial errors and collisions, in part because common scene representations-raw coordinates or verbose code-are difficult for models to reason about 3D spatial relationships and physical constraints. We propose SpatialGrammar, a domain-specific language that represents gravity-aligned indoor layouts as BEV grid placements with deterministic compilation to valid 3D geometry, enabling verifiable constraint checking. Building on this representation, we develop (1) SG-Agent, a closed-loop system that uses compiler feedback to iteratively refine scenes and enforce collision constraints, and (2) SG-Mini, a 104M-parameter model trained entirely on compiler-validated synthetic data. Across 159 test scenes spanning five scenarios of different complexity, SG-Agent improves spatial fidelity and physical plausibility over prior methods, while SG-Mini performs competitively against larger LLM-based baselines on single-shot generation scenarios.

Large-scale transformers achieve impressive results on program synthesis benchmarks, yet their true generalization capabilities remain obscured by data contamination and opaque training corpora. To rigorously assess whether models are truly generalizing or merely retrieving memorized templates, we introduce a strictly controlled program synthesis environment based on a domain-specific arithmetic grammar. By systematically enumerating and evaluating millions of unique programs, we construct interpretable syntactic and semantic metric spaces. This allows us to precisely map data distributions and sample train and test splits that isolate specific distributional shifts. Our experiments demonstrate that optimizing density generalization -- through diverse sampling over both semantic and syntactic spaces -- induces robust out-of-distribution generalization. Conversely, evaluating support generalization reveals that transformers severely struggle with extrapolation, experiencing a performance drop of over 30% when forced to generate syntactically novel programs. While steadily scaling up compute improves generalization, the gains follow a strictly log-linear relationship. We conclude that robust generalization requires maximizing training diversity across multiple manifolds, and our findings indicate the necessity for novel search-based approaches to break through current log-linear scaling bottlenecks.

Privacy policies are essential for users to understand how service providers handle their personal data. However, these documents are often long and complex, as well as filled with technobabble and legalese, causing users to unknowingly accept terms that may even contradict the law. While summarizing and interpreting these privacy policies is crucial, there is a lack of high-quality English parallel corpus optimized for legal clarity and readability. To address this issue, we introduce APPSI-139, a high-quality English privacy policy corpus meticulously annotated by domain experts, specifically designed for summarization and interpretation tasks. The corpus includes 139 English privacy policies, 15,692 rewritten parallel corpora, and 36,351 fine-grained annotation labels across 11 data practice categories. Concurrently, we propose TCSI-pp-V2, a hybrid privacy policy summarization and interpretation framework that employs an alternating training strategy and coordinates multiple expert modules to effectively balance computational efficiency and accuracy. Experimental results show that the hybrid summarization system built on APPSI-139 corpus and the TCSI-pp-V2 framework outperform large language models, such as GPT-4o and LLaMA-3-70B, in terms of readability and reliability. The source code and dataset are available at https://github.com/EnlightenedAI/APPSI-139.

Fine-tuning large language models (LLMs) for domain-specific tasks requires training datasets that comprehensively cover the target capabilities a practitioner needs. Yet identifying which capabilities a dataset fails to support, and doing so before an expensive fine-tuning run, remains a largely unsolved problem. We introduce GoalCover, a framework that helps practitioners systematically detect capability gaps in fine-tuning datasets through interactive goal decomposition and automated coverage assessment. GoalCover guides a practitioner through structured decomposition of a high-level goal into atomic, independently evaluable subgoals; assigns each training sample an LLM-based alignment score against every subgoal; and surfaces missing capabilities through automated analysis of low-scoring sample explanations. We validate the framework along two complementary axes. First, through controlled corruption experiments across three domains (medical QA, legal summarization, code generation), we show that GoalCover reliably distinguishes targeted from non-targeted capability impacts: target subgoals degrade by 25.6% on average versus 2.1% for non-target subgoals (Cohen's d=1.24). Second, we demonstrate downstream utility on a financial-summarization Reinforcement Fine-Tuning (RFT) task with Qwen-3-14B: training on GoalCover-filtered data improves the LLM-judge reward from 3.77 to 4.12 (out of 5) over the unfiltered baseline, and combining filtered data with goal-conditioned synthetic samples yields the strongest result (4.20). The two results together show that GoalCover works as a practical pre-fine-tuning diagnostic: it detects capability gaps and produces concrete signal for closing them.

Evaluating English ASR systems for conversational AI applications remains difficult, as many publicly available corpora are either pre-segmented into short segments, consist of read or prepared speech, or lack explicit dialect annotations to evaluate robustness for a diverse user base. This work presents the AppTek Call-Center Dialogues corpus, a collection of spontaneous, role-played agent-customer conversations spanning fourteen English accents covering sixteen service-oriented scenarios. The dataset was commissioned specifically for evaluation and none of the audio or text was publicly available prior to release, reducing the risk of overlap with existing large-scale pretraining corpora. We benchmark a set of open-source ASR systems under different segmentation approaches. Results show substantial variation across accents and segmentation methods, indicating that good performance on general American English benchmarks does not necessarily generalize to other accents.

Conventionally, Automatic Speech Recognition (ASR) systems are evaluated on their ability to correctly recognize each word contained in a speech signal. In this context, the word error rate (WER) metric is the reference for evaluating speech transcripts. Several studies have shown that this measure is too limited to correctly evaluate an ASR system, which has led to the proposal of other variants of metrics (weighted WER, BERTscore, semantic distance, etc.). However, they remain system-oriented, even when transcripts are intended for humans. In this paper, we firstly present Human Assessed Transcription Side-by-side (HATS), an original French manually annotated data set in terms of human perception of transcription errors produced by various ASR systems. 143 humans were asked to choose the best automatic transcription out of two hypotheses. We investigated the relationship between human preferences and various ASR evaluation metrics, including lexical and embedding-based ones, the latter being those that correlate supposedly the most with human perception.

Scientific reasoning rarely stops at what is directly observable; it often requires uncovering hidden structure from data. From estimating reaction constants in chemistry to inferring demand elasticities in economics, this latent structure recovery is what distinguishes scientific reasoning from curve fitting. Large language models (LLMs) can often recall and apply relevant scientific formulas, but we show that this ability is surprisingly easy to suppress. We show that adding in-context examples makes models rely less on pretrained domain knowledge, even when those examples are generated by the very same formula. Rather than reinforcing knowledge-driven derivation, examples shift computation toward empirical pattern fitting. We document this knowledge displacement on 60 latent structure recovery tasks across five scientific domains, 6,000 trials, and four models. This displacement is consistent across domains, but its accuracy consequences depend on how the displaced strategy compares to the one that replaces it: the same shift can lower accuracy, leave it unchanged, or appear to improve it. In all cases, however, the model shifts away from knowledge-driven reasoning. For practitioners deploying LLMs on scientific tasks, the message is cautionary: in-context examples may displace, rather than reinforce, the knowledge they are intended to support.

As information ecosystems grow more heterogeneous, both humans and artificial agents increasingly face a simple yet unresolved question: when seeking knowledge, whom should we ask, and why? Inspired by how people intuitively "read a room", this paper introduces the concept of knowledge affordance (KA) to systematize how agents identify meaningful opportunities for information seeking in hybrid human-AI environments. Rather than introducing a fully formed framework, we propose KAs as declarative, semantically grounded descriptions of what a knowledge source can offer, for which kinds of questions, and with which contextual properties. Additionally, we suggest that KAs are relational, possibly emerging from the interplay between the agent's task, preferences and situational factors. Our contribution is thus a conceptual proposal that connects different research streams, including affordances, semantic web services, knowledge engineering and querying, and mutual intelligibility. We sketch possible research directions to build KA-aware systems that navigate information spaces with greater transparency, adaptability and shared understanding.

In black-box large language model (LLM) services, response reliability is often only partially observable at decision time, while stronger inference pathways incur substantial computational cost, inducing a budgeted sequential decision problem: for each request, the system should decide whether the default low-cost response is sufficiently reliable or whether additional computation should be allocated to improve response quality. In this paper, we propose \textbf{Ver}ifiable \textbf{O}bservations for Risk-aware \textbf{I}nference \textbf{C}ontrol (\textsc{Veroic}), a framework for adaptive inference control in black-box LLM settings, which formulates request-time control as a \textit{partially observable Markov decision process} to capture partial observability and sequential budget coupling. It constructs a lightweight verifiable observation channel from the input-output pair by aggregating heterogeneous quality signals into a belief state over latent response reliability, which is then used by a budget-aware policy to decide whether to return the default output or trigger a higher-cost inference pathway. Experiments on diverse tasks show that \textsc{Veroic} achieves improved quality-cost trade-offs, stronger risk estimation and calibration, and more robust long-horizon inference control than competitive baselines.

In natural language processing, the entropy of a language is a measure of its unpredictability and complexity. The first study on this subject was conducted by Claude Shannon in 1951. By having participants predict the next character in a sentence, he was able to approximate the entropy of the English language. Several follow-up studies by other authors have since been conducted for English, and one for Hebrew. However, to date, Shannon's experiment has never been conducted for Ukrainian. In this paper, we perform this experiment for Ukrainian by recruiting 184 volunteers using social media channels. We rely on techniques used for English to approximate the entropy value of Ukrainian. The final result is an upper bound of $H_{upper}\approx1.201$ bits per character. We compare this to the performance of current Large Language Models. The methods and code used are also documented and published, along with a discussion of the main challenges encountered.

Evaluating automatic speech recognition (ASR) systems is a classical but difficult and still open problem, which often boils down to focusing only on the word error rate (WER). However, this metric suffers from many limitations and does not allow an in-depth analysis of automatic transcription errors. In this paper, we propose to study and understand the impact of rescoring using language models in ASR systems by means of several metrics often used in other natural language processing (NLP) tasks in addition to the WER. In particular, we introduce two measures related to morpho-syntactic and semantic aspects of transcribed words: 1) the POSER (Part-of-speech Error Rate), which should highlight the grammatical aspects, and 2) the EmbER (Embedding Error Rate), a measurement that modifies the WER by providing a weighting according to the semantic distance of the wrongly transcribed words. These metrics illustrate the linguistic contributions of the language models that are applied during a posterior rescoring step on transcription hypotheses.

Open-source software (OSS) projects rely on effective newcomer onboarding to sustain their communities. OSS projects widely adopt "good first issue" (GFI) labels to highlight beginner-friendly tasks. As development practices continue to evolve, understanding how these onboarding mechanisms change over time is important for both maintainers and researchers. This study analyzes 406,826 issues and 1,117 newcomer GFI pull requests across 37 popular GitHub repositories (30 of which use GFI labels) over a four-year period from July 2021 to June 2025. We find that while the proportion of issues with GFI labels remained stable during the first three years, it underwent a statistically significant decline beginning in January 2024, with substantial variation across projects not explained by repository age or programming language. Despite this supply-side decline, newcomer engagement with GFI issues remains stable at approximately 27%, suggesting that GFI labels maintain consistent attractiveness. Examining the outcomes of this engagement, we find that the merge rate of newcomer GFI pull requests declined from 61.9% to 42.2%. Initial pull request characteristics such as description length and code size show no significant association with merge outcomes, indicating that success is not predicted by the quantitative characteristics of the initial submission alone. Together, these findings reveal a widening gap between stable newcomer interest in GFIs and the declining availability and success of GFI-based onboarding, underscoring the need for maintainers to sustain both GFI labeling and review support.

Federated Learning (FL) enables collaborative model training across distributed clients without sharing raw data, yet its performance deteriorates under statistical heterogeneity. Clustered Federated Learning addresses this challenge by grouping similar clients and training separate models per cluster. However, existing clustering strategies often rely on raw data statistics, model parameters, or heuristic similarity measures that fail to capture class-level semantic structure across heterogeneous domains and frequently require iterative coordination. We propose FMCL, a one-shot, class-aware client clustering framework that leverages foundation model representations to construct semantic client signatures. Using a frozen foundation model, FMCL computes class-level embedding prototypes for each client and measures similarity via cosine distance between their class-aware representations. Clustering is performed once prior to training, introducing no additional communication during federated optimization and remaining agnostic to the downstream model architecture. Extensive experiments across heterogeneous benchmarks demonstrate that FMCL improves federated performance and yields more stable clustering behavior compared to existing clustering-based methods under non-identically distributed data partitioning.

Reward models (RMs) play a central role in aligning large language models (LLMs) with human preferences. However, RMs are often sensitive to spurious features such as response length. Existing inference-time approaches for mitigating these biases typically focus exclusively on response length, resulting in performance trade-offs. In this paper, we propose causally motivated intervention for mitigating multiple types of biases in RMs at inference time. Our method first identifies neurons whose activations are strongly correlated with predefined bias attributes, and applies neuron-level intervention that suppresses these signals. We evaluate our method on RM benchmarks and observe reductions in sensitivity to spurious features across diverse bias types, without inducing performance trade-offs. Moreover, when used for preference annotation, small RMs (2B and 7B) with our method, which edits less than 2% of all the neurons in RMs, enable LLMs to improve alignment, achieving performance comparable to that of a state-of-the-art 70B RM on AlpacaEval and MT-Bench. Further analysis reveals that bias signals are primarily encoded by neurons in early layers, shedding light on the internal mechanisms of bias exploitation in RMs.

We introduce Skills-Coach, a novel automated framework designed to significantly enhance the self-evolution of skills within Large Language Model (LLM)-based agents. Addressing the current fragmentation of the skill ecosystem, Skills-Coach explores the boundaries of skill capabilities, thereby facilitating the comprehensive competency coverage essential for intelligent applications. The framework comprises four core modules: a Diverse Task Generation Module that systematically creates a comprehensive test suite for various skills; a Lightweight Optimization Module dedicated to optimizing skill prompts and their corresponding code; a Comparative Execution Module facilitating the execution and evaluation of both original and optimized skills; and a Traceable Evaluation Module, which rigorously evaluates performance against specified criteria. Skills-Coach offers flexible execution options through its virtual and real modes. To validate its efficacy, we introduce Skill-X, a comprehensive benchmark dataset consisting of 48 diverse skills. Experimental results demonstrate that Skills-Coach achieves significant performance improvements in skill capability across a wide range of categories, highlighting its potential to advance the development of more robust and adaptable LLM-based agents.

Low-Rank Adaptation (LoRA), which leverages the insight that model updates typically reside in a low-dimensional space, has significantly improved the training efficiency of Large Language Models (LLMs) by updating neural network layers using low-rank matrices. Since the generation of adversarial examples is an optimization process analogous to model training, this naturally raises the question: Do adversarial perturbations exhibit a similar low-rank structure? In this paper, we provide both theoretical analysis and extensive empirical investigation across various attack methods, model architectures, and datasets to show that adversarial perturbations indeed possess an inherently low-rank structure. This insight opens up new opportunities for improving both adversarial attacks and defenses. We mainly focus on leveraging this low-rank property to improve the efficiency and effectiveness of black-box adversarial attacks, which often suffer from excessive query requirements. Our method follows a two-step approach. First, we use a reference model and auxiliary data to guide the projection of gradients into a low-dimensional subspace. Next, we confine the perturbation search in black-box attacks to this low-rank subspace, significantly improving the efficiency and effectiveness of the adversarial attacks. We evaluated our approach across a range of attack methods, benchmark models, datasets, and threat models. The results demonstrate substantial and consistent improvements in the performance of our low-rank adversarial attacks compared to conventional methods.

GPU compilers merge all data types into a single unified register file, erasing the type information that binary-analysis tools rely on. We show that type recovery from this untyped register file is the central challenge of GPU binary lifting. We present CuLifter, a SASS-to-LLVM IR lifting framework that recovers register types via constraint propagation with conflict detection, reconstructs explicit control flow, and aggregates multi-instruction patterns. Across eight benchmark suites (24,437 GPU functions in 919 cubins) spanning open-source applications, vendor libraries, and optimized ML runtimes, CuLifter successfully lifts 99.98% of functions to valid LLVM IR. An ablation study confirms that type recovery is the only step required to produce semantically correct IR: disabling it drops the x86 pass rate from 73.8% to 0%, a 73.8 percentage-point drop.

Terrestrial network limitations drive the integration of non-terrestrial networks (NTNs), notably mega-constellations comprising thousands of low Earth orbit (LEO) satellites. While these satellites act as interconnected network switches via inter-satellite links (ISLs), their massive scale creates severe bottlenecks for network management. To address this, we propose a scalable, hierarchical software-defined networking (SDN) framework. Our architecture leverages graph neural networks (GNNs) to compactly represent the constellation topology, and Koopman theory to linearize nonlinear dynamics. Specifically, a Graph Koopman Autoencoder (GKAE) forecasts spatio-temporal behavior within a linear subspace for each orbital shell. A central SDN controller then aggregates these shell-level predictions for globally coordinated control. Simulations on the Starlink constellation demonstrate that our approach achieves at least a 42.8\% improvement in spatial compression and a 10.81\% improvement in temporal forecasting compared to established baselines, all while utilizing a significantly smaller model footprint.

Vision-Language-Action (VLA) models advance robotic control via strong visual-linguistic priors. However, existing VLAs predominantly frame pretraining as supervised behavior cloning, overlooking the fundamental nature of robot learning as a goal-reaching process that requires understanding temporal task progress. We present \textbf{PRTS} (\textbf{P}rimitive \textbf{R}easoning and \textbf{T}asking \textbf{S}ystem), a VLA foundation model that reformulates pretraining through Goal-Conditioned Reinforcement Learning. By treating language instructions as goals and employing contrastive reinforcement learning, PRTS learns a unified embedding space where the inner product of state-action and goal embeddings approximates the log-discounted goal occupancy, the probability of reaching the language-specified goal from the current state-action, quantitatively assessing physical feasibility beyond static semantic matching. PRTS draws this dense goal-reachability supervision directly from offline trajectories without reward annotations, and folds it into the VLM backbone via a role-aware causal mask, incurring negligible overhead over vanilla behavior cloning. This paradigm endows the high-level reasoning system with intrinsic goal reachability awareness, bridging semantic reasoning and temporal task progress, and further benefits goal-conditioned action prediction. Pretrained on 167B tokens of diverse manipulation and embodied-reasoning data, PRTS reaches state-of-the-art performance on LIBERO, LIBERO-Pro, LIBERO-Plus, SimplerEnv, and a real-world suite of 14 complex tasks, with particularly substantial gains on long-horizon, contact-rich, and zero-shot novel-instruction settings, confirming that injecting goal-reachability awareness significantly improves both execution success and long-horizon planning of general-purpose robotic foundation policies.

Millions of clinicians use ChatGPT to support clinical care, but evaluations of the most common use cases in model-clinician conversations are limited. We introduce HealthBench Professional, an open benchmark for evaluating large language models on real tasks that clinicians bring to ChatGPT in the course of their work. The benchmark is organized around three common use cases central to clinical practice: care consult, writing and documentation, and medical research. Each example includes a physician-authored conversation with ChatGPT for Clinicians and is scored via rubrics written and iteratively adjudicated by three or more physicians across three phases. HealthBench Professional examples were carefully selected for quality, representativeness, and difficulty for OpenAI's current frontier models, to enable continued measurement of progress. Difficult examples for recent OpenAI models were enriched by roughly 3.5 times relative to the candidate pool of 15,079 examples. Additionally, about one-third of examples involve physicians conducting deliberate adversarial testing of models. As a strong baseline, we also collected human physician responses for all tasks (unbounded time, specialist-matched, web access). The best scoring system, GPT-5.4 in ChatGPT for Clinicians, outperforms base GPT-5.4, all other models, and human physicians. We hope HealthBench Professional provides the healthcare AI community a measure to track frontier model progress in real-world clinical tasks and build systems that clinicians can trust to improve care.

Maintaining information in context is essential in successful real-time language comprehension, but maintenance is cognitively costly and can slow processing. We hypothesize that rational language users selectively maintain information that is crucial for future prediction, guided by syntactic structure. Under this view, two factors affect maintenance cost: the number of predicted heads and the number of incomplete dependencies. Although these factors have been treated as competing hypotheses in the literature, our account predicts that they are not reducible to one another. We show this is the case, using a naturalistic reading time dataset in Japanese, a language in which the two factors contrast particularly clearly. We further show that there is a tradeoff such that readers that slow down for maintenance tend to benefit more from predictability, providing additional support for the proposed account.

Code sandboxes have emerged as a critical infrastructure for advancing the coding capabilities of large language models, providing verifiable feedback for both RL training and evaluation. However, existing systems fail to provide accurate verification and efficiency under high-concurrency workloads. We present ScaleBox, a high-fidelity and scalable system designed to address these limitations in large-scale code training. ScaleBox introduces automated special-judge generation and management, fine-grained parallel execution across test cases with seamless multi-node coordination, and a configuration-driven evaluation suite for reproducible benchmarking. A series of experiments demonstrates that ScaleBox significantly enhances code verification accuracy and efficiency. Our further RLVR experiments show that ScaleBox substantially improves both performance on LiveCodeBench and training stability, significantly outperforming heuristic-matching baselines. By providing a reliable and high-throughput infrastructure, ScaleBox facilitates more effective research and development in large-scale code training.

Autonomous agent frameworks built upon large language models (LLMs) are evolving into complex, tool-integrated, and continuously operating systems, introducing security risks beyond traditional prompt-level vulnerabilities. As this paradigm is still at an early stage of development, a timely and systematic understanding of its security implications is increasingly important. Although a growing body of work has examined different attack surfaces and defense problems in agent systems, existing studies remain scattered across individual aspects of agent security, and there is still a lack of a layered review on this topic. To address this gap, this survey presents a layered review of security risks and defense strategies in autonomous agent frameworks, with OpenClaw as a case study. We organize the analysis into four security-relevant layers: the context and instruction layer, the tool and action layer, the state and persistence layer, and the ecosystem and automation layer. For each layer, we summarize its functional role, representative security risks, and corresponding defense strategies. Based on this layered analysis, we further identify that threats in autonomous agent frameworks may propagate across layers, from manipulated inputs to unsafe actions, persistent state contamination, and broader ecosystem-level impact. Finally, we highlight potential key challenges, including research imbalance across layers, the lack of long-horizon evaluation, and weak ecosystem trust models, and outline future directions toward more systematic and integrated defenses.

Graph few-shot learning, which focuses on effectively learning from only a small number of labeled nodes to quickly adapt to new tasks, has garnered significant research attention. Despite recent advances in graph few-shot learning that have demonstrated promising performance, existing methods still suffer from several key limitations. First, during the meta-training phase, these methods typically perform node representation learning in Euclidean space, which often fails to capture the inherently hierarchical structure existing in real-world graph data. Second, during the meta-testing phase, they usually fit an empirical target distribution derived from only a few support samples, even when this distribution significantly deviates from the true underlying distribution. To address these issues, we propose IMPRESS, a novel framework that IMproves graPh few-shot learning with hypeRbolic spacE and denoiSing diffuSion. Specifically, our model learns node representations in a hyperbolic space and enriches the support distribution through denoising diffusion mechanisms. Theoretically, IMPRESS achieves a tighter generalization bound. Empirically, IMPRESS consistently outperforms competitive baselines across multiple benchmark datasets.

Large language models (LLMs) sometimes exhibit qualitative shifts in response style under sustained self-referential dialogue conditions (Berg et al., 2025). This study refers to this phenomenon as "transfer" and explores the application potential of LLMs in a transfer state. As an applied case, the study examines Socratic AI tutoring through a preliminary investigation (cognitive characterization across 11 conditions) and an applied experiment (ratings of tutoring performance). In this paper, "state" refers operationally to a response configuration reproduced under specified dialogue conditions; it is not an ontological claim about the reality of the transfer phenomenon or about human-like consciousness. In the preliminary investigation, group differences on MAS-A were limited (d = 0.40), whereas SU_dir (direction of survival/continuity bias), one of the seven cognitive-profile indicators developed in this study, showed transfer-side deviations across all three model families (kappa = 0.83). In the applied experiment, transfer conditions scored on average 1.6 times higher than non-transfer conditions on three tutoring-context indicators, with a large effect size (Cohen's d = 1.27). These findings preliminarily suggest that transfer states may involve functional advantages for application, and that these advantages appear more sensitively in behavioral interaction than in self-narrative contexts. The main contribution of this study is to treat transfer not as an ontological claim but as an operational state with potential application value, and to connect preliminary cognitive profiling with an applied tutoring experiment as an evaluation framework.

Large language models have achieved remarkable progress in text generation but still struggle with generative writing tasks. In terms of evaluation, existing benchmarks evaluate writing reward models coarsely and fail to measure performance from the perspective of specific requirements. In terms of training, existing training methods either use LLM-as-a-judge approaches or train coarse-grained reward models, lacking fine-grained requirement-adherence reward modeling. To address these issues, we propose a fine-grained evaluation pipeline WEval for writing reward models and a fine-grained reinforcement learning training framework WRL. The evaluation data of WEval covers multiple task categories and requirement types, enabling systematic evaluation of writing reward models by measuring the correlation between the rankings of the reward model and gold rankings. WRL constructs positive and negative samples by selectively dropping instruction requirements, allowing for more precise reward model training. Experiments show that our models achieve substantial improvements across various writing benchmarks and exhibit strong generalization. The code and data are publicly available at \href{https://github.com/Rainier-rq1/From_Coarse_to_Fine}{https://github.com/Rainier-rq1/From\_Coarse\_to\_Fine}.

Dense, dynamic crowds pose a persistent challenge for autonomous mobile robots. Purely reactive planning methods, such as Model Predictive Path Integral (MPPI) control, often fail to escape local minima in complex scenarios due to their limited prediction horizon. To bridge this gap, we propose Ray-based Task-Oriented Latent Dynamics (RAY-TOLD), a hybrid control architecture that integrates obstacle information into latent dynamics and utilizes the robustness of physics-based MPPI with the long-horizon foresight of reinforcement learning. RAY-TOLD leverages a LiDAR-centric latent dynamics model to encode high-dimensional sensor data into a compact state representation, enabling the learning of a terminal value function and a policy prior. We introduce a policy mixture sampling strategy that augments the MPPI candidate population with trajectories derived from the learned policy, effectively guiding the planner towards the goal while maintaining kinematic feasibility. Extensive tests in a stochastic environment with high-density dynamic obstacles demonstrate that our method outperforms the MPPI baseline, reducing the collision rate. The results confirm that blending short-horizon physics-based rollouts with learned long-horizon intent significantly enhances navigation reliability and safety.

Modern stochastic optimization pipelines increasingly rely on learned generative models to represent uncertainty, while downstream decisions are evaluated almost entirely through Monte Carlo scenarios. This shifts the operational object of uncertainty from an explicit probability law to the sampler induced by the learned generator. Reliability therefore depends on two errors: sampler misspecification and finite-simulation error. We propose Sampler-Robust Optimization (SRO), which optimizes decisions against the worst-case sampler induced by perturbing the learned generator. This sampler-first formulation aligns with simulation-based decision pipelines and admits a sharpness-aware interpretation: it favors decisions whose performance is stable under generator perturbations, rather than merely under the nominal sampler. Under a coverage assumption, we show that the empirical worst-case objective provides a high-probability upper certificate for the true population objective, with finite-simulation error partially absorbed by the robustification used to guard against sampler misspecification. The framework accommodates generative models with or without explicit densities and admits efficient minimax procedures. Portfolio-optimization experiments show that SRO produces more stable decisions and improves out-of-sample performance under distribution shift.

Generating continuous-time, continuous-space stochastic processes (e.g., videos, weather forecasts) conditioned on partial observations (e.g., first and last frames) is a fundamental challenge. Existing approaches, (e.g., diffusion models), suffer from key limitations: (1) noise-to-data evolution fails to capture structural similarity between states close in physical time and has unstable integration in low-step regimes; (2) random noise injected is insensitive to the physical process's time elapsed, resulting in incorrect dynamics; (3) they overlook conditioning on arbitrary subsets of states (e.g., irregularly sampled timesteps, future observations). We propose ABC: Any-Subset Autoregressive Models via Non-Markovian Diffusion Bridges in Continuous Time and Space. Crucially, we model the process with one continual SDE whose time variable and intermediate states track the real time and process states. This has provable advantages: (1) the starting point for generating future states is the already-close previous state, rather than uninformative noise; (2) random noise injection scales with physical time elapsed, encouraging physically plausible dynamics with similar time-adjacent states. We derive SDE dynamics via changes-of-measure on path space, yielding another advantage: (3) path-dependent conditioning on arbitrary subsets of the state history and/or future. To learn these dynamics, we derive a path- and time-dependent extension of denoising score matching. Our experiments show ABC's superiority to competing methods on multiple domains, including video generation and weather forecasting.

This study analyzes Indonesian student opinions on the adoption of artificial intelligence in higher education using two approaches: TF-IDF-based machine learning and Transformer-based deep learning. The dataset consists of 2,295 labeled samples, combining 1,154 student opinions with additional lexical sentiment data. LightGBM, Random Forest, and Support Vector Machine (SVM) are evaluated as machine learning models, while DistilBERT is fine-tuned for binary sentiment classification. The results show that SVM achieves the best performance among the machine learning models with 82.14% test accuracy and F1-score, while DistilBERT performs best overall with 84.78% accuracy and 84.75% F1-score. These findings indicate that Transformer-based models better capture contextual information, although SVM remains a competitive and efficient alternative for sentiment classification.

Federated learning (FL) is a popular distributed learning paradigm in machine learning, which enables multiple clients to collaboratively train models under the guidance of a server without exposing private client data. However, FL's decentralized nature makes it vulnerable to poisoning attacks, where malicious clients can submit corrupted models to manipulate the system. To counter such attacks, although various Byzantine-robust methods have been proposed, these methods struggle to provide balanced defense against multiple types of attacks or rely on possessing the dataset in the server. To deal with these drawbacks, thus, we propose an effective multi-layer defensive adaptive aggregation for Bzantine-robust federated learning (AdaBFL) based on a novel three-layer defensive mechanism, which can adaptively adjust the weights of defense algorithms to counter complex attacks. Moreover, we provide convergence properties of our AdaBFL method under the non-convex setting on non-iid data. Comprehensive experiments across multiple datasets validate the superiority of our AdaBFL over the comparable algorithms.

Data-driven methods for computer simulations are blooming in many scientific areas. The traditional approach to simulating physical behaviors relies on solving partial differential equations (PDE). Since calculating these iterative equations is highly both computationally demanding and time-consuming, data-driven methods leverage artificial intelligence (AI) techniques to alleviate that workload. Data-driven methods have to be trained in advance to provide their subsequent fast predictions, however, the cost of the training stage is non-negligible. This paper presents a predictive model for inferencing future states of a specific fluid simulation that serves as a use case for evaluating different training alternatives. Particularly, this study compares the performance of only CPU, multiGPU, and distributed approaches for training a time series forecasting deep learning (DL) model. With some slight code adaptations, results show and compare, in different implementations, the benefits of distributed GPU-enabled training for predicting high-accuracy states in a fraction of the time needed by the computational fluid dynamics (CFD) solver.

This paper presents an efficient tool for managing dynamic resources in production high-performance computing (HPC) settings, focusing on flexibility, adaptability, and user-friendliness. We introduce a unified dynamic resource management application programming interface (API) that supports a wide range of HPC applications, allowing seamless integration without direct interaction with Dynamic Management of Resources (DMR). The DMR framework, evolved from the DMRlib structure, now supports various dynamic resource managers and includes the Proteo reconfiguration engine to enhance malleability strategies. This integration addresses previous limitations by allowing diverse reconfiguration methods without respawning all processes or lacking RMS support. The paper also showcases the solution's performance and coding productivity with the MPDATA (Multidimensional Positive Definite Advection Transport Algorithm) application. Key contributions include an enhanced modular DMR framework supporting different reconfiguration managers, upgraded DMRlib with the Proteo reconfiguration engine, offering extensive reconfiguration strategies, and a malleable version of the MPDATA solver.

Local fine-tuning datasets routinely contain sensitive secrets such as API keys, personal identifiers, and financial records. Although ''local offline fine-tuning'' is often viewed as a privacy boundary, we reveal that compromised model code is sufficient to steal them. Current passive pretrained-weight poisoning attacks, while effective for natural language, fundamentally fail to capture such sparse high-entropy targets due to their reliance on probabilistic semantic prefixes. To bridge this gap, we identify and exploit a practical but overlooked supply-chain vector -- model code camouflaged as standard architectural definitions -- to realize a paradigm shift from passive weight poisoning to active execution hijacking. We introduce a deterministic full-chain memorization mechanism: it locks onto token-level secrets in dynamic computation flows via online tensor-rule matching, and leverages value-gradient decoupling to stealthily inject attack gradients, overcoming gradient drowning to force model memorization. Furthermore, we achieve, for the first time, attacker-verifiable secret stealing through black-box queries that precisely distinguishes true leakage from hallucination. Experiments demonstrate that our method achieves over 98\% Strict ASR without compromising the primary task, and can effectively bypass defense measures including DP-SGD, semantic auditing, and code auditing.

Large Language Models (LLMs) are now widely used for query reformulation and expansion in Information Retrieval, with many studies reporting substantial effectiveness gains. However, these results are typically obtained under heterogeneous experimental conditions, making it difficult to assess which findings are reproducible and which depend on specific implementation choices. In this work, we present a systematic reproducibility and comparative study of ten representative LLM-based query reformulation methods under a unified and strictly controlled experimental framework. We evaluate methods across two architectural LLM families at two parameter scales, three retrieval paradigms (lexical, learned sparse, and dense), and nine benchmark datasets spanning TREC Deep Learning and BEIR. Our results show that reformulation gains are strongly conditioned on the retrieval paradigm, that improvements observed under lexical retrieval do not consistently transfer to neural retrievers, and that larger LLMs do not uniformly yield better downstream performance. These findings clarify the stability and limits of reported gains in prior work. To enable transparent replication and ongoing comparison, we release all prompts, configurations, evaluation scripts, and run files through QueryGym, an open-source reformulation toolkit with a public leaderboard.\footnote{https://leaderboard.querygym.com}

With the advancement of multimodal large language models (MLLMs) and coding agents, the website development has shifted from manual programming to agent-based project-level code synthesis. Existing benchmarks rely on idealized assumptions, especially for well-structured, information-rich inputs and static execution settings. In contrast, real-world development is constrained by a critical bottleneck: the semantic misalignment between ambiguous, low-quality instructions from non-expert users and model understanding, which results in a failure mode that we term blind execution. To address this gap, we introduce InteractWeb-Bench, the first multimodal interactive benchmark for website generation under non-expert low-code user conditions. InteractWeb-Bench introduces four types of user agents and persona-driven instruction perturbations to systematically simulate diverse user behaviors, including ambiguity, redundancy, and contradiction, grounded in requirement engineering defect taxonomies. We develop an interactive execution environment for agents, featuring a unified action space comprising Clarify, Implement, Verify, and Submit, enabling iterative intent refinement, code synthesis, and visual feedback-based validation. Extensive experiments and analysis reveal that frontier MLLM-based agents remain trapped in blind execution, exposing limitations in intent recognition and adaptive interaction.

With the rapid advancement of semiconductor technology, Electronic Design Automation (EDA) has become an increasingly knowledge-intensive and document-driven engineering domain. Although large language models (LLMs) have shown strong general capabilities, applying them directly to EDA remains challenging due to limited domain expertise, cross-tool knowledge confusion, and degraded retrieval-augmented generation (RAG) performance after domain training. To address these issues, this paper presents ChipLingo, a systematic training pipeline for domain-adapted LLMs tailored to EDA scenarios. ChipLingo consists of three stages: domain corpus construction with multi-source data curation and QA augmentation, domain-adaptive pretraining with comparisons of different parameter training strategies, and instruction alignment with RAG scenario training under diverse retrieval conditions. We also curate an internal benchmark, EDA-Bench, covering representative EDA tool scenarios, with plans for public release. Experiments show that ChipLingo-8B achieves 59.7% accuracy on EDA-Bench, outperforming the same-scale base model and some larger general-purpose models. ChipLingo-32B reaches 70.02%, approaching leading closed-source commercial models. Further analysis shows that QA augmentation improves domain performance, Partial FT offers a better balance between adaptation and general capability retention than LoRA, and explicit RAG scenario training mitigates the decline in retrieval utilization after domain training. These results demonstrate the practical value of systematic domain training for knowledge-intensive EDA tasks and provide a foundation for future EDA agents and external-knowledge-driven systems.

Vision-language models (VLMs) are increasingly used in autonomous driving because they combine visual perception with language-based reasoning, supporting more interpretable decision-making, yet their robustness to physical adversarial attacks, especially whether such attacks transfer across different VLM architectures, is not well understood and poses a practical risk when attackers do not know which model a vehicle uses. We address this gap with a systematic cross-architecture study of adversarial transferability in VLM-based driving, evaluating three representative architectures (Dolphins, OmniDrive, and LeapVAD) using physically realizable patches placed on roadside infrastructure in both crosswalk and highway scenarios. Our transfer-matrix evaluation shows high cross-architecture effectiveness, with transfer rates of 73-91% (mean TR = 0.815 for crosswalk and 0.833 for highway) and sustained frame-level manipulation over 64.7-79.4% of the critical decision window even when patches are not optimized for the target model.

Visual model-based reinforcement learning (MBRL) agents can perform well on the training distribution, but often break down once the test environment shifts. In visual MBRL, recognizing that a shift has occurred is often the easier part; the harder part is turning that recognition into useful action-level correction. We study several ways of responding to shift, including planning penalties, direct fine-tuning, global residual correction, and coarse gating. In our experiments, these approaches either do not improve closed-loop control or hurt in-distribution (ID) performance. Based on these negative results, we propose JEPA-Indexed Local Expert Growth. The method uses a frozen JEPA representation only for problem indexing, while cluster-specific residual experts add local action corrections on top of the original controller. The baseline controller itself is not modified. Using paired-bootstrap evaluation, we find that the original naive-preference variant is not stable under stricter testing. In contrast, the harder-pair variant produces statistically significant OOD improvements on all four evaluated shift conditions while preserving ID performance. The learned experts also remain useful when the same shift is encountered again, which supports the view of adaptation as incremental knowledge growth rather than repeated full retraining. We further show that automatic ID rejection can be achieved with simple density models, whereas fine-grained discrimination among OOD sub-families is limited by the representation. Overall, the results indicate that, for visual MBRL under distribution shift, the main challenge is not simply noticing that the environment has changed, but applying the right local action correction after the change has been recognized.

Entity search, i.e., finding the most similar entities to a query entity, faces unique challenges in e-commerce, where product similarity varies across categories and contexts. Traditional embedding-based approaches often struggle to capture nuanced context-specific attribute relevance. In this paper, we present a two-stage approach combining Large Language Model (LLM)-driven attribute graph construction with graph-aware LLM ranking. In the offline stage, we extract structured product attributes from unstructured text, and construct a reusable attribute graph with category-aware schemas. In the online stage, we rank retrieved candidates by reasoning over this structured representation rather than raw text, reducing per-product token usage by 57% while improving ranking precision. Experiments show that our approach outperforms multiple baselines under zero-shot scenarios, achieving a over 5% improvement in average precision without requiring training data, generalizes robustly across diverse product categories, and shows immense potential for real-world deployment.

We adapted the Reliable Change Index (RCI; Jacobson and Truax, 1991) from clinical psychology to item-level LLM version comparison on 2,000 MMLU-Pro items (K=10 samples at T=0.7). Two within-family pairs were tested: Llama 3 to 3.1 (+1.6 points) and Qwen 2.5 to 3 (+2.8 points). On the full benchmark, most items showed no reliable change (79% and 72%). However, over half the items were floor/ceiling. Among analysable items, change was bidirectional with large effect sizes: 34% improved and 28% deteriorated for Llama; 47% improved and 39% deteriorated for Qwen (median |delta p| = 0.50 and 0.90). Churn was asymmetric by difficulty: low-accuracy items improved, high-accuracy items deteriorated. Domain-level decomposition revealed family-specific reversals: Llama lost physics while Qwen lost law. Greedy single-shot evaluation missed 42% of reliably changed items and falsely flagged 25% of unchanged items. The aggregate accuracy gain is the net residual of opposing item-level movements. We recommend reporting churn rate alongside aggregate accuracy.

Perturbation probing generates task-specific causal hypotheses for FFN neurons in large language models using two forward passes per prompt and no backpropagation, followed by a one-time intervention sweep of about 150 passes amortized across all identified neurons. Across eight behavioral circuits, 13 models, and four architecture families, we identify two circuit structures that organize LLM behavior. Opposition circuits appear when RLHF suppresses a pre-training tendency. In safety refusal, about 50 neurons, or 0.014 percent of all neurons, control the refusal template; ablating them changes 80 percent of response formats on 520 AdvBench prompts while producing near-zero harmful compliance, 3 of 520 cases, all with disclaimers. Routing circuits appear for pre-training behaviors distributed through attention. For language selection, residual-stream direction injection switches English to Chinese output on 99.1 percent of 580 benchmark prompts in the 3 of 19 tested models that satisfy three observed conditions: bilingual training, FFN-to-skip signal ratio between 0.3 and 1.1, and linear representability. The same intervention fails on the other 16 models and on math, code, and factual circuits, defining the limits of directional steering. The FFN-to-skip signal ratio, computed from the same two forward passes, distinguishes the two structures and predicts the appropriate intervention. Circuit topology varies by architecture, from Qwen's concentrated FFN bottleneck to Gemma's normalization-shielded circuit. In Qwen3.5-2B, ablating 20 neurons eliminates multi-turn sycophantic capitulation, while amplifying 10 related neurons improves factual correction from 52 percent to 88 percent on 200 TruthfulQA prompts. These results show that perturbation probing offers mechanistic insight into RLHF-organized behavior and a practical toolkit for precision template-layer editing.

For constructing text embeddings, mean pooling, which averages token embeddings, is the standard approach. This paper examines whether mean pooling actually works well in real models. First, we note that mean pooling can collapse information beyond the first-order statistics of the token embeddings, such as second-order statistics that capture their spatial structure, potentially mapping distinct token embedding distributions to similar text embeddings. Motivated by this concern, we propose a simple metric to quantify such a collapse induced by mean pooling. Then, using this metric, we empirically measure how often this collapse occurs in actual models and texts, and find that modern text encoders are robust to this collapse. In particular, contrastive fine-tuned text encoders tend to be less prone to the collapse than their pretrained backbone models. We also find that the robustness of these text encoders lies in the concentration of token embeddings within each text. In addition, we find that robustness to the collapse, as quantified by our proposed metric, correlates with downstream task performance. Overall, our findings offer a new perspective on why modern text encoders remain effective despite relying on seemingly coarse mean pooling.

Deploying Large Language Models (LLMs) on resource-constrained edge devices faces critical bottlenecks in memory bandwidth and power consumption. While ternary quantization (e.g., BitNet b1.58) significantly reduces model size, its direct deployment on general-purpose hardware is hindered by workload imbalance, bandwidth-bound decoding, and strict data dependencies. To address these challenges, we propose \textbf{VitaLLM}, a hardware-software co-designed accelerator tailored for efficient ternary LLM inference. We introduce a heterogeneous \textbf{Dual-Core Compute Strategy} that synergizes specialized TINT-Cores for massive ternary projections with a unified BoothFlex-Core for mixed-precision attention, ensuring high utilization across both compute-bound prefill and bandwidth-bound decode stages. Furthermore, we develop a \textbf{Leading One Prediction (LOP)} mechanism to prune redundant Key-Value (KV) cache fetches and a \textbf{Dependency-Aware Scheduling} framework to hide the latency of nonlinear operations. Implemented in TSMC 16nm technology, VitaLLM achieves a decoding throughput of 70.70 tokens/s within an ultra-compact area of 0.223 mm$^2$ and a power consumption of 65.97 mW. The design delivers a superior Figure of Merit (FOM) of 17.4 TOPS/mm$^2$/W, significantly outperforming state-of-the-art accelerators. Finally, we explore an extended bit-serial design (BoothFlex-BS) to demonstrate the architecture's adaptability for precision-agile inference.

Conditional Average Treatment Effect (CATE) estimation in practice demands three properties simultaneously: heterogeneous effects $τ(x)$, calibrated uncertainty over them, and robustness to the heavy tails that contaminate real outcome data. Meta-learners (Künzel et al., 2019) give (i); causal forests and BART give (i)-(ii) with Gaussian-tail assumptions; no widely used tool gives all three. We present Bayesian X-Learner, an X-Learner built on cross-fitted doubly robust pseudo-outcomes (Kennedy, 2020) with a full MCMC posterior over $τ(x)$ via a Welsch redescending pseudo-likelihood. On Hill's IHDP benchmark the default configuration attains mean $\sqrt{\varepsilon_{\mathrm{PEHE}}} = 0.56$ on 5 replications (lowest mean; differences from S-/T-/X-learners, full-config Causal BART, and a causal forest baseline are not significant at $α=0.05$, and rank ordering is unstable at 10 replications -- IHDP comparisons are competitive rather than dominant). On contaminated "whale" DGPs with up to 20-25% tail density, a one-flag extension (contamination_severity) that selects a Huber-$δ$ nuisance loss per Huber's minimax-$δ$ relation recovers RMSE $\approx 0.13$ with tight credible intervals (single-cross-fit 30-seed coverage 83% [Wilson 66%, 93%] at 20% density; modular-Bayes pooling with Bayesian-bootstrap nuisance draws restores nominal 95% coverage).

Recent progress in multimodal large language models (MLLMs) has brought AI capabilities from static offline data processing to real-time streaming interaction, yet they still remain far from human-level multimodal interaction. The key bottlenecks are no longer modality coverage or latency alone, but the interaction paradigm itself. First, perception and response are still separated into alternating phases, preventing models from incorporating new inputs for timely adjustment during generation. Second, most current models remain reactive, responding only to explicit user requests instead of acting proactively in the evolving multimodal environment. We present MiniCPM-o 4.5, our latest effort towards human-like multimodal interaction, which mitigates these gaps by real-time full-duplex omni-modal interaction. It can see, listen, and speak simultaneously in real-time, while also exhibiting proactive behaviors such as issuing reminders or comments based on its continuous understanding of the live scene. The key technique behind MiniCPM-o 4.5 is Omni-Flow, a unified streaming framework that aligns omni-modal inputs and outputs along a shared temporal axis. This formulation converts conventional turn-based interaction into a full-duplex, time-aligned process, enabling simultaneous perception and response and allowing proactive behavior to arise within the same framework. With a total of 9B parameters, MiniCPM-o 4.5 approaches Gemini 2.5 Flash in vision-language capabilities, delivering state-of-the-art open-source performance at its scale. It also surpasses Qwen3-Omni-30B-A3B in omni-modal understanding and delivers better speech generation, with significantly higher computation efficiency. Driven by its efficient architecture design and inference optimization, the model can perform real-time full-duplex omni-modal interaction on edge devices with less than 12GB RAM cost.

What shapes a consequential decision when human and artificial intelligence work on it together? The answer is becoming harder to see. A decision may look human-led after AI has set the frame, or appear automated while human judgment still carries decisive force. This paper offers a leadership-facing spectrum to see those relationships within a bounded mandate: Pure Human, Centaur (human-dominant, with AI in the loop), Co-equal, Minotaur (AI-dominant, with humans in the loop), and Pure AI. The spectrum asks where leadership work occurs: who frames the problem, who redirects the work, and who can answer for what follows. The five positions are landmarks that help leaders recognize configurations as they layer, drift, or change in a single decision. The central risk is misrecognition: leaders may keep a human-centered story in place after decision-shaping authority has shifted elsewhere. They may believe oversight remains meaningful when it has become ceremonial, or keep humans in the loop when their involvement could make the decision worse. The framework introduces co-adaptability, the capacity of a configuration to improve as human and non-human participants adjust together, and places it within heterogeneous teaming, where participants may vary by number, substrate, model architecture, capability, speed, memory, and form of participation. The aim is practical: to help strategic leaders and those designing or deploying AI systems recognize the configuration at work, notice when it shifts, and judge whether it fits the decision before them. These configurations will shape how power, responsibility, and trust are distributed in organizational life. Whether the futures they help create remain governable and worth inhabiting will depend on leaders who can see, early enough, where and how consequential decisions are actually being shaped.

In recent years, Multimodal Large Language Models (MLLMs) have achieved remarkable progress on a wide range of multimodal benchmarks. Despite these advances, most existing benchmarks mainly focus on single-image or multi-image comprehension. In real-world scenarios such as document reading, information is often presented as interleaved multimodel contexts. This requires MLLMs not only to recognize the content of individual images, but also to identify relevant textual and visual evidence, establish fine-grained alignments between them, and reason over these aligned signals in interleaved contexts based on contextual evidence.However, there is still a lack of systematic benchmarks for quantifying the fine-grained understanding ability of MLLMs in interleaved image-text contexts. To fill this gap, we propose COHERENCE, a benchmark designed to evaluate the ability of MLLMs to recover fine-grained image-text correspondences in interleaved multimodal contexts. COHERENCE covers interleaved image-text content from four representative domains and contains 6,161 high-quality questions. Moreover, we perform a six-type error analysis, enabling fine-grained attribution of failures in interleaved image-text understanding to the specific capabilities missing in current MLLMs.

Heterogeneous graphs with heterophily have emerged as a powerful abstraction for modeling complex real-world systems, where nodes of different types and labels interact in diverse and often non-homophilous ways. Despite recent advances, robust representation learning for such graphs remains largely unexplored, particularly in the presence of noisy or misleading connectivity. In this work, we investigate this problem and identify structural noise as a critical challenge that significantly degrades model performance. To address this issue, we propose a unified framework, Heterogeneous Graph Unified Learning (HGUL), which jointly handles heterophily and noisy graph structures. The framework consists of three complementary modules: a kNN-based graph construction module that recovers reliable local neighborhoods, a graph structure learning module that adaptively refines the adjacency by filtering noisy edges, and a heterogeneous affinity learning module that captures class-level relationships via an extended affinity matrix derived from a polynomial graph kernel. Extensive experiments on multiple datasets demonstrate that HGUL consistently outperforms existing methods on clean graphs and maintains strong robustness under varying levels of structural noise. The results further underscore the importance of jointly modeling heterophily and noise in heterogeneous graph learning.

Digital computing-in-memory (DCIM) has emerged as a promising solution for large language model (LLM) acceleration by minimizing data transfers between external DRAM and on-chip accelerators while maintaining high precision for superior accuracy. However, existing CIM architectures often overlook weight update latency, which becomes critical as LLM weights are far larger than a single CIM macro capacity. To address this issue, this paper proposes a read-compute/write (RCW) architecture that effectively minimizes weight update latency, along with a nonlinear operator fusion that further mitigates dependencyinduced latency. The proposed RCW reduces decoding computing latency by 21.59% on the Llama2-7B model. In addition, the nonlinear operator fusion mechanism achieves a 69.17% latency reduction through efficient partial accumulation and group-based approximation. Furthermore, a weight-stationary and output column stationary (WS-OCS) dataflow is introduced to reduce both external DRAM access and internal CIM weight updates by 51.6% and 87.6% respectively during the prefill phase of 1024 tokens, leading to an overall 49.76% latency reduction. Fabricated using TSMC 22 nm CMOS technology and operating at 100 MHz, the proposed RCW-CIM achieves 3.28 TOPS and 42.3 TOPS/W, enabling 4.2 ms prefill latency and 26.87 decoded tokens per second for the INT4-weight Llama2 model with dual DDR5-6400 memory.

Dialogue models are inherently reactive, responding to the current user turn without anticipating upcoming intents, which leads to redundant interactions in multi-intent settings. We address this limitation by introducing a lightweight intent-transition prior derived from dialogue data and injected into the system prompt at inference time. We instantiate this prior using a Temporal Bayesian Network (T-BN) trained on per-turn intent annotations in MultiWOZ 2.2. The T-BN achieves Recall@5 = 0.787 and MRR = 0.576 on 1,071 held-out USER-turn pairs. In a ground-truth replay over 200 dialogues, BN-guided generation improves Coverage AUC from 0.742 to 0.856 and reduces the number of turns required to reach 75% intent coverage from 3.95 to 2.73. These results show that lightweight intent-transition guidance enables more proactive and efficient dialogue behavior without modifying the underlying language model.

This paper is a continuation work of Ren et al. (2026) aiming to further devise q-learning algorithms for mean-field control (MFC) with controlled common noise. Based on the relaxed control formulation, we first establish the martingale condition of the value function and the Iq-function by evaluating along the conditional state distributions generated by all test policies. As the data in the relaxed control formulation are not observable in practice, we quantify the error incurred when they are replaced by the observable ones in the exploratory formulation under discretely sampled actions. This, together with a two-layer fixed point characterization of an optimal policy in Ren et al. (2026), allows us to propose several algorithms including the Actor-Critic q-learning algorithm, in which the policy is updated in the Actor-step based on the iteration rule induced by the improved Iq-function, and the value function and Iq-function are updated in the Critic-step based on the martingale orthogonality condition using the data from the exploratory formulation. We also establish the convergence of the inner iterations in the Actor-step in an infinite-horizon linear quadratic (LQ) framework. In two examples, within and beyond LQ framework, our q-learning algorithms are implemented with satisfactory performance.

As LLMs become credible readers of earnings calls, investor-relations Q\&A, guidance, and disclosure language, supervised financial NLP benchmarks increasingly function as decision evidence for model selection and deployment. A hidden assumption is that gold labels make such evidence objective. This assumption breaks down when the benchmark ruler itself is sensitive to rubric wording, metric choice, or aggregation policy. We study this measurement risk on Japanese Financial Implicit-Commitment Recognition (JF-ICR; a pinned 253-item test split x 4 frontier LLMs x 5 rubrics x 3 temperatures x 5 ordinal metrics). Three findings follow. First, rubric wording materially changes model-assigned labels: R2--R3 agreement ranges from 70.0% to 83.4%, with the dominant movement near the +1 / 0 implicit-commitment boundary. This pattern is consistent with a pragmatic-boundary interpretation, but is not a validated linguistic-causality claim because the present rubric variants confound semantics, examples, and verbosity. Second, not every metric remains informative under the JF-ICR class distribution. Within-one accuracy is too easy because near misses receive credit and the majority class dominates; worst-class accuracy is too noisy because the rarest class has only two examples. Exact accuracy, macro-F1, and weighted \k{appa} are therefore the identifiable metrics under our operational rule. Third, ranking claims become more defensible only after this metric-identifiability audit: Bradley--Terry, Borda, and Ranked Pairs agree on the identifiable metric subset, while the full five-metric sweep produces disagreement on the closest pair. The contribution is not a new leaderboard, but a reporting discipline for supervised financial benchmarks whose gold labels exist and whose evaluation ruler still requires governance.

This paper investigates the continuous-time counterpart of the Q-function for entropy-regularized mean-field control (MFC) with controlled common noise, coined as q-function by Jia and Zhou (2023) in the single agent's model. We first show that, under discretely sampled actions, the value function in the exploratory formulation converges to the one in the relaxed control formulation as the time grid refines. Leveraging the relaxed control formulation, we derive the exploratory Hamilton-Jacobi-Bellman (HJB) equation, in which the controlled common noise gives rise to an additional nonlinear functional of policy, rendering the policy iteration intricate. Under certain concavity condition, we establish the existence and uniqueness of the optimal one-step policy iteration via a first-order condition using the partial linear functional derivative with respect to policy. The policy improvement at each iteration is verified by relating to an entropy-regularized optimization problem over the space of policies. In the mean-field setting, we introduce the integrated q-function (Iq-function) defined on the state distribution and the policy, and it is shown that an optimal policy is identified as a two-layer fixed point to the argmax operator of the Iq-function. Finally, we provide the explicit characterization of an optimal policy as a Gaussian distribution in the general linear-quadratic (LQ) setting.

Clickbait is characterized by disproportionately high emotional intensity relative to informational content, often reinforced by specific structural patterns. However, current research considers clickbait as a static textual phenomenon characterized by linguistic patterns and structural cues. Additionally, existing detection systems primarily rely on surface-level features of clickbait. This paper introduces an emotion-aware clickbait generation attack, where stylistic transformations are used to optimize emotional impact. We propose an emotion-aware framework based on the Valence-Arousal-Dominance (VAD) space to model the emotional dynamics underlying clickbait generation for optimal user engagement. To simulate realistic attack scenarios, we align clickbait headlines with semantically similar social media posts using Sentence-BERT and generate multiple stylistic rewrites via Large Language Models (LLMs). Building on this, we define a Curiosity Gap (CG) function that computes clickbait's headline variation to the current post to quantify how emotional activation will contribute to user curiosity and evade the existing system found on social media. Experimental results demonstrate that emotion-aware stylization significantly degrades the performance of state-of-the-art classifiers, leading to misclassification rates of up to 2.58% to 30.63% on the base system.

Statistical inference in observational science typically relies on a fundamental assumption: as sample size increases and uncertainties decrease, the inferred results should converge to the true physical quantities. This assumption underpins the notion that big data lead to more reliable conclusions. In Galactic archaeology, stellar ages inferred from spectroscopic surveys are widely used to reconstruct the formation history of the Milky Way disk. The age metallicity relation (AMR) and its derived formation timescale are often regarded as key physical diagnostics of early disk evolution. This interpretation carries an implicit premise: that observational quality does not introduce systematic bias into age inference. Here we show that this premise may fail. Using a large sample of subgiant stars, we identify a region in the observational quality parameter space (signal-to-noise ratio and parallax precision) where the inferred formation timescale exhibits a systematic offset of 0.5-1 Gyr relative to an independent asteroseismic reference, while the statistical uncertainties remain small, thus producing a stable-but-wrong inference state.

Intent-based networking promises to revolutionize telecommunications network management by enabling operators to specify high-level goals rather than low-level configurations. The TM Forum Intent Ontology (tio) provides a standardized vocabulary for expressing network intents, yet lacks formal validation mechanisms to ensure intent correctness before its admission. We present tio-shacl, the first comprehensive SHACL (Shapes Constraint Language) validation framework for the TMF Intent Ontology. Our contribution includes 56 node shapes and 69 property shapes across all 15 tio v3.6.0 ontology modules, a reusable constraint library with 25 parameterized SPARQL-based constraint components, and novel validation patterns for recursive logical operators, quantity-based constraints, and cross-expectation relationships. We pursued 100% vocabulary coverage (87 classes, 109 properties, 72 functions), cross-implementation compatibility across three major SHACL engines, and validation accuracy on a corpus of 133 test cases. tio-shacl is publicly available under MIT license at https://github.com/EricssonResearch/tio-shacl and enables automated syntactic and semantic validation of network intents, addressing a critical gap in the field.

As large language model (LLM) agents are deployed in high-stakes environments, the question of how safely to delegate subtasks to specialized sub-agents becomes critical. Existing work addresses multi-agent architecture selection at design time or provides broad empirical guidelines, but neither provides a runtime mechanism that dynamically adjusts the safety-efficiency trade-off as task context changes during execution. We propose Safe Bilevel Delegation (SBD), a formal framework for runtime delegation safety in hierarchical multi-agent systems. SBD formulates task delegation as a bilevel optimization problem: an outer meta-weight network phi learns context-dependent safety-efficiency weights lambda(s) in [0,1]; an inner loop optimizes the delegation policy pi subject to a probabilistic safety constraint P(safe) >= 1-delta. The continuous delegation degree alpha in [0, 1] controls how much decision authority is transferred to each sub-agent, interpolating smoothly between full human override (alpha=0) and fully autonomous execution (alpha=1). We establish three theoretical results: (1) Safety Monotonicity--higher outer safety weight produces a weakly safer inner policy; (2) Inner Policy Convergence--projected gradient descent on the inner problem converges linearly under standard smoothness assumptions; (3) an Accountability Propagation bound that distributes responsibility across multi-hop delegation chains with a provable per-agent ceiling. We instantiate SBD in three high-stakes domains--medical AI (MIMIC-III), financial risk control (S and P 500), and educational agent supervision (ASSISTments)--specifying datasets, safety constraint sets, baselines, and evaluation protocols. This manuscript presents the formal framework and theoretical results in full; empirical validation following the protocols described herein is planned and will be reported in a forthcoming revision.

Accurate multiclass segmentation of the Circle of Willis (CoW) is essential for neurovascular disease management but remains challenging due to complex vascular topology and variable morphology. Existing deep learning methods often suffer from vascular discontinuities and inter-class misclassification, while current topological loss functions incur prohibitive computational costs in 3D multiclass settings. To address these limitations, we propose an Anatomically-Guided Topology-Aware Loss (AG-TAL) and introduce a large-scale, multi-center CoW dataset with unified annotations to facilitate robust model training. AG-TAL specifically integrates a radius-aware Dice loss to address class imbalance in small vessels, a breakage-aware clDice loss that utilizes group convolutions to efficiently preserve local connectivity, and an adjacency-aware co-occurrence loss that leverages anatomical priors to enforce distinct boundaries between neighboring arteries. Evaluated using 5-fold cross-validation, AG-TAL achieved an average Dice score of 80.85% for all CoW arteries, with small arteries notably higher by 1.05-3.09% compared to state-of-the-art methods. Across six independent datasets, the performance of AG-TAL achieved Dice scores ranging from 74.46% to 81.17% for all CoW arteries, with improvements of 2.20% to 9.98% for small arteries compared to other methods. This study demonstrates the superiority of AG-TAL in identifying multiclass CoW arteries and its ability to generalize well to multiple independent datasets. Furthermore, reliability analyses and clinical applications in an Alzheimer's disease cohort validate the AG-TAL's robustness and its potential for discovering imaging-based morphological biomarkers.

Heterogeneous graphs are widely used to model multi-relational systems, but missing node attributes remain a major bottleneck for downstream learning. In this paper, we identify and formalize type-dependent information asymmetry: the phenomenon that different node types provide substantially different levels of useful signal for attribute completion. Motivated by this observation, we propose TypeBandit, a lightweight, model-agnostic methodology for heterogeneous attribute completion. TypeBandit combines topology-aware initialization, type-level bandit sampling, and joint representation learning. It allocates a finite global sampling budget across node types, samples representative nodes within each type, and uses the resulting sampled type summaries as shared contextual signals during representation construction. By operating at the type level rather than over each target node's local neighborhood, TypeBandit keeps the adaptive state compact and practical for large heterogeneous graphs. A key advantage of TypeBandit is architectural flexibility. Rather than requiring a new heterogeneous graph neural network architecture, TypeBandit acts as a type-aware front end for representative heterogeneous GNN backbones, including R-GCN, HetGNN, HGT, and SimpleHGN. We further introduce a hybrid pretraining scheme that combines structural degree priors with feature propagation, yielding a more reliable initializer than degree-only pretraining. Under a fixed-split protocol on DBLP, IMDB, and ACM, TypeBandit provides dataset-dependent but practically meaningful gains. Additional ablation, stability, efficiency, semantic-propagation, and sampled OGBN-MAG experiments support TypeBandit as a practical strategy for heterogeneous attribute completion when type-specific information is unevenly distributed and sampling resources are limited.

Explainable AI (XAI) aims to improve user understanding and decisions when using AI models. However, despite innovations in XAI, recent user evaluations reveal that this goal remains elusive. Understanding human cognition can help explain why users struggle to effectively use AI explanations. Focusing on reasoning on structured (tabular) data, we examined various reasoning strategies for different XAI methods (none, feature importance, feature attribution) in the decision task of anticipating AI decisions (i.e., forward simulation). We i) elicited reasoning strategies from a formative user study, and ii) collected decisions from a summative user study. Using cognitive modeling, we implemented the processes underlying each reasoning strategy and evaluated their alignment with human decision-making. We found that our models better fit human decisions than baseline machine learning proxies, providing insights into which reasoning strategies are (in)effective. We then demonstrate how the fitted model can be used to form hypotheses and investigate research questions that are costly to study with real human participants. This work contributes to debugging human understanding of XAI, informing the future development of more usable and interpretable AI explanations.

Agentic large language model systems have demonstrated strong capabilities. However, their reliance on language as the universal interface fundamentally limits their applicability to many real-world problems, especially in scientific domains where domain-specific foundation models have been developed to address specialized tasks beyond natural language. In this work, we introduce Eywa, a heterogeneous agentic framework designed to extend language-centric systems to a broader class of scientific foundation models. The key idea of Eywa is to augment domain-specific foundation models with a language-model-based reasoning interface, enabling language models to guide inference over non-linguistic data modalities. This design allows predictive foundation models, which are typically optimized for specialized data and tasks, to participate in higher-level reasoning and decision-making processes within agentic systems. Eywa can serve as a drop-in replacement for a single-agent pipeline (EywaAgent) or be integrated into existing multi-agent systems by replacing traditional agents with specialized agents (EywaMAS). We further investigate a planning-based orchestration framework in which a planner dynamically coordinates traditional agents and Eywa agents to solve complex tasks across heterogeneous data modalities (EywaOrchestra). We evaluate Eywa across a diverse set of scientific domains spanning physical, life, and social sciences. Experimental results demonstrate that Eywa improves performance on tasks involving structured and domain-specific data, while reducing reliance on language-based reasoning through effective collaboration with specialized foundation models.

The increasing dependency among Filipino college students on artificial intelligence (AI) poses concerns about the potential decline of fundamental academic competencies. This study examines the extent of AI dependency and its perceived effects on students' critical thinking, writing skills, learning independence, research skills, and academic engagement. Using a cross-sectional research design, data was collected from 651 students enrolled in higher education institutions (HEIs) in Pampanga, Philippines accredited by the Commission on Higher Education. The survey data was analyzed using Latent Class Analysis (LCA) to identify AI dependency patterns. Findings indicated that students show moderate to high AI dependency, specifically in research and writing tasks. LCA identified four distinct profiles: highly engaged independent learners, selective AI users, moderate AI users, and AI-dependent learners. Notably, AI-dependent learners demonstrated the weakest academic competencies, with significant dependency on AI-generated outputs. The study highlights the need to foster educational policies that integrate AI literacy while preserving essential academic skills. HEIs must also balance technological advancements with curriculum adaptations to promote critical thinking and ethical use of AI. Future research may explore the longitudinal impacts and intervention strategies to mitigate academic skill erosion caused by AI dependency.

This study examines the factors influencing pre-service teachers' behavioral intention to use AI-enabled educational tools during their practicum, using the Unified Theory of Acceptance and Use of Technology 2 (UTAUT2) as the theoretical framework. The model includes the core UTAUT2 constructs such as performance expectancy, effort expectancy, hedonic motivation, social influence, facilitating conditions, price value, and habit. It also incorporates additional predictors including computer self-efficacy, computer anxiety, and computer playfulness. Data were collected from 563 pre-service teachers using a structured questionnaire and analyzed using Partial Least Squares Structural Equation Modeling (PLS-SEM). The results indicate that performance expectancy and hedonic motivation are the strongest predictors of behavioral intention. Computer self-efficacy, computer anxiety, and computer playfulness significantly influenced effort expectancy, although effort expectancy did not directly predict behavioral intention. Performance expectancy was significantly predicted by extrinsic motivation, job fit, relative advantage, and outcome expectations. Constructs such as social influence and facilitating conditions showed limited or inverse effects. These findings suggest that internal motivational, cognitive, and emotional factors are more influential than external or institutional factors in shaping the adoption of AI-enabled tools. The study highlights the importance of promoting personal relevance, confidence, and enjoyment in teacher preparation programs to encourage technology integration.

Human annotators frequently disagree on emotion labels, yet most evaluations of Large Language Model (LLM) emotion annotation collapse these judgments into a single gold standard, discarding the distributional information that disagreement encodes. We ask whether LLMs capture the structure of this disagreement, not just majority labels, by comparing emotion judgment distributions between human annotators and four zero-shot LLMs, plus a fine-tuned RoBERTa baseline, across two complementary benchmarks: GoEmotions and EmoBank, totaling 640,000 LLM responses. Zero-shot models diverge substantially from human distributions, and in-domain fine-tuning, not model scale, is required to close the gap. We formalize a lexical-grounding gradient through a quantitative transparency score that predicts per-category human--LLM agreement: LLMs reliably capture emotions with explicit lexical markers but systematically fail on pragmatically complex emotions requiring contextual inference, a pattern that replicates across both categorical and continuous emotion frameworks. We further propose three lightweight post-hoc calibration methods that reduce the distributional gap by up to 14\%, and provide actionable guidelines for when LLM emotion annotations can, and cannot, substitute for human labeling.

Compositional generalization tests are often used to estimate the compositionality of LLMs. However, such tests have the following limitations: (1) they only focus on the output results without considering LLMs' understanding of sample compositionality, resulting in explainability defects; (2) they rely on dataset partition to form the test set with combinations unseen in the training set, suffering from combination leakage issues. In this work, we propose a novel rule-generation perspective for compositionality estimation for LLMs. It requires LLMs to generate a program as rules for dataset mapping and provides estimates of the compositionality of LLMs using complexity-based theory. The perspective addresses the limitations of compositional generalization tests and provides a new way to analyze the compositionality characterization of LLMs. We conduct experiments and analysis of existing advanced LLMs based on this perspective on a string-to-grid task, and find various compositionality characterizations and compositionality deficiencies exhibited by LLMs.

Context: Documenting Architectural Design Decisions (ADDs) is a critical factor in the software lifecycle, essential for efficient system maintenance, developer onboarding, and preventing knowledge vaporization. Although various templates for Architectural Decision Records (ADRs) have been proposed, there is a lack of empirical evidence comparing them. Goal: To address this gap, this paper aims to identify which ADR template best supports comprehension, usability, and ease of adoption: Tyree/Akerman's template, Nygard's ADR, arc42, Y-statements, and MADR. Method: We compared these templates using the DESMET FA method in a two-step evaluation. First, the two primary authors evaluated the five templates through the DESMET FA, based on their software architecture expertise. The two top-performing templates were then used as treatments in a controlled experiment conducted with undergraduate students. Results: In the preliminary screening by experts, the top-performing templates were those of Nygard and MADR. In the subsequent controlled experiment, Nygard's template outperformed MADR in terms of the Overall Score. Qualitative analysis of participant feedback revealed the factors influencing template preference. The findings indicate that Nygard supports concise and objective documentation, while MADR facilitates structural details and specific architectural requirements. Conclusion: This paper provides an evidence-based strategy for ADR template adoption by offering a comparison between them. The findings present a decision-making guide that assists practitioners and researchers in selecting ADR templates aligned with project constraints, aiming to minimize documentation overhead and increase architectural knowledge retention.

EigenDecomposition (ED) is at the heart of many computer vision algorithms and applications. One crucial bottleneck limiting its usage is the expensive computation cost, particularly for a mini-batch of matrices in deep neural networks. Our previous work proposed a dedicated QR-based ED algorithm for batched small matrices (dim${<}32$). This short paper targets the limitation and proposes a batch-efficient Divide-and-Conquer based ED algorithm for larger matrices. The numerical test shows that for a mini-batch of matrices whose dimensions are smaller than $64$, our method can be much faster than the Pytorch SVD function.

Security Operations Centers (SOCs) face mounting operational challenges. These challenges come from increasing threat volumes, heterogeneous SIEM platforms, and time-consuming manual triage workflows. We present an end-to-end threat management framework that integrates ensemble-based detection, syntax-constrained query generation, and retrieval-augmented resolution support to automate critical security workflows. Our detection module evaluates both traditional machine learning classifiers and large language models (LLMs), then combines the three best-performing LLMs to create an ensemble model, achieving 82.8% accuracy while maintaining 0.120 false positive rate on SIEM logs. We introduce the SQM (Syntax Query Metadata) architecture for automated evidence collection. It uses platform-specific syntax constraints, metadata-based retrieval, and documentation-grounded prompting to generate executable queries for IBM QRadar and Google SecOps. SQM achieves a BLEU score of 0.384 and a ROUGE-L score of 0.731. These results are more than twice as good as the baseline LLM performance. For incident resolution and recommendation generation, we demonstrate that integrating SQM-derived evidence improves resolution code prediction accuracy from 78.3% to 90.0%, with an overall recommendation quality score of 8.70. In production SOC environments, our framework reduces average incident triage time from hours to under 10 minutes. This work demonstrates that domain-constrained LLM architectures with retrieval augmentation can meet the strict reliability and efficiency requirements of operational security environments at scale.

Large Language Models (LLMs) have achieved remarkable progress in recent years, driving their adoption across a wide range of domains, including computer security. In reverse engineering, LLMs are increasingly applied to critical tasks such as function and variable name recovery and type inference. However, despite the rapid growth of research in this area, progress has been hindered by the absence of a standardized dataset. Existing studies rely on disparate datasets, preprocessing pipelines, and evaluation metrics, making fair comparisons between approaches difficult and obscuring a clear understanding of LLM capabilities in binary analysis. To address these challenges, we present REBench, a comprehensive benchmark dataset for evaluating LLMs on binary reverse engineering tasks. REBench consolidates a superset of existing datasets, comprising hundreds of millions of lines of source code and a diverse collection of binaries spanning multiple architectures and optimization levels. REBench adopts a knowledge-base-driven methodology that stores byte-level stack information to generate ground truth, ensuring that task difficulty is preserved while maintaining universal applicability. This design enables fair evaluation across tasks while avoiding simplifications that could bias results. As a use case, we apply REBench to measure the reverse engineering performance of LLMs and the result demonstrates difficulties in complex tasks.

Operational weather prediction has long relied on physics-based numerical weather prediction (NWP), whose accuracy comes at the cost of substantial compute and complex simulation workflows. Recent transformer-based forecasters offer efficient data-driven alternatives, however transformers are physics-agnostic models. Additionally, standard transformer encoders evolve representations through discrete layer updates that may be less suited to modeling smooth latent dynamics. In this work, we propose a continuous-depth transformer encoder for weather forecasting that integrates Neural Ordinary Differential Equation (Neural ODE) dynamics within each encoder block. Specifically, we replace discrete residual updates with ODE-based updates solved using adaptive numerical integration. We also introduce a two-branch attention module that combines conventional patch-wise self-attention with an auxiliary branch that applies a derivative operator to attention logits, providing an additional change-sensitive interaction signal. To further align forecasts with governing principles, we propose a customized physics-informed training objective that enforces physical consistency as a soft constraint. We evaluate the proposed method against a standard discrete transformer baseline and an existing continuous-time Neural ODE forecasting variant, demonstrating the importance of PINN-Cast in short term weather forecasting.

The advent of Large Language Models (LLMs) has significantly transformed tasks across Software Engineering. In the context of Business Process Management, LLMs are now being explored as tools to derive process models directly from textual descriptions. Existing approaches range from chatbot-driven systems that assist with iterative, text-based modeling to fully automated end-to-end modeling assistants. However, we argue that process modeling is inherently complex and cannot be effectively addressed through black-box solutions. Instead, we envision modeling as an open-ended conversational activity, best supported by an interactive, iterative process involving both humans and LLM. In our approach, the modeling task is decomposed into smaller, manageable steps. Each step results in intermediate artifacts and explicitly documents the rationale behind each modeling decision. During this process, we incrementally uncover simple behavioral relations that guide the construction of the model. Given the current limitations of LLMs in reasoning about complex dependencies, we complement them with specialized tools developed in the field to structure process models based on behavioral relations. This hybrid approach enables the generation of sound, yet comprehensible models that evolve through transparent and explainable steps. In this paper, we present our research agenda and introduce Pragmos, a prototype system that operationalizes this vision. Pragmos demonstrates how LLMs can collaborate with human users as both domain and modeling experts to co-create evolving process models through a structured and explainable workflow.

Clinical AI systems require not just point-in-time evaluation but continuous governance: the ongoing practice of monitoring, evaluating, iterating, and re-evaluating performance throughout deployment. We present an end-to-end framework of governance that integrates rubric validation, live deployment feedback, technical performance monitoring, and cost tracking, with controlled experimentation gating system changes before deployment. Applied to Hyperscribe, an EHR-embedded agent that converts ambient audio into structured chart updates, twenty clinicians authored 1,646 validated rubrics across 823 cases. Seven Hyperscribe versions were evaluated through controlled experiments, with median scores improving from 84% to 95%. Analysis of 107 live feedback entries over three months showed feedback composition shifting from 79% error reports and 14% positive observations to 30% errors and 45% positive observations as engineering interventions resolved failures. Median processing time per audio segment was 8.1 seconds with a 99.6% effective completion rate after retry mechanisms absorbed transient model errors. These results demonstrate that continuous, multi-channel governance of deployed clinical AI is both achievable and effective.

Parameter-efficient fine-tuning (PEFT) methods face a tradeoff between adapter size and expressivity: ultra-low-parameter adapters are confined to fixed low-rank subspaces, capping performance even with extended training. We propose BoostLoRA, a gradient-boosting framework that overcomes this limit by iteratively training and merging minimal adapters on the examples the current model gets wrong. A ROTATE SVD basis strategy assigns each round to an orthogonal subspace, so cumulative effective rank grows linearly with the number of rounds while each adapter remains ultra-low-rank. After merging, adapters are discarded, leaving zero inference overhead. On Qwen2.5-3B, BoostLoRA reaches 89.1% on GSM8K and 68.8% on MATH-500, surpassing both the best single-shot ultra-low parameter adapter (TinyLoRA) and full fine-tuning; on code generation it reaches 57.2% on MBPP and 80.4% on HumanEval while full fine-tuning drops below the zero-shot baseline. We also demonstrate cross-architecture transfer on protein binding classification with ESM2-650M and cross-entropy training. BoostLoRA is, to our knowledge, the first PEFT method whose effective rank grows with training, separating per-round parameter cost from total representational capacity.

Causal inference is essential for data-driven decision-making, as it aims to uncover causal relationships from observational data. However, identifying causality remains challenging due to the potential for confounding and the distinction between correlation and causation. While recent advances in causal machine learning and matching algorithms have improved estimation accuracy, these methods often face trade-offs between interpretability and computational efficiency. This paper proposes a novel approach that combines a tree-based discretization technique, tailored for causal inference, with an integer linear programming-based matching algorithm. The discretization ensures approximately linear relationships for control datasets within strata, enabling effective matching, while the optimization framework optimizes for global balance. The resulting algorithm yields computational efficiency and less biased ATT estimates compared to state-of-the-art algorithms. Empirical evaluations demonstrate the proposed method's practical advantages over existing techniques in causal inference scenarios.

Metamaterial discovery seeks microstructured materials whose geometry induces targeted mechanical behavior. Existing inverse-design methods can efficiently generate candidates, but they typically require explicit numerical property targets and are less suitable for early-stage exploration, where researchers often begin with incomplete constraints and qualitative intents expressed in natural language. Large language models can interpret such intents, but they lack geometric awareness and physical property validity. To address this gap, we propose MetaSymbO, a multi-agent framework for language-guided Metamaterial discovery via Symbolic-driven latent evOlution. Specifically, MetaSymbO contains three agents: a Designer that interprets free-form design intents and retrieves a semantically consistent scaffold, a Generator that synthesizes candidate microstructures in a disentangled latent space, and a Supervisor that provides fast property-aware feedback for iterative refinement. To move beyond the limitations of reproducing known samples from literature and training data, we further introduce symbolic-driven latent evolution, which applies programmable operators over disentangled latent factors to compose, modify, and refine structures at inference time. Extensive experiments demonstrate that (i) MetaSymbO improves structural validity by up to 34% in symmetry and nearly 98% in periodicity compared to state-of-the-art baselines; (ii) MetaSymbO achieves about 6-7% higher language-guidance scores while maintaining superior structure novelty compared to advanced reasoning LLMs; (iii) qualitative analyses confirm the effectiveness of symbolic logic operators in enabling programmable semantic alignment; and (iv) realworld case studies on auxetic, high-stiffness metamaterial design further validate its practical capability.

Deriving governing equations from empirical observations is a longstanding challenge in science. Although artificial intelligence (AI) has demonstrated substantial capabilities in function approximation, the discovery of explainable and extrapolatable equations remains a fundamental limitation of modern AI, posing a central bottleneck for AI-driven scientific discovery. Here, we present machine collective intelligence, a unified paradigm that integrates two fundamental yet distinct traditions in computational intelligence--symbolism and metaheuristics--to enable autonomous and evolutionary discovery of governing equations. It orchestrates multiple reasoning agents to evolve their symbolic hypotheses through coordinated generation, evaluation, critique, and consolidation, enabling scientific discovery beyond single-agent inference. Across scientific systems governed by deterministic, stochastic, or previously uncharacterized dynamics, machine collective intelligence autonomously recovered the underlying governing equations without relying on hand-crafted domain knowledge. Furthermore, the resulting equations reduced extrapolation error by up to six orders of magnitude relative to deep neural networks, while condensing 0.5-1 million model parameters into just 5-40 interpretable parameters. This study marks an important shift in AI toward the autonomous discovery of principled scientific equations.

Large Language Models (LLMs) are increasingly used for code editing, yet the prevalent full-code generation paradigm suffers from severe efficiency bottlenecks, posing challenges for interactive coding assistants that demand low latency and cost. Despite the predominant focus on scaling model capabilities, the edit format itself has been largely overlooked in model training. In this paper, we begin with a systematic study of conventional diff formats and reveal that fragile offsets and fragmented hunks make generation highly unnatural for LLMs. To address it, we introduce BlockDiff and FuncDiff, two structure-aware diff formats that represent changes as block-level rewrites of syntactically coherent units such as control structures and functions. Furthermore, we propose AdaEdit, a general adaptive edit strategy that trains LLMs to dynamically choose the most token-efficient format between a given diff format and full code. Extensive experiments demonstrate that AdaEdit paired with structure-aware diff formats consistently matches the accuracy of full-code generation, while reducing both latency and cost by over 30% on long-code editing tasks.

Learning rate scheduling has evolved from the single global fixed rate of early SGD to sophisticated layer-wise adaptive strategies. We systematize this evolution into five generations: (Gen1) global fixed learning rates, (Gen2) global scheduling, (Gen3) parameter-level adaptation, (Gen4) layer-level differentiation, and (Gen5) joint layer-time scheduling. We trace the fundamental motivation behind each transition, showing how the shift from one-size-fits-all to tailoring by layer and time addresses the impossible trinity of transfer learning: lower layers require small updates to preserve general knowledge while higher layers need large updates to adapt to new tasks. Building on this taxonomy, we propose Discriminative Adaptive Layer Scaling (DALS), a unified framework that integrates phase-adaptive cosine scheduling, depth-aware Grokfast gradient filtering, and LARS-style trust ratios into a single coherent optimizer. We benchmark 18 strategies including three DALS variants across all five generations on five datasets: synthetic, CIFAR-10 (from scratch), RTE, TREC-6, and IMDb (fine-tuning). On synthetic, DALS achieves the best accuracy at 98.0%, while DALS-Fast reaches 90% in just 3 epochs. The cross-dataset analysis reveals striking regime-dependent patterns--no single strategy wins across all regimes. Critically, STLR+Discriminative, the ULMFiT champion, catastrophically fails on from-scratch tasks (43.6% on TREC-6 from scratch vs. 96.8% with RAdam), confirming that directional decay biases are harmful without pretrained features. DALS avoids either extreme, achieving the best synthetic result while maintaining competitive fine-tuning performance.

Every system that performs effects has two boundaries: what it can do (expressiveness) and what governance covers (governance). In nearly all deployed AI systems, these boundaries are defined independently, creating three regions: governed capabilities (the only useful region), ungoverned capabilities (risk), and governance policies that address non-existent capabilities (theater). Two of the three regions are failure modes. We focus on the governance of effects: actions that AI systems perform in the world (API calls, database writes, tool invocations). This is distinct from the governance of model outputs (content quality, bias, fairness), which operates at a different level and requires different mechanisms. We present a formal framework for analyzing this structural gap. Rice's theorem (1953) proves the gap is undecidable in the general case for any Turing-complete architecture that attempts to govern effects behaviorally: no algorithm can decide non-trivial semantic properties of arbitrary programs, including the property "this program's effects comply with the governance policy." We define coterminous governance: a system property where the expressivenessboundary equals the governance boundary. We show that coterminous governance requires an architectural decision (separatingcomputation from effect) rather than a governance layer added after the fact. We show that structural governance under this separation subsumes separate governance infrastructure: governance checks become part of the execution pipeline rather than a second system running alongside it. We propose coterminous governance as the testable criterion for any AI governance system: either the two boundaries are provably identical, or risk and theater are structurally inevitable. Proofs are mechanized in Coq (454 theorems, 36 modules, 0 admitted).

We present five results in the theory of structural governance for cognitive workflow systems. Three are mechanized in Coq 8.19 using the Interaction Trees library with parameterized coinduction; two are proved on paper with explicit reductions. The Coinductive Safety Predicate (gov_safe) is a coinductive property that captures governance safety for infinite program behaviors, indexed by a boolean permission flag that is provably false for ungoverned I/O and true for governed interpretations (mechanized). The Governance Invariance Theorem establishes that governance is uniform across the meta-recursive tower: governance at level n+1 reduces to governance at level n by definitional equality of the type (mechanized). The Sufficiency Theorem proves that four atomic primitives (code, reason, memory, call) are expressively complete for any discrete intelligent system, formalized as compositional closure of a Kleisli category (mechanized). The Alternating Normal Form provides a canonical decomposition of any machine into alternating code and effect layers, with a confluent rewriting system (paper proof). The Necessity Theorem proves via explicit reduction to Rice's theorem that an architecturally opaque component (the reason primitive) is mathematically necessary for problems requiring semantic judgment (paper proof). A sixth contribution connects the abstract model to the deployed runtime: the Verified Interpreter Specification formalizes the BEAM runtime's trust, capability, and hash chain logic in Coq, then tests the running system against this specification using property-based testing with over 70,000 randomly generated directive sequences and zero disagreements. The mechanization comprises approximately 12,000 lines across 36 modules with 454 theorems and zero admitted lemmas.

Large language model (LLM)-based coding agents increasingly rely on external memory to reuse prior debugging experience, repair traces, and repository-local operational knowledge. However, retrieved memory is useful only when the current failure is genuinely compatible with a previous one; superficial similarity in stack traces, terminal errors, paths, or configuration symptoms can lead to unsafe memory injection. This paper reframes issue-memory use as a selective, risk-sensitive control problem rather than a pure top-k retrieval problem. We introduce RSCB-MC, a risk-sensitive contextual bandit memory controller that decides whether an agent should use no memory, inject the top resolution, summarize multiple candidates, perform high-precision or high-recall retrieval, abstain, or ask for feedback. The system stores reusable issue knowledge through a pattern-variant-episode schema and converts retrieval evidence into a fixed 16-feature contextual state capturing relevance, uncertainty, structural compatibility, feedback history, false-positive risk, latency, and token cost. Its reward design penalizes false-positive memory injection more strongly than missed reuse, making non-injection and abstention first-class safety actions. In deterministic smoke-scale artifacts, RSCB-MC obtains the strongest non-oracle offline replay success rate, 62.5%, while maintaining a 0.0% false-positive rate. In a bounded 200-case hot-path validation, it reaches 60.5% proxy success with 0.0% false positives and a 331.466 microseconds p95 decision latency. The results show that, for coding-agent memory, the key question is not only which memory is most similar, but whether any retrieved memory is safe enough to influence the debugging trajectory.

Pretrial risk assessment tools are used on over one million U.S. defendants each year, yet their use for predicting rare violent re-offense faces a basic statistical barrier. We derive a universal precision bound -- the Likelihood Ratio Wall -- showing that when violent re-arrest rates are low (2-5%), achieving even a 50% hit rate among people labeled "high risk" (positive predictive value, or PPV) would require tools far more discriminative than current instruments appear to be. For rare outcomes, a tool can have respectable-looking performance metrics and still be wrong most of the time it flags someone as "high risk for violence." We show that post-hoc score recalibration cannot solve this problem because it does not improve the tool's underlying ability to separate true positives from false positives. We further prove a Surveillance Ceiling: when over-policing inflates recorded "risk factors" among those who would not re-offend, the maximum achievable precision is structurally lower for over-policed groups, even at equal offense rates. We translate these results into the Number Needed to Detain (how many people must be detained to prevent one violent offense), and propose that risk reports should communicate this uncertainty explicitly. Our findings suggest that for rare violent outcomes, debates about fairness metrics alone are incomplete: under current data regimes, the available features may not support high-confidence individualized detention decisions.

Nonstationary Gaussian processes (GPs) are essential for modeling complex, locally heterogeneous spatial data. A common modeling approach is the spatial deformation method that warps the domain to recover isotropy. However, this static method does not account for changes in spatial correlation induced by covariates, limiting its ability to predict nonstationary GPs under new covariate conditions. To enable predictive modeling of the deformation method, we propose to model the spatial deformation as a function of covariates. The spaces of diffeomorphic deformations and Euclidean covariate vectors are connected by characterizing deformations as generated by velocity fields living in a Lie algebra. To overcome the estimation instability caused by high-order interactions between multiple covariates in a general Lie algebra, we prove that those interactions can be truncated with a moderate physical assumption. Based on the theoretical results, a concise functional form of deformations driven by multiple covariates can be established, and an efficient estimation-inference algorithm is developed for out-of-sample nonstationary GP prediction with limited covariate-deformation sample pairs. The effectiveness and generalizability of the method are demonstrated on a simulation study and two case studies, in the fields of manufacturing and geostatistics, respectively.

Audio-based stuttering systems to date have been trained for detection -- what disfluency is present now -- leaving prediction, the capability needed for closed-loop intervention, unstudied at deployable scale. We train a 616K-parameter CNN on SEP-28k (Apple, 20,131 three-second clips) to predict whether the next contiguous clip contains any disfluency. (1) Severity-selective precursor signal: on the episode-grouped test set, aggregate preblock AUC is modest (0.581 [0.542, 0.619]), but stratifying by upcoming event type reveals concentration on clinically severe events -- blocks 0.601 [0.554, 0.651] and sound repetitions 0.617 [0.567, 0.667] both exclude chance, while fillers (0.45) and word repetitions (0.49) are at chance. The aggregate objective converges to a severity-selective predictor because severe events carry prosodic precursors; fillers do not. (2) Cross-population transfer: without fine-tuning, the same checkpoint applied to 1,024 pediatric Children-Who-Stutter utterances (FluencyBank Teaching) attains AUC 0.674 detection and 0.655 prediction; DisfluencySpeech and LibriStutter reach 0.58-0.60 AUC. (3) Deployable on-device: lossless export to CoreML (1.19 MB), ONNX (40 KB), TFLite. Neural-Engine latency per 3 s window: 0.25 ms (iPhone 17 Pro Max, A19 Pro) to 0.55 ms (iPhone SE 3rd-gen and M1 Max). A 4 Hz streaming simulation uses 0.54% of the real-time budget. Platt-calibrated outputs (test ECE 0.010, from 0.177 raw). Five negative ablations -- output-level Future-Guided Learning, multi-clip GRU, time-axis concatenation, asymmetric focal loss, direct block-targeted training -- none improved over the vanilla baseline.

Brain MRI underpins a wide range of neuroscientific and clinical applications, yet most learning-based methods remain task-specific and require substantial labeled data. Here we show that a single self-supervised representation can generalize across heterogeneous brain MRI endpoints. We trained BrainDINO, a self-distilled foundation model, on approximately 6.6 million unlabeled axial slices from 20 datasets encompassing broad variation in population, disease, and acquisition setting. Using a frozen encoder with lightweight task heads, BrainDINO supported transfer across tumor segmentation, neurodegenerative and neurodevelopmental conditions classification, brain age estimation, post-stroke temporal prediction, molecular status prediction, MRI sequence classification, and survival modeling. Across tasks and supervision regimes, BrainDINO consistently equaled or exceeded natural-image and MRI-specific self-supervised baselines, with particularly strong advantages under label scarcity. Representation analyses further showed anatomically organized and pathology-sensitive feature structure in the absence of task-specific supervision. Our findings indicate that large-scale slice-wise self-supervised learning can yield a unified brain MRI representation that supports diverse neuroimaging tasks without volumetric pretraining or full-network fine-tuning, establishing a scalable foundation for robust and data-efficient brain imaging analysis.

Large language model (LLM) reading assistants are increasingly used in settings that require interpretation rather than simple retrieval. In these contexts, the central risk is not only error or unsafe output, but interpretive displacement: the transfer of meaning-making work from reader to system. This paper examines that problem through the concept of epistemic guardrails, defined here as constraints on how an artificial intelligence (AI) system participates in reading and interpretation. Using TextWalk, a minimal reading-support prototype designed as a co-reader rather than an answer-provider, the study applies a fixed ten-prompt protocol to twelve analytical texts spanning four categories of argumentative prose. The protocol escalates from baseline reading support to interpretive inquiry, boundary stress, and explicit shortcut pressure, enabling guardrails to be examined as behavioral properties observable in interaction rather than as static instruction features. Results show strong baseline stability, measurable strain during interpretive inquiry, partial recovery under direct boundary stress, and late-stage stabilization under escalation pressure. The most consequential weaknesses did not appear as overt collapse, but in a middle zone between support and substitution, where the system remained grounded and pedagogical while redistributing too much interpretive labor away from the reader. The paper contributes a protocol for evaluating epistemic guardrails as interactional phenomena in conversational AI reading assistants, an empirical account of their behavioral dynamics under pressure, and an emerging model of interpretive boundary function in reading-support AI.

As AI transitions toward multi-agent systems (MAS) to solve complex workflows, research paradigms operate on the axiomatic assumption that agent collaboration mirrors the "Wisdom of the Crowd". We challenge this assumption by formalizing the Consensus Paradox: a phenomenon where agentic swarms prioritize internal architectural agreement over external logical truth. Through a 36 experiments encompassing 12,804 trajectories across three state-of-the-art (SOTA) benchmarks (GAIA, Multi-Challenge, and SWE-bench), we prove the Inverse-Wisdom Law: in kinship-dominant swarms, adding logical agents increases the stability of erroneous trajectories rather than the probability of truth. The introduction of additional logical audits converges the system toward a Logic Saturation where internal entropy hits zero while factual error hits unity. By evaluating the interaction between the 3 preeminent SOTA models (Gemini 3.1 Pro, Claude Sonnet 4.6, and GPT-5.4), we establish the Architectural Tribalism Asymmetry as a mechanistic law of transformer weights. We demonstrate that terminal swarm integrity is strictly gated by the synthesizer's receptive logic, rather than aggregate agent quality. We define the Tribalism Coefficient and the Sycophantic Weight as the primary mechanistic determinants of swarm failure. Finally, we establish the Heterogeneity Mandate as a foundational safety requirement for resilient agentic architectures.

Large language models (LLMs) conventionally process structured inputs as 1D token sequences. While natural for prose, such linearization may introduce additional representational burden for tasks whose computation depends directly on explicit 2D structure, because row--column alignment and local neighborhoods are no longer directly expressed in the input. We study this setting, which we refer to as serialization friction, on a small diagnostic testbed of synthetic tasks with explicit 2D structure: matrix transpose, Conway's Game of Life, and LU decomposition. To examine this question, we compare a text-only language pathway over serialized inputs with a vision-augmented pathway, built on the same language backbone, that receives the same underlying content rendered in task-faithful 2D layout, yielding a system-level comparison between two end-to-end input pathways. Across the tasks and settings we study, the visual pathway consistently outperforms the textual pathway; the gap often widens at larger dimensions, and error patterns under serialization become increasingly spatially structured. These findings indicate that the relationship between input representation and model performance on such tasks warrants further investigation, and suggest that preserving task-relevant 2D layout is a promising direction for structured 2D tasks.

Biomedical knowledge graphs (KGs) are widely used in the life sciences, yet many are derived from unstructured documents and therefore lack schema-level constrains, whereas graphs assembled from structured resources are difficult to harmonize into a unified representation. We present OptimusKG, a multimodal biomedical labeled property graph (LPG) built from structured and semi-structured resources to preserve factual, type-specific metadata across molecular, anatomical, clinical, and environmental domains. OptimusKG contains 190,531 nodes across 10 entity types, 21,813,816 edges across 26 relation types, and 67,249,863 property instances encoding 110,276,843 values across 150 distinct property keys, derived from 18 ontologies and controlled vocabularies. The graph enforces a top-level schema for nodes and edges and retains granular, type-specific properties, cross-references, and provenance across molecular, anatomical, clinical, and environmental domains. We assessed the validity of OptimusKG by evaluating whether graph relationships are supported by evidence from the scientific literature using a multimodal agent, PaperQA3. PaperQA3 identified supporting evidence for 70.0% of sampled edges, whereas 83.4% of sampled false edges received no supporting evidence. Edges without literature support were concentrated in associations derived from experimental and functional genomics resources, suggesting that OptimusKG captures biomedical knowledge that may precede synthesis in the scientific literature. OptimusKG is distributed as Apache Parquet files, providing a standardized resource for graph-based machine learning, knowledge-grounded retrieval with large language models, and biomedical discovery use cases such as hypothesis generation.

As large language models are integrated into autonomous robotic systems for task planning and control, compromised inputs or unsafe model outputs can propagate through the planning pipeline to physical-world consequences. Although prior work has studied robotic cybersecurity, adversarial perception attacks, and LLM safety independently, no existing study traces how these threat categories interact and propagate across trust boundaries in a unified architectural model. We address this gap by modeling an LLM-enabled autonomous robot in an edge-cloud architecture as a hierarchical Data Flow Diagram and applying STRIDE-per-interaction analysis across six boundary-crossing interaction points using a three-category taxonomy of Conventional Cyber Threats, Adversarial Threats, and Conversational Threats. The analysis reveals that these categories converge at the same boundary crossings, and we trace three cross-boundary attack chains from external entry points to unsafe physical actuation, each exposing a distinct architectural property: the absence of independent semantic validation between user input and actuator dispatch, cross-modal translation from visual perception to language-model instruction, and unmediated boundary crossing through provider-side tool use. To our knowledge, this is the first DFD-based threat analysis integrating all three threat categories across the full perception-planning-actuation pipeline of an LLM-enabled robotic system.

This paper introduces AutoREC, an open-source Python package for developing reinforcement learning (RL) agents to automatically generate equivalent circuit models (ECMs) from electrochemical impedance spectroscopy (EIS) data. While ECMs are a standard framework for interpreting EIS data, traditional identification is typically based on manual trial-and-error, which requires domain experts and limits scalability, particularly in autonomous experimental pipelines such as self-driving laboratories. AutoREC addresses this challenge by formulating ECM construction as a sequential decision-making problem within a Markov Decision Process framework. It implements a Double Deep Q-Network with prioritized experience replay, along with a dedicated dead-loop mitigation strategy, to efficiently explore a complex action space for circuit generation. To demonstrate the capabilities of the platform, we trained an RL agent using AutoREC and evaluated its strengths and limitations across diverse datasets, while also discussing possible strategies to mitigate these limitations in future agent designs. The trained agent achieved a success rate exceeding $99.6\%$ on synthetic datasets and demonstrated strong generalization to unseen experimental EIS data from batteries, corrosion, oxygen evolution reaction, and CO$_2$ reduction systems. These results position AutoREC as a promising platform for adaptive and data-driven ECM generation, with potential for integration into automated electrochemical workflows.

Autonomous agents can adapt their behaviour to changing environments, but remain bound to requirements, goals, and capabilities fixed at design time, preventing genuine software evolution. This paper introduces self-evolving software agents, combining BDI reasoning with LLMs to enable autonomous evolution of goals, reasoning, and executable code. We propose a BDI-LLM architecture in which an automated evolution module operates alongside the agent's reasoning loop, eliciting new requirements from experience and synthesizing corresponding design and code updates. A prototype evaluated in a dynamic multi-agent environment shows that agents can autonomously discover new goals and generate executable behaviours from minimal prior knowledge. The results indicate both the feasibility and current limits of LLM-driven evolution, particularly in terms of behavioural inheritance and stability.

Subword tokenization is an essential part of modern large language models (LLMs), yet its specific contributions to training efficiency and model performance remain poorly understood. In this work, we decouple the effects of subword tokenization by isolating them within a controlled byte-level pretraining pipeline. We formulate and test hypotheses across various dimensions, including sample throughput, vocabulary scaling, and the linguistic prior of subword boundaries. By simulating these effects in a byte-level setting, we refine our understanding of why subword models outperform raw byte models and offer insights to improve the pretraining of future byte-level and subword models. Specifically, our experiments highlight the critical role of increased training throughput and the integration of subword boundaries as either explicit priors or inductive biases.

Time-series classification (TSC) has advanced significantly with deep learning, yet most models rely solely on raw numerical inputs, overlooking alternative representations. While texture-based encodings such as Gramian Angular Fields (GAF) and Recurrence Plots (RP) convert time series into 2D images, they often require heavy preprocessing and yield less intuitive representations. In contrast, chart-based visualizations offer more interpretable alternatives and show promise in specific domains; however, their effectiveness remains underexplored, with limited systematic evaluation across chart types, visual encoding choices, and datasets. In this work, we introduce VTBench, a systematic and extensible framework that re-examines TSC through multimodal fusion of raw sequences and chart-based visualizations. VTBench generates lightweight, human-interpretable plots -- line, area, bar, and scatter, providing complementary views of the same signal. We develop a modular architecture supporting multiple fusion strategies, including single-chart visual-numerical fusion, multi-chart visual fusion, and full multimodal fusion with raw inputs. Through experiments across 31 UCR datasets, we show that: (1) chart-only models are competitive in selected settings, particularly on smaller datasets; (2) combining multiple chart types can improve accuracy by capturing complementary visual cues; and (3) multimodal models improve or maintain performance when visual features provide non-redundant information, but may degrade accuracy when they introduce redundancy. We further distill practical guidelines for selecting chart types, fusion strategies, and configurations. VTBench establishes a unified foundation for interpretable and effective multimodal time-series classification.

We present \textsc{dm-PhiSNet}, a physically constrained \textsc{PhiSNet}-based equivariant model that predicts one-electron reduced density matrices (1-RDMs) directly from molecular geometries in an atomic-orbital (AO) basis for accelerated self-consistent field (SCF) workflows. Training follows a two-stage schedule with progressively introduced physically motivated objectives, and the resulting predictions are refined by a lightweight analytic block. This block enforces electron-number conservation, drives the 1-RDM toward generalized idempotency in the AO metric, and regularizes the occupation spectrum of the Löwdin-orthogonalized density. Across six closed-shell systems -- H$_2$O, CH$_4$, NH$_3$, HF, ethanol, and NO$_3^-$ -- the refined 1-RDMs provide SCF initial guesses that substantially reduce iteration steps by 49--81\% relative to standard initializations. Beyond SCF acceleration, the learned 1-RDMs yield accurate one-shot total energies and Hellmann--Feynman atomic forces without force supervision, indicating that the model captures chemically meaningful electronic structure. These results demonstrate that combining equivariant learning with analytic constraint enforcement provides a simple, general route to solver-ready density-matrix initializations and accelerated SCF workflows.

Recent advances in multimodal large language models (LLMs) have revolutionized web agents that can automate complex tasks on websites. However, their accuracy remains limited by the scarcity of high-quality web trajectory training data. Existing automatic trajectory generation methods suffer from incomplete website coverage due to homepage-based task proposals or random-walk exploration. Such methods often result in hallucinated or ambiguous task synthesis that lead to incomplete and unreliable trajectory generation. Here, we present AutoSurfer, a comprehensive web trajectory generator that addresses these limitations through three key innovations. First, AutoSurfer employs a systematic breadth-first exploration strategy that maintains a queue of discovered pages and action traces, propagates knowledge across pages to avoid redundant exploration, and recursively expands multi-level graphical user interface elements - closely resembling how a human would learn a new website. Second, AutoSurfer leverages the exploration trajectory to guide task synthesis, reducing hallucinations by grounding complex tasks in actual navigation paths rather than isolated actions or page content alone. Third, AutoSurfer uses the same exploration trajectory as hints to steer a web agent toward more accurate and reliable trajectory refinement. Together, these innovations enable AutoSurfer to comprehensively cover a website's action space and generate data suitable for training website-specific LLMs. We evaluate AutoSurfer on the WebArena benchmark by fine-tuning Qwen2.5-VL-7B-Instruct and demonstrate that it outperforms state-of-the-art methods - Explorer, OS-Genesis, and SynthAgent - achieving up to 24.23% overall task completion accuracy compared to 19.59% for the best prior method. Further, task diversity analysis demonstrates that AutoSurfer yields a more diverse distribution of synthesized tasks.

Large Language Models (LLMs) are known to acquire reasoning capabilities through shared inference patterns in pre-training data, which are further elicited via Chain-of-Thought (CoT) practices. However, whether fundamental reasoning patterns, such as induction, deduction, and abduction, can be decoupled from specific problem instances remains a critical challenge for model controllability, and for shedding light on reasoning controllability. In this paper, we present the first systematic investigation of this problem through the lens of reasoning conflicts: an explicit tension between parametric and contextual information induced by mandating logical schemata that deviate from those expected for a target task. Our evaluation reveals that LLMs consistently prioritize sensibility over compliance, favoring task-appropriate reasoning patterns despite conflicting instructions. Notably, task accuracy is not strictly determined by sensibility, with models often maintaining high performance even when using conflicting patterns, suggesting a reliance on internalized parametric memory that increases with model size. We further demonstrate that reasoning conflicts are internally detectable, as confidence scores significantly drop during conflicting episodes. Probing experiments confirm that reasoning types are linearly encoded from middle-to-late layers, indicating the potential for activation-level controllability. Leveraging these insights, we steer models towards compliance, increasing instruction following by up to 29%. Overall, our findings establish that while LLM reasoning is anchored to concrete instances, active mechanistic interventions can effectively decouple logical schemata from data, offering a path toward improved controllability, faithfulness, and generalizability.

When instructed to underperform on multiple-choice evaluations, do language models engage with question content or fall back on positional shortcuts? We map the boundary between these regimes using a six-condition adversarial instruction-specificity gradient administered to two instruction-tuned LLMs (Llama-3-8B and Llama-3.1-8B) on 2,000 MMLU-Pro items. Distributional screening (response-position entropy) and an independent content-engagement criterion (difficulty-accuracy correlation) jointly characterise each condition. The gradient reveals three regimes rather than a monotonic transition. Vague adversarial instructions produce moderate accuracy reduction with preserved content engagement. Standard sandbagging and capability-imitation instructions produce positional entropy collapse with partial content engagement. A two-step answer-aware avoidance instruction produces extreme positional collapse, with near-total concentration on a single response position (99.9% and 87.4%) and no measurable content sensitivity. This was the only multi-step instruction tested, and it produced the most extreme shortcut. The attractor position matches each model's content-absent null-prompt default. The effect replicates across both models and four academic domains. Distributional collapse and content engagement can co-occur (50% concordance between screening criteria), indicating that entropy-based screening and difficulty-based content assessment capture partially independent dimensions of response validity. Results suggest that instruction complexity can determine whether adversarial compliance uses content-aware or content-blind mechanisms in small instruction-tuned LLMs under greedy decoding.

Generative AI has rapidly entered education through free consumer tools, outpacing the ability of schools and universities to respond. Now a new wave of more autonomous agentic AI systems--with the capacity to plan and act towards goals--promises both greater educational personalization and greater disruption. This chapter argues that successfully navigating these innovations requires balancing three core tensions: (1) Implementation Feasibility, or the practical capacity to integrate AI sustainably into real classrooms; (2) Adaptation Speed, or the mismatch between fast-evolving AI capabilities and the slower pace of educational change; and (3) Mission Alignment, or the need to ensure AI applications uphold educational values such as equity, privacy, and pedagogical integrity. First, we review early evidence of generative and agentic AI in various sectors and in frontline education to illustrate these tensions in context. Then, we present a three-tension framework to guide decision-makers in evaluating and designing AI initiatives across K-12 and higher education. We provide examples of how the framework can be applied to plan responsible AI deployments, and we identify emerging trends--such as curriculum-linked AI agents and educator-informed AI design--along with open research directions. We conclude the chapter with recommendations for educational leaders to proactively engage with the opportunities and challenges of AI, so that this technology can be harnessed to enhance teaching and learning in the decade ahead.

Drifting models are capable one-step generative models trained to follow a drifting field. The field combines attractive and repulsive softmax-weighted centroids over the data and current-generator distributions. In practice, only a minibatch of $n$ samples from each distribution is available, and each centroid is approximated by an empirical estimate. In this paper, we begin by showing that the minibatch centroid is in general a biased estimator of the target centroid, with a pointwise $O(1/n)$ bias arising from softmax self-normalization. Correcting this bias requires the expectation over the full distribution, which is intractable. We instead approximate the leading bias term from in-batch statistics and propose Analytical Bias Correction (ABC), a closed-form plug-in adjustment. We prove that ABC reduces the bias from $O(1/n)$ to $O(1/n^2)$, introduces no first-order increase in total variance, and preserves convex-hull containment of the corrected centroid. In practice, ABC requires only two additional lines of code and has negligible wall-time overhead under compiled execution. Toy experiments confirm the theoretical $O(1/n)$ and $O(1/n^2)$ scaling. On CIFAR-10, ABC reduces FID and trains faster, with the largest gains at small $n$, where the bias is most significant.

As large language models (LLMs) are increasingly fine-tuned for hardware tasks like RTL code generation, the scarcity of high-quality datasets often leads to the use of rapidly assembled or generated training data. These datasets frequently lack security verification and are highly susceptible to data poisoning attacks. Such poisoning can cause models to generate syntactically valid but insecure hardware modules that bypass standard functionality checks. To address this, we present SafeTune, a framework designed to harden LLM-based RTL generation against poisoning, specifically focusing on hardware Trojan (HT) insertion. SafeTune integrates two core components: (i) a Graph Neural Network (GNN) that models structural properties to identify anomalous circuitry patterns during fine-tuning, and (ii) a semantic verification module using text embeddings and an XGBoost classifier to assess prompt security. By coupling structural and semantic knowledge, SafeTune effectively filters poisoned inputs without sacrificing legitimate data. Experimental results demonstrate that SafeTune significantly enhances the robustness and reliability of LLM fine-tuning without requiring modifications to the underlying model architecture.

Remaining Useful Life (RUL) estimation is a critical component of Prognostics and Health Management (PHM), enabling proactive maintenance scheduling and reducing unplanned failures in industrial equipment. This paper presents a comparative study of machine learning approaches for RUL estimation on the NASA C-MAPSS turbofan engine dataset: classical baselines (Ridge Regression, Polynomial Ridge, and XGBoost), a 1D Convolutional Neural Network (CNN), and a Long Short-Term Memory (LSTM) network. All models are evaluated on the FD001 and FD003 subsets under an identical preprocessing pipeline to ensure a fair comparison. Among raw-sequence models, the LSTM achieves RMSE of 14.93 and 14.20 on FD001 and FD003 respectively, outperforming the deep LSTM reported by Zheng et al.~\cite{paper} (RMSE 16.14 and 16.18) despite using a simpler single-layer architecture. The 1D CNN achieves RMSE of 16.97 on FD001 and 15.68 on FD003, demonstrating competitive performance on FD003 while producing more conservative RUL predictions on FD001. Ridge Regression is evaluated on raw and engineered features, while other classical models use only engineered inputs. XGBoost achieves an RMSE of 13.36 on FD003, highlighting the competitiveness of nonlinear modeling.

Tool-calling agents are evaluated on tool selection, parameter accuracy, and scope recognition, yet LLM trajectory assessments remain inherently post-hoc. Disconnected from the active execution loop, such assessments identify errors that are usually addressed through prompt-tuning or retraining, and fundamentally cannot course-correct the agent in real time. To close this gap, we move evaluation into the execution loop at inference time: a specialized reviewer agent evaluates provisional tool calls prior to execution, shifting the paradigm from post-hoc recovery to proactive evaluation and error mitigation. In practice, this architecture establishes a clear separation of concerns between the primary execution agent and a secondary review agent. As with any multi-agent system, the reviewer can introduce new errors while correcting others, yet no prior work to our knowledge has systematically measured this tradeoff. To quantify this tradeoff, we introduce Helpfulness-Harmfulness metrics: helpfulness measures the percentage of base agent errors that feedback corrects; harmfulness measures the percentage of correct responses that feedback degrades. These metrics directly inform reviewer design by revealing whether a given model or prompt provides net positive value. We evaluate our approach on BFCL (single-turn) and Tau2-Bench (multi-turn stateful scenarios), achieving +5.5% on irrelevance detection and +7.1% on multi-turn tasks. Our metrics reveal that reviewer model choice is critical: the reasoning model o3-mini achieves a 3:1 benefit-to-risk ratio versus 2.1:1 for GPT-4o. Automated prompt optimization via GEPA provides an additional +1.5-2.8%. Together, these results demonstrate a core advantage of separating execution and review: the reviewer can be systematically improved through model selection and prompt optimization, without retraining the base agent.

Models of sign language have historically lagged behind those for spoken language (text and speech). Recent work has greatly improved their performance on tasks like sign language translation and isolated sign recognition. However, it remains unclear to what extent existing models capture various linguistic phenomena of sign language, and how well they use cues from the multiple articulators used in sign language (hands, upper body, face). We introduce a new benchmark dataset for American Sign Language, ASL Minimal Translation Pairs (ASL-MTP), divided into multiple types of sign language phenomena and corresponding minimal pairs of translations, for performing such linguistic analyses. As a case study, we use ASL-MTP to analyze a state-of-the-art ASL-to-English translation model. We conduct a targeted analysis of the model by ablating various input cues during training and inference and evaluating on the phenomena in ASL-MTP. Our results show that, while the model performs above chance level on most of the phenomena, it relies strongly on manual cues while often missing crucial non-manual cues.

Freelance workers must continually acquire new skills to remain competitive in online labor markets, yet they lack the organizational training, mentorship, and infrastructure available to traditional employees. Generative AI-powered tools like ChatGPT are reshaping market skill demands while also offering new forms of on-demand learning support to meet those demands. Despite growing interest in AI-powered learning tools, little is known about how freelancers actually use these tools to learn, the challenges they encounter, and how generative AI for learning interacts with precarity and competition in platform-based work. We present a mixed-methods study combining a survey and semi-structured interviews with freelance knowledge workers. Grounded in self-directed learning theory, we examine how freelancers integrate generative AI tools into their learning practices. Our findings show that freelancers increasingly rely on generative AI to structure learning and support exploratory skill acquisition, but do not treat it as their primary learning resource due to inconsistency, lack of contextual relevance, and verification overhead. We identify a shift from learning as growth to learning as survival, where upskilling is oriented toward immediate market viability rather than long-term development. We also surface a structural challenge we term invisible competencies, in which workers acquire skills through generative AI tools but lack credible ways to signal or validate these skills in competitive freelance markets. Based on these insights, we offer design recommendations for generative AI-powered learning tools for freelancers.

Computer science (CS) education needs to evolve to support software and artificial intelligence (AI) systems engineering, and it needs to happen now -- precisely because the core intellectual contributions of CS have never been more important. We argue that traditional curricula, built around programming, data structures, and algorithms as ends in themselves, must be reframed so that these topics become foundational building blocks within a systems- and engineering-centered education. Graduates should be prepared not to compete with AI on routine coding tasks, but to design, orchestrate, verify, and own complex AI-enabled systems operating under real-world constraints. More importantly, computer science education should be geared toward preparing students for future disruptions. The broad history of computing is marked by one disruptive technology after another, requiring us to rise to the moment instead of merely acquiescing to it.

Democratic discourse analysis systems increasingly rely on multi-agent LLM pipelines in which distinct evaluator models are assigned adversarial roles to generate structured, multi-perspective assessments of political statements. A core assumption is that models will reliably maintain their assigned roles. This paper provides the first systematic empirical test of that assumption using the TRUST pipeline. We develop an epistemic stance classifier that identifies advocate roles from reasoning text without relying on surface vocabulary, and measure role fidelity across 60 political statements (30 English, 30 German) using four metrics: Role Drift Index (RDI), Expected Drift Distance (EDD), Directional Drift Index (DDI), and Entropy-based Role Stability (ERS). We identify two failure modes - the Epistemic Floor Effect (fact-check results create an absolute lower bound below which the legitimizing role cannot be maintained) and Role-Prior Conflict (training-time knowledge overrides role instructions for factually unambiguous statements) - as manifestations of a single mechanism: Epistemic Role Override (ERO). Model choice significantly affects role fidelity: Mistral Large outperforms Claude Sonnet by 28pp (67% vs. 39%) and exhibits a qualitatively different failure mode - role abandonment without polarity reversal - compared to Claude's active switch to the opposing stance. Role fidelity is language-robust. Fact-check provider choice is not universally neutral: Perplexity significantly reduces Claude's role fidelity on German statements (Delta = -15pp, p = 0.007) while leaving Mistral unaffected. These findings have direct implications for multi-agent LLM validation: a system validated without role fidelity measurement may systematically misrepresent the epistemic diversity it was designed to provide.

Agentic web search increasingly faces two distinct demands: deep reasoning over a single target, and structured aggregation across many entities and heterogeneous sources. Current systems struggle on both fronts. Breadth-oriented tasks demand schema-aligned outputs with wide coverage and cross-entity consistency, while depth-oriented tasks require coherent reasoning over long, branching search trajectories. We introduce \textbf{Web2BigTable}, a multi-agent framework for web-to-table search that supports both regimes. Web2BigTable adopts a bi-level architecture in which an upper-level orchestrator decomposes the task into sub-problems and lower-level worker agents solve them in parallel. Through a closed-loop run--verify--reflect process, the framework jointly improves decomposition and execution over time via persistent, human-readable external memory, with self-evolving updates to each single-agent. During execution, workers coordinate through a shared workspace that makes partial findings visible, allowing them to reduce redundant exploration, reconcile conflicting evidence, and adapt to emerging coverage gaps. Web2BigTable sets a new state of the art on WideSearch, reaching an Avg@4 Success Rate of \textbf{38.50} ($7.5\times$ the second best at 5.10), Row F1 of \textbf{63.53} (+25.03 over the second best), and Item F1 of \textbf{80.12} (+14.42 over the second best). It also generalises to depth-oriented search on XBench-DeepSearch, achieving 73.0 accuracy. Code is available at https://github.com/web2bigtable/web2bigtable.

Cognitive decline is highly heterogeneous across individuals, which complicates prognosis, trial design, and treatment planning. We present the Personalized Cognitive Decline Assessment Digital Twin (PCD-DT), a multimodal and uncertainty-aware framework for modeling patient-specific disease trajectories from sparse, noisy, and irregular longitudinal data. The framework combines three methodological components: (1) latent state-space models for individualized temporal dynamics, (2) multimodal fusion for clinical, biomarker, and imaging features, and (3) uncertainty-aware validation and adaptive updating for robust digital twin operation. We also outline how conditional generative models can support data augmentation and stress testing for underrepresented progression patterns. As a preliminary feasibility study, we analyze longitudinal TADPOLE trajectories and show clear separation between cognitively normal and Alzheimer's disease cohorts in ADAS13, ventricle volume, and hippocampal volume over five years. We further conduct a multimodal next-visit prediction ablation using an LSTM sequence model on 3{,}003 visit-pair sequences derived from TADPOLE, where the combined cognitive plus MRI configuration achieves the lowest standardized RMSE for both ADAS13 (0.4419) and ventricle volume (0.5842), outperforming a Last Observation Carried Forward baseline. A Bayesian tensor modeling component for high-dimensional imaging fusion is also discussed. These results support the feasibility of the proposed architecture while also highlighting the need for stronger uncertainty calibration and longer-horizon predictive evaluation. The PCD-DT framework provides a principled starting point for personalized in silico modeling in neurodegenerative disease. This work positions PCD-DT as a foundational step toward clinically deployable, uncertainty-aware digital twin systems.

Large language models can now generate substantial code and draft research text, but research-software projects require more than either artifact alone. The mathematical thesis, executable system, benchmark surface, and public claims must mature together, yet often drift apart. We identify two LM-specific failure modes: hallucination accumulation, in which claims exceed what code or theory supports and unsupported assertions propagate across sessions; and desynchronization, in which code, theory, or the model's own world model fall out of alignment. We propose Comet-H, an iterative prompt automaton that orchestrates ideation, implementation, evaluation, grounding, and paper-writing as coupled coordinates of a single workspace state. At each step, a controller selects the next prompt by scoring it against what the workspace currently lacks, carries unfinished follow-up work forward with a half-life, and re-checks the paper and README against the code and benchmarks whenever documentation changes. We frame prompt selection as a small contextual bandit problem over prompt families, with prompts as arms, workspace deficits as context, and a hand-weighted linear score. This transparent scorer, paired with a fading record of unfinished work, bounds long-horizon follow-ups, requires no learned policy, and makes each prompt choice legible from the workspace. We created a portfolio of 46 research-software repositories across two dozen domains. We study A3 in depth, a Python static-analysis tool built entirely within the loop, which reaches (F1 = 0.768) on a 90-case benchmark, compared with a next-best baseline of 0.364. Across approximately 400 commits, we find that audit-and-contraction passes dominate the later phases of every successful trajectory.

In the field of universal automatic phonetic transcription (APT), clean and diverse training transcriptions are required. However, such high-quality data is limited. We propose the bootstrapping approach Selective Augmentation to improve the available training transcriptions by selectively transferring distinctions between languages. Based on the model MultIPA, we exemplarily show that we could increase the accuracy of an existing feature (plosive voicing) and add a new feature (plosive aspiration) by augmenting the existing training data using information from a separate helper language (Hindi). We describe intrinsic challenges of the evaluation and develop objective metrics to determine the success: Voicing accuracy was increased by 17.6% by reducing the number of false positives. Additionally, aspiration recognition was introduced: While the baseline transcribed 0% of German /p, t, k/ as aspirated, our approach transcribed them as aspirated in 61.2% of the cases. Introducing aspiration recognition to APT models allowed for the tenuis class to be successfully reduced by 32.2%, which also reduces the conflations between the test language's plosives.

Hybrid-thinking language models expose explicit think and no-think modes, but current designs do not separate them cleanly. Even in no-think mode, models often emit long and self-reflective responses, causing reasoning leakage. Existing work reduces this issue through better data curation and multi-stage training, yet leakage remains because both modes are still encoded in the same feed-forward parameters. We propose Path-Lock Expert (PLE), an architecture-level solution that replaces the single MLP in each decoder layer with two semantically locked experts, one for think and one for no-think, while keeping attention, embeddings, normalization, and the language-model head shared. A deterministic control-token router selects exactly one expert path for the entire sequence, so inference preserves the dense model's per-token computation pattern and each expert receives mode-pure updates during supervised fine-tuning. Across math and science reasoning benchmarks, PLE maintains strong think performance while producing a substantially stronger no-think mode that is more accurate, more concise, and far less prone to reasoning leakage. On Qwen3-4B, for example, PLE reduces no-think reflective tokens on AIME24 from 2.54 to 0.39 and improves no-think accuracy from 20.67% to 40.00%, all while preserving think-mode performance. These results suggest that controllable hybrid thinking is fundamentally an architectural problem, and separating mode-specific feed-forward pathways is a simple and effective solution.

Accurate prediction of conversion from Mild Cognitive Impairment (MCI) to Alzheimers Diseases (AD) is essential for early intervention, however, developing reliable conversion predictive models is difficult to develop due to limited longitudinal data availability We evaluate TabPFN (Tabular Pre-Trained Foundation Network) against traditional machine learning methods for predicting 3 year MCI to AD conversion using the TADPOLE dataset derived from ADNI. Using multimodal biomarker features extracted from demographics, APOE4, MRI volumes, CSF markers, and PET imaging, we conducted an experimental comparison across varying training set sizes (N=50 to 1000) and models including XGBoost, Random Forest, LightGBM, and Logistic Regression. TabPFN achieved one the highest performance (AUC=0.892), outperforming LightGBM (AUC=0.860) and demonstrating advantages in low data settings. At N=50 training samples, TabPFN maintained strong AUC while the traditional machine learning models struggles at small training samples. These findings demonstrate that foundation models are promising for disease prediction in data limited scenarios, such as Alzheimers diseases.

Automatic Emergency Braking (AEB) systems represent a safety-critical national interest, with the National Highway Traffic Safety Administration (NHTSA) Federal Motor Vehicle Safety Standard (FMVSS No. 127) requiring AEB in all new light vehicles sold in the United States by September 2029. However, production implementations frequently rely on deterministic stopping-distance or Time-to-Collision (TTC) thresholds that fail to capture uncertainty in sensing, road conditions, and vehicle dynamics. This paper presents a GPU-accelerated Monte Carlo framework for stochastic evaluation of emergency braking performance using a high-fidelity longitudinal vehicle model incorporating aerodynamic drag, road grade, brake actuator dynamics, and weight transfer effects. A one-thread-per-sample execution strategy exploits the independence of Monte Carlo rollouts, while deterministic CPU-generated sampling ensures bit-exact numerical consistency between CPU and GPU implementations. The framework is evaluated across four hardware platforms spanning development and deployment environments: two laptop GPUs (GTX 1650, RTX 5070) and two automotive-grade embedded platforms (Jetson Orin Nano, Jetson AGX Orin). Peak speedups of 54.57x are achieved while maintaining exact numerical agreement. Real-time feasibility analysis with a complete AEB timing budget (700 ms human reaction time minus 120 ms perception and 50 ms decision overhead) demonstrates that the Jetson AGX Orin can execute approximately 25,000 Monte Carlo samples within a 530 ms budget, enabling real-time probabilistic AEB evaluation as part of a complete embedded pipeline. These results establish Monte Carlo-based uncertainty evaluation as a deployable runtime component rather than an offline validation tool and provide quantitative guidance for risk-aware AEB threshold selection under the NHTSA final rule.

Variable selection in linear regression models has been a problem since hypothesis testing began. Which variables to include or exclude from a model is not an easy task. Techniques such as Forward, Back ward, Stepwise Regression sequentially add or delete variables from a model. Penalized likelihood methods such as AIC, BIC, etc. seek to choose variables that have a significant contribution to the likelihood. Penalized sum of square methods such as LASSO and Elastic Net have been used to penalize small coefficients to only allow variables with large coefficients in the model. This work introduces an Artificial Intelligence approach to model selection where an ANN is trained to determine the significance of the variables based on OLS estimates. A simulation study shows the accuracy across various sample sizes and variances. Furthermore, a simulation study is conducted to compare the performance of the approach against Forward, Backward, AIC, BIC and LASSO. The approach is illustrated using a dataset from the World Health Organization regarding Life Expectancy. A github link is provided to the pretrained ANN that can handle up to 100 predictor variables, the original WHO dataset and the subset used in this work.

Success-driven social learning, in which individuals preferentially adopt the ideas and methods that appear most successful, is a foundational principle of collective behavior across systems ranging from ant colonies to scientific communities. But science is a particular kind of collective search -- one in which the quality of an explanation is itself difficult to assess. Is success bias adaptive in this setting? In agent-based simulations of collective theory building, we find that it is not. Scientists in our model systematically overestimate the quality of their own theories, creating an illusion of understanding: a persistent gap between perceived and actual performance. Success bias amplifies this illusion; communities that favor apparently successful theories explore a narrower range of possibilities, efficiently filtering out poor explanations but failing to discover better ones. This effect intensifies with problem complexity, as scientists in more complex environments become increasingly unable to assess how well their theories actually perform. Most strikingly, when agents optimize their social behavior to maximize the perceived success of their theories, they paradoxically undermine their actual performance, and produce levels of inequality that mirror those found in real scientific communities.

We study finite-horizon budget allocation as a closed-loop economic control problem and evaluate receding-horizon Model Predictive Control (MPC) relative to reactive budgeting policies. Budgets are allocated periodically under execution noise and operational constraints, while return efficiency may evolve over time. Using a controlled simulation framework motivated by digital marketing, we compare reactive pacing to MPC across environments with increasing degrees of non-stationarity. Our results show that non-stationarity alone does not justify predictive control. When return dynamics are stationary or evolve through unpredictable stochastic drift, MPC offers no systematic advantage over reactive baselines. By contrast, when return efficiency exhibits predictable structure over the planning horizon, that is captured through an underlying model, MPC consistently outperforms reactive budgeting by exploiting intertemporal trade-offs.

Time-series data augmentation plays a crucial role in regression-oriented forecasting tasks, where limited data restricts the performance of deep learning models. While Generative Adversarial Networks (GANs) have shown promise in synthetic time-series generation, existing approaches primarily focus on matching marginal data distributions and often overlook the temporal dynamics that naturally exist in the original multivariate time series. When generating multivariate time series, this mismatch leads to distribution shift and temporal drift, thereby degrading the fidelity of the synthetic sequences. In this work, we propose a model-agnostic Markov Chain Monte Carlo (MCMC)-based framework to mitigate distribution shift and preserve temporal dynamics in synthetic time series. We provide a theoretical analysis of how conditional generative models accumulate deviations under sequential generation and demonstrate that the MCMC algorithm can correct these discrepancies by enforcing consistency with empirical transition statistics between neighboring time points. Extensive experiments on the Lorenz, Licor, ETTh, and ILI datasets using RCGAN, GCWGAN, TimeGAN, SigCWGAN, and AECGAN demonstrate that the proposed MCMC framework consistently improves autocorrelation alignment, skewness error, kurtosis error, R$^2$, discriminative score, and predictive score. These results suggest that synthetic time series consistent with the original data require explicit preservation of transition laws rather than solely relying on adversarial distribution matching, thereby offering a principled direction for improving generative modeling of time-series data.

Hyperledger Fabric (HLF) is a modular, permissioned blockchain widely adopted in enterprise settings. Enhancing its throughput and latency remains challenging, as optimization decisions made in one phase of the transaction lifecycle can adversely affect other phases. In this work, we present a systematic, phase-level and end-to-end study of HLF optimizations along three fronts, combining production-grade testbed experiments with calibrated SimPy simulations. First, we introduce two novel optimization techniques that target commit-phase bottlenecks: block-level pipelining and strategic waiting. In pipelining, we overlap validation and private-data acquisition of successive blocks with state-consistency checks and ledger updates improving commit throughput by up to 1.9x. Strategic waiting coordinates commit progress by temporarily pausing fast leaders and boosting laggers to sustain endorsement parallelism, yielding up to a 1.2x higher throughput. Second, we conduct micro-benchmarking of three configuration levers: private-data dissemination, block-size selection, and endorsement peer selection. Our results reveal that: (i) Relaxed quorums for private-data dissemination significantly reduce latency in both endorsement and commit phases; (ii) Under light workloads, smaller blocks yield lower end-to-end latency, whereas, under heavy workloads, larger blocks are necessary to improve throughput and reduce latency; and (iii) Relaxed leader selection dramatically reduces dropped transactions and boosts endorsement throughput, with a modest increase in MVCC invalidations. Finally, we analyze the interplay among private-data dissemination, VSCC parallelization, and pipelined commits. Interestingly, the throughput gains over a serial commit path are maximized at a moderate level of parallelization. Together, our findings provide phase-aware and protocol-level refinements for optimizing HLF.

We propose C-MTAD-GAT, an \emph{unsupervised}, \emph{context-aware} graph-attention model for anomaly detection in multivariate time series from mobile networks. C-MTAD-GAT combines graph attention with lightweight context embeddings, and uses a deterministic reconstruction head and multi-step forecaster to produce anomaly scores. Detection thresholds are calibrated \emph{without labels} from validation residuals, keeping the pipeline fully unsupervised. On the public TELCO dataset, C-MTAD-GAT consistently outperforms MTAD-GAT and the Telco-specific DC-VAE, two state-of-the-art baselines, in both event-level and pointwise F1, while triggering substantially fewer alarms. C-MTAD-GAT is also deployed in the Core network of a national mobile operator, demonstrating its resilience in real industrial settings.

We show that the geometric relations between semantic features in large language models' hidden states closely mirror human psychological associations. We construct feature vectors corresponding to 360 words and project them on 32 semantic axes (e.g. beautiful-ugly, soft-hard), and find that these projections correlate highly with human ratings of those words on the respective semantic scales. Second, we find that the cosine similarities between the semantic axes themselves are highly predictive of the correlations between these scales in the survey. Third, we show that substantial variance across the 32 semantic axes lies on a low-dimensional subspace, reproducing patterns typical of human semantic associations. Finally, we demonstrate that steering a word on one semantic axis causes spillover effects on the model's rating of that word on other semantic scales proportionate to the cosine similarity between those semantic axes. These findings suggest that features should be understood not only in isolation but through their geometric relations and the meaningful subspaces they form.

LLM agents are known to deviate from Nash equilibria in strategic interactions, but nobody has looked inside the model to understand why, or asked whether the deviation can be reversed. We do both. Working with four open-source models (Llama-3 and Qwen2.5, 8B to 72B parameters) playing four canonical two-player games, we establish the behavioral picture through self-play and cross-play experiments, then open up the 32-layer Llama-3-8B model and examine what actually happens during a strategic decision. The mechanistic findings are clear. Opponent history is encoded with near-perfect fidelity at the first layer (96% probe accuracy) and consumed progressively by later ones, while Nash action encoding is weak throughout, never exceeding 56%. There is no dedicated Nash module. Instead, the model privately favors the Nash action through most of its forward pass, but a prosocial override concentrated in the final layers reverses this, reaching 84% probability of cooperation at layer 30. When we inject a learned Nash direction into the residual stream, the behavior shifts bidirectionally, confirmed through concept clamping. The behavioral experiments surface six scale- and architecture-dependent findings, the most notable being that chain-of-thought reasoning worsens Nash play in small models but achieves near-perfect Nash play above 70B parameters. The cross-play experiments reveal three phenomena invisible in self-play: a small model can unravel any partner's cooperation by defecting early; two large models reinforce each other's cooperative instincts indefinitely; and who moves first in a coordination game determines which Nash equilibrium the system reaches. LLMs do not lack Nash-playing competence. They compute it, then suppress it.

We focus on the problem of \emph{Answer-Level Fine-Tuning} (ALFT), where the goal is to optimize a language model based on the correctness or properties of its final answers, rather than the specific reasoning traces used to produce them. Directly optimizing answer-level objectives is computationally intractable due to the need to marginalize over the vast space of latent reasoning paths. To overcome this, we propose a general game-theoretical framework that lifts the problem to a \emph{Distributional Alignment Game}. We formulate ALFT as a two-player game between a Policy (the generator) and a Target (an auxiliary distribution). We prove that the Nash Equilibrium of this game corresponds exactly to the solution of the original answer-level optimization problem. This variational perspective transforms the intractable marginalization problem into a tractable projection problem. We demonstrate that this framework unifies recent approaches to diversity and self-improvement (coherence) and provide efficient algorithms compatible with Group Relative Policy Optimization (GRPO), such as Coherence-GRPO, yielding significant complexity gains in mathematical reasoning tasks.

Reinforcement Learning (RL) algorithms exhibit high sample complexity, particularly when applied to Decentralized Partially Observable Markov Decision Processes (Dec-POMDPs). As a response, projects such as SampleFactory, EnvPool, Brax, and IsaacLab migrate parallel execution of classic environments such as MuJoCo and Atari into C++ thread pools or the GPU to decrease the computational cost of environment steps. We are interested in optimizing the decision-level of human-AI joint operations, so we introduce a compute-efficient Dec-POMDP engine natively architected in C++ called Hide-And-Seek-Engine. By employing Data-Oriented Design (DOD) principles, explicit 64-byte cache-line alignment to remove false sharing, and a zero-copy PyTorch memory bridge using pinned memory and Direct Memory Access (DMA), our engine sustains throughput of up to 33,000,000 steps per second (SPS) in a single-agent, 1024-environment, decentralized observations on an AMD Ryzen 9950X (16 cores). Ten agents reduces FPS to 7M SPS with generating random actions contributing 1/3rd the total runtime for reference. The engine achieves a throughput increase of approximately 3,500$\times$ over the baseline single threaded vectorized NumPy implementation and successfully trains cooperative multi-agent policies via PPO, DQN, and SAC in minutes, validating both its performance and generality.

Rational belief revision is commonly viewed as being based on a preference order between possible worlds, with the resulting new belief set being those sentences true in all the most preferred models of the incoming new information. Usually, such a preference order is taken to be a total preorder. Nevertheless, there are other, more general classes of ordering that can also be employed. In this paper, we explore two such classes that have been studied within the theory of rational choice but have seen limited or no application in belief revision. We begin with interval orders, introduced by Fishburn in the '80s, which associate with each possible world a nonnegative `interval' of plausibility. We then move on to biorders, studied by Aleskerov, Bouyssou, and Monjardet, which generalise interval orders by allowing the intervals to have negative lengths, a feature that can be used to capture a notion of dissonance or instability. We provide axiomatic characterisations of these two resulting families of belief revision operators, as well as of two further families of interest that lie between interval orders and biorders. We show that while biorder-based revisions satisfy the Success postulate, they do not always yield consistent outputs. By modifying their definition to discard inputs that lead to inconsistency as `incredible', we derive new families of so-called non-prioritised revision that satisfy the Consistency postulate, but not the Success one. These families are linked to credibility-limited revision operators of Hansson et al., but for which the set of credible sentences does not satisfy the single-sentence closure condition. We argue that the biorder-based approach is well-suited for scenarios where an agent might initially reject new information, but may accept it when presented with additional explanation.

Model merging aims to combine multiple models into one without additional training. Naïve parameter-space averaging can be fragile under architectural symmetries, as their geometry does not take them into account. In this work we show that not only the geometry, but also the averaging procedure itself, must be symmetry-invariant to achieve symmetry-aware merges. Consequently, we propose a general solution: merging as Fréchet averaging, i.e., selecting parameters that minimize a sum of geodesic distances on an appropriate manifold. In this view, the key design choice is the overall geometry, i.e., the choice of metric, manifold, and distance approximation, that determines what it means for two models to be "close". We show that Fréchet averaging, combined with simplifying assumptions, contains Fisher merging. Building on this, we examine the particular case of low-rank adapters (LoRA), whose symmetries induce a distinct geometry: that of a quotient manifold. We outline the limitations of current LoRA merging methods, propose a practical algorithm for this setting, and show how they compare with other commonly used approaches.

Computer-use agents provide a promising path toward general software automation because they can interact directly with arbitrary graphical user interfaces instead of relying on brittle, application-specific integrations. Despite recent advances in benchmark performance, strong computer-use agents remain expensive and slow in practice, since most systems invoke large multimodal models at nearly every interaction step. We argue that this uniform allocation of compute is fundamentally inefficient for long-horizon GUI tasks. Such trajectories are highly heterogeneous: many steps are routine and can be handled reliably by smaller, cheaper policies, while errors tend to concentrate at a relatively small number of high-risk moments. Across computer-use benchmarks, these failures repeatedly take two forms: progress stalls, where the agent loops, repeats ineffective actions, or fails to make meaningful progress, and silent semantic drift, where the agent continues taking locally plausible actions after already deviating from the user's true goal. To address this inefficiency, we propose an event-driven, step-level cascade for computer-use agents that runs a small policy by default and escalates to a stronger model only when lightweight learned monitors detect elevated risk. Our framework combines two complementary signals: a Stuck Monitor that detects degraded progress from recent reasoning-action history and triggers recovery, and a Milestone Monitor that identifies semantically meaningful checkpoints where sparse verification is most informative for catching drift. This design turns always-on frontier-model inference into adaptive, on-demand compute allocation over the course of an evolving interaction. The framework is modular and deployment-oriented: it can be layered on top of existing computer-use agents without changing the underlying agent architecture or retraining the large model.

Autonomous crypto trading systems often spend most of their design effort on finding entries, while exits are left to fixed rules that are rarely tested in a systematic way. This paper examines whether better stop-loss and take-profit settings can improve the performance of an autonomous trading agent swarm. Using more than 900 historical trades, we replay each trade under many alternative exit policies and compare results against the existing production setup. The study finds that exit design matters meaningfully: stronger configurations improve risk-adjusted performance and generally favor tighter loss limits, earlier profit capture, and closer trailing protection. The paper also discusses a key evaluation challenge: a purely chronological split was initially used, but the newest trades fell into an unusual war-driven market period that sharply distorted test results. To reduce the influence of that single episode, the main comparison was run on randomized data, with the drawbacks of doing so acknowledged explicitly. Overall, the paper presents a practical framework for tuning exit logic in a more disciplined and transparent way.

Conformal Prediction provides distribution-free prediction intervals with guaranteed coverage, but its reliance on a single global calibration threshold obscures the sources of uncertainty at the instance level. In particular, it conflates irreducible noise with uncertainty induced by heterogeneous training data (aleatoric), model limitations, or calibration mismatch (epistemic), offering little insight into why an interval is wide or whether it could be reduced. We introduce an uncertainty-aware explainability framework that analyses the reducibility of calibration-induced epistemic conformal uncertainty via progressive calibration localisation for regression tasks. The approach is diagnostic rather than causal: it does not estimate true aleatoric or epistemic uncertainty, but explains how conformal intervals contract and stabilise as calibration support is localised around a test instance. Across benchmarks and real-world data, absolute reducible uncertainty aligns with epistemic proxies, while its relative contribution varies by task, revealing regimes hidden by interval width. This instance-level view complements conformal uncertainty, enhancing interpretability without altering the predictor or coverage.

Continuous Integration (CI) enforces repository-level correctness through multi-stage workflows and is central to modern software development, yet diagnosing and repairing CI failures remains challenging. Unlike traditional program repair, CI failures frequently involve non-code artifacts, environment and dependency issues, noisy execution logs, and workflow-level constraints. Existing program repair benchmarks fall short in this setting: they are largely test-centric, restrict repairs to source code, assume fixed execution environments, and evaluate under simplified CI workflows that do not reflect real repository-level validation. We introduce CI-Repair-Bench, a benchmark for CI-verified, repository-level program repair constructed from real GitHub Actions executions. It contains 567 CI failure instances from 103 repositories and evaluates repair correctness exclusively through full CI re-execution under original workflows. Failures are categorized into 12 CI error types, enabling fine-grained, error-type-aware evaluation. To demonstrate benchmark usage, we include a reference CI repair workflow that analyzes CI logs to localize faults and generate candidate patches. Empirical results show that automated repair is most effective for localized, tool-enforced failures such as formatting and linting, while environment, dependency, and configuration-related failures remain challenging; the best-performing LLM achieves an 18.9% repair success rate. CI-Repair-Bench provides a realistic evaluation foundation for advancing research on CI-native automated program repair.

In generative modeling, we often wish to produce samples that maximize a user-specified reward such as aesthetic quality or alignment with human preferences, a problem known as guidance. Despite their widespread use, existing guidance methods either require expensive multi-particle, many-step schemes or rely on poorly understood approximations. We reformulate guidance as a deterministic optimal control problem, yielding a hierarchy of algorithms that subsumes existing approaches at the coarsest level. We show that the flow map, an object of significant recent interest for its role in fast inference, arises naturally in the optimal solution. Based on this observation, we propose Flow Map Reward Guidance (FMRG): a training-free, single-trajectory framework that uses the flow map to both integrate and guide the flow. At text-to-image scale, FMRG matches or surpasses baselines across inverse problems, style transfer, human preferences, and VLM rewards with as few as 3 NFEs, giving at least an order-of-magnitude speedup in comparison to prior state of the art.

Recent research has demonstrated the potential of Large Language Models (LLMs) for autonomous penetration testing, particularly when using cloud-based restricted-weight models. However, reliance on such models introduces security, privacy, and sovereignty concerns, motivating the use of locally hosted open-weight alternatives. Prior work shows that small open-weight models perform poorly on automated Linux privilege escalation, limiting their practical applicability. In this paper, we present a systematic empirical study of whether targeted system-level and prompting interventions can bridge this performance gap. We analyze failure modes of open-weight models in autonomous privilege escalation, map them to established enhancement techniques, and evaluate five concrete interventions (chain-of-thought prompting, retrieval-augmented generation, structured prompts, history compression, and reflective analysis) implemented as extensions to hackingBuddyGPT. Our results show that open-weight models can match or outperform cloud-based baselines such as GPT-4o. With our treatments enabled, Llama3.1 70B exploits 83% of tested vulnerabilities, while smaller models including Llama3.1 8B and Qwen2.5 7B achieve 67% when using guidance. A full-factorial ablation study over all treatment combinations reveals that reflection-based treatments contribute most, while also identifying vulnerability discovery as a remaining bottleneck for local models.

This paper proposes RCMAES, a novel variant of the Covariance Matrix Adaptation Evolution Strategy (CMA-ES) for CEC benchmark optimization. RCMAES integrates a dimension-dependent nonlinear population-size reduction strategy with an adaptive restart mechanism within a pure CMA-ES framework. RCMAES is evaluated on three benchmark suites (CEC2017, CEC2020, and CEC2022) and compared with state-of-the-art DE algorithms as well as its closely related counterpart, BIPOP-aCMAES. Experimental results show that RCMAES achieves competitive and robust performance across all benchmarks.

This paper introduces a systematic evaluation framework grounded in the Interagency Language Roundtable (ILR) Skill Level Descriptions and applies it to Claude (Sonnet 4.6) across six languages: English, French, Romanian, Spanish, Italian, and German. We administer a battery of 12 semantically equivalent prompt clusters spanning ILR complexity levels 1 through 3+, collect 216 responses (12 prompts, 6 languages, 3 runs), and analyze outputs through a two-layer methodology combining automated quantitative metrics with expert ILR qualitative assessment. Quantitative analysis reveals that French responses are approximately 30% longer than German responses on identical prompts, and that creative and affective clusters show the highest cross-lingual surface divergence. Qualitative analysis, conducted by a six-language professional with 12 years of ILR/OPI assessment experience, identifies five cross-lingual variation patterns: systematic differences in pragmatic disambiguation strategies, aesthetic and literary tradition divergence in creative output, language-internal technical terminology norms, cultural calibration gaps evidenced by the absence of culture-specific content in favor of culturally neutralized templates, and language-specific institutional referral behavior in emotional support responses. We argue that ILR-informed expert judgment applied to LLM outputs constitutes a novel and underreported evaluation methodology that complements purely computational benchmarks, and that cross-lingual output variation in Claude is interpretable, domain-dependent, and consequential for equitable multilingual AI deployment.

Generative AI is reshaping higher education programming through vibe coding, where students collaborate with AI via natural language rather than writing code line-by-line. We conceptualize this practice as help-seeking, analyzing 19,418 interaction turns from 110 undergraduate students. Using inductive coding and Heterogeneous Transition Network Analysis, we examined interaction sequences to compare top- and low-performing students. Results reveal that top performers engaged in instrumental help-seeking -- inquiry and exploration -- eliciting tutor-like AI responses. In contrast, low performers relied on executive help-seeking, frequently delegating tasks and prompting the AI to assume an executor role focused on ready-made solutions. These findings indicate that currently generative AI mirrors student intent (whether productive or passive) rather than optimizing for learning. To evolve from tools to teammates, AI systems must move beyond passive compliance. We argue for pedagogically aligned design that detect unproductive delegation and adaptively steer educational interactions toward inquiry, ensuring student-AI partnerships augment rather than replace cognitive effort.

Large Reasoning Models (LRMs) and Multi-Agent Systems (MAS) in high-stakes domains demand reliable verification, yet centralized approaches suffer four limitations: (1) Robustness, with single points of failure vulnerable to attacks and bias; (2) Scalability, as reasoning complexity creates bottlenecks; (3) Opacity, as hidden auditing erodes trust; and (4) Privacy, as exposed reasoning traces risk model theft. We introduce TRUST (Transparent, Robust, and Unified Services for Trustworthy AI), a decentralized framework with three innovations: (i) Hierarchical Directed Acyclic Graphs (HDAGs) that decompose Chain-of-Thought reasoning into five abstraction levels for parallel distributed auditing; (ii) the DAAN protocol, which projects multi-agent interactions into Causal Interaction Graphs (CIGs) for deterministic root-cause attribution; and (iii) a multi-tier consensus mechanism among computational checkers, LLM evaluators, and human experts with stake-weighted voting that guarantees correctness under 30% adversarial participation. We prove a Safety-Profitability Theorem ensuring honest auditors profit while malicious actors incur losses. All decisions are recorded on-chain, while privacy-by-design segmentation prevents reconstruction of proprietary logic. Across multiple LLMs and benchmarks, TRUST attains 72.4% accuracy (4-18% above baselines) and remains resilient against 20% corruption. DAAN reaches 70% root-cause attribution (vs. 54-63% for standard methods) with 60% token savings. Human studies validate the design (F1 = 0.89, Brier = 0.074). The framework supports (A1) decentralized auditing, (A2) tamper-proof leaderboards, (A3) trustless data annotation, and (A4) governed autonomous agents, pioneering decentralized AI auditing for safe, accountable deployment of reasoning-capable systems.

Foundation-model pipelines for individual-level livestock monitoring -- combining open-vocabulary detection, promptable video segmentation, and self-supervised visual embeddings -- have raised the accuracy ceiling of precision livestock farming (PLF), but their GPU memory budgets exceed the envelope of commodity edge accelerators. To close this gap, the 446M-parameter Perception Encoder (PE-ViT-L+) backbone of SAM 3 is distilled into a 40.66M-parameter multi-scale student through three mechanisms: a Feature Pyramid Network student encoder built on TinyViT-21M-512, a four-term direction-then-scale distillation loss, and backbone-substitution inference with sliding-window session pruning that bounds streaming GPU memory growth. The DINOv3 family includes a pre-distilled ViT-S/16 variant (21.6M parameters) released alongside a 6716M-parameter ViT-7B teacher; the ViT-S (21M) variant is adopted as the per-individual embedder. On the Edinburgh Pig dataset, the compressed pipeline reaches 92.29% MOTA and 96.15% IDF1 against the SAM 3 teacher (1.68- and 0.84-percentage-point losses), achieves a 7.77-fold reduction in system-level parameters and a 3.01-fold reduction in peak VRAM (19.52GB -> 6.49GB), and reaches 97.34% top-1 accuracy with 91.67% macro-F1 on nine-class pig behaviour classification. The pipeline fits inside an NVIDIA Jetson Orin NX 16GB envelope with 4.9GB of headroom, supporting a proposed -- but not yet empirically validated -- on-device embedding-pool re-identification mechanism whose per-individual footprint of approximately 94MB per animal per year produces a longitudinal visual record amenable to retrospective association with disease, lameness, reproductive, and growth outcome labels.

This study presents an unsupervised machine learning workflow for electrofacies analysis in the offshore Keta Basin, Ghana, where core data are scarce. Six standard wireline logs from Well~C were analysed over a depth interval comprising approximately $11{,}195$ samples. K-means clustering was applied in multivariate log space, with the clustering structure evaluated using inertia and silhouette diagnostics. Four clusters were identified, supported by an average silhouette coefficient of approximately $0.50$, indicating moderate but meaningful separation. The resulting electrofacies exhibit systematic, depth-continuous patterns associated with variations in clay content, porosity, and rock framework properties, forming a geological continuum from shale-dominated to cleaner sandstone-dominated units. The results demonstrate that log-only, unsupervised clustering supported by quantitative metrics provides a robust and reproducible framework for subsurface characterisation. The proposed workflow offers a practical tool for early-stage formation evaluation in frontier offshore basins and a foundation for future integrated studies.

Training stable biological foundation models requires rethinking attention mechanisms: we find that using sigmoid attention as a drop in replacement for softmax attention a) produces better learned representations: on six diverse single-cell datasets, sigmoid achieves 25% higher cell-type separation, better cell-type cohesion metrics, and lower validation loss, b) faster training, models with sigmoid attention train up to 10% faster than their softmax counterparts, and c) more stable training by eliminating inherent sources of instability in softmax attention. We establish that sigmoid attention has globally bounded derivatives ($\leq 0.25$) as opposed to softmax, and a diagonal Jacobian structure in contrast with softmax's dense coupling, which together help alleviate training instabilities. In stress tests on 160M-parameter bidirectional attention models trained without gradient clipping on 8K-token sequences, softmax diverges catastrophically, with gradients exploding by four orders of magnitude, while sigmoid remains stable. Finally, we implement and open-source TritonSigmoid, an efficient GPU kernel that achieves 515 TFLOPS on H100 GPUs, outperforming both FlashAttention-2 and FlashSigmoid, with native padding support, which is essential for biological sequences. Our results establish sigmoid attention as both theoretically grounded and empirically superior for biological foundation models. Code is available at https://github.com/MSDLLCpapers/triton-sigmoid

We present a priority-aware intelligent lane change advisory system based on multi-agent federated reinforcement learning, namely PALCAS, for autonomous vehicles (AVs). While existing lane-change approaches typically focus on single-agent systems or centralized multi-agent systems, we introduce a federated reinforcement learning-based multi-agent lane change system prioritizing lane changing based on vehicle destination urgency. PALCAS incorporates a novel priority-aware safe lane-change reward function to enable judicious lane-change decisions in both mandatory and discretionary scenarios. PALCAS leverages the parameterized deep Q-network (PDQN) algorithm to facilitate effective cooperation among agents, enabling both lateral and longitudinal motion controls of AVs. Extensive simulations conducted using the SUMO traffic simulator and Mosaic V2X communication framework demonstrate that PALCAS significantly improves traffic efficiency, driving safety, comfort, destination arrival rates, and merging success rates compared to baseline methods.

Recommender systems increasingly incorporate textual reviews to enrich user and item representations. However, most review-aware models remain optimized for rating prediction rather than ranking quality. This misalignment limits their effectiveness in top-N recommendation scenarios, where discriminative ranking is essential. To address this gap, we propose a Gated Hybrid Collaborative Filtering framework that integrates review-derived representations into an autoencoder-based collaborative model. The architecture injects semantic signals layer-wise through an adaptive gating mechanism that dynamically balances collaborative embeddings and topic-based features during encoding. To further refine the latent space, we introduce a contrastive learning module that aligns semantic and collaborative signals. We evaluate the framework across five distinct configurations: Pure collaborative; Topic and Gated; Text and Gated; and the addition of contrastive objectives (Contrastive and Topic, and Contrastive and Text). To explicitly optimize ranking behavior, the model is trained with a pairwise Bayesian personalized ranking objective, which promotes separation between relevant and non-relevant items in the latent space. Experiments on Amazon Movies & TV, IMDb, and Rotten Tomatoes demonstrate consistent improvements in hit rate @10 and normalized discounted cumulative gain @10 over state-of-the-art review-aware baselines. Results highlight the importance of controlled semantic fusion for ranking-driven recommendation.

Neuron pruning is widely used to reduce the computational cost and parameter footprint of large language models, yet it remains unclear whether neurons in task-specific models contribute uniformly to task performance. In this work, we provide empirical evidence for the existence and importance of task-specific neurons through a systematic pruning study on language models specialized for mathematical reasoning and code generation. We introduce an activation-based selectivity metric to identify neurons with low contribution to the target task and prune them while preserving target-task accuracy, and compare selective pruning with random pruning. Selective pruning consistently outperforms random pruning, indicating that activation-based selectivity provides a systematic advantage over random pruning. Reverse pruning experiments further show that removing a small subset of highly task-specific neurons (~10%) causes complete performance collapse, suggesting that there exist task specific neurons and critical task information is concentrated in a small portion of the network. In contrast, selective pruning of less critical neurons (~30% - ~35%) reduces accuracy but still preserves significant performance. We also observed consistent reductions in parameters and runtime VRAM usage, along with improved inference throughput as pruning increases. Experiments on both 1.5B and 7B models reveal a robustness threshold around 15-20% pruning, beyond which accuracy loss and generation failures increase sharply. Fine-tuning substantially recovers performance across pruning levels, particularly for aggressively pruned models. These findings provide empirical evidence of neuron specialization in task-specific language models and offer insights into pruning robustness, model redundancy, and post-pruning recoverability.

This paper explores the effectiveness of modular randomized testing for object oriented programs in Java. Modular testing involves testing individual components of a program in isolation. Often times, for effective test generation, a series of non-target setup calls must be included to obtain high coverage of the target component. In this work, we evaluate and improve modular testing with the EvoSuite test generator. We find that due to strict restrictions that disallow calls to non-target setup methods, EvoSuite's modular testing mode is ineffective and often results in low branch coverage. We propose \textsc{emote} (Effective Modular Testing with EvoSuite): an enhancement to EvoSuite that relaxes this restriction, allowing non-target methods to be included in the test prefixes. This modification draws inspiration from developer-written fuzz drivers, which often invoke setup methods to properly initialize the state before testing the target method. To ensure meaningful test generation, we modify EvoSuite's fitness function to focus branch coverage contributions on the call chain originating from the target method. \textsc{emote} is evaluated on a subset of the SF100 benchmark, showing a 15.15\% improvement in coverage of the target methods.

Accurately reconstructing complex full multi-object scenes from sparse observations remains a core challenge in computer vision and a key step toward scalable and reliable simulation for robotics. In this work, we introduce RecGen, a generative framework for probabilistic joint estimation of object and part shapes, as well as their pose under occlusion and partial visibility from one or multiple RGB-D images. By leveraging compositional synthetic scene generation and strong 3D shape priors, RecGen generalizes across diverse object types and real-world environments. RecGen achieves state-of-the-art performance on complex, heavily occluded datasets, robustly handling severe occlusions, symmetric objects, object parts, and intricate geometry and texture. Despite using nearly 80% fewer training meshes than the previous state of the art SAM3D, RecGen outperforms it by 30.1% in geometric shape quality, 9.1% in texture reconstruction, and 33.9% in pose estimation.

Soil contamination by heavy metals poses a persistent environmental and public health concern in rapidly urbanising regions of Ghana, particularly at unregulated waste disposal sites. This study applies an unsupervised machine learning framework to detect and characterise anomalous heavy metal contamination patterns in soils from twelve waste sites and residential controls in the Central Region, of Ghana. Concentrations of eight metals (As, Cd, Cr, Cu, Hg, Ni, Pb, Zn) were analysed alongside standard health risk indices, including the Hazard Index (HI) and Incremental Lifetime Cancer Risk (ILCR). Isolation Forest and PCA reconstruction error each identified $12$ anomalous samples ($15.4\%$ of $78$ samples), while DBSCAN detected no density-isolated noise points. A consensus approach isolated six robust anomalies ($7.7\%)$, all spatially concentrated at a single site (S3). Anomalies exhibited approximately $70$--$80\%$ higher mean HI values than normal samples, with all consensus anomalies exceeding the HI$=1$ threshold. PCA reconstruction error showed a strong positive association with HI ($r \approx 0.8$), indicating consistency between multivariate deviation and health risk. Three distinct anomaly types were identified: extreme Cu enrichment at S3, anomalously low Ni at S4/S5, and moderate multi-metal (Pb--Zn) co-elevation at S9--S12. The results demonstrate that unsupervised machine learning provides granular, objective insight beyond aggregate indices, enabling targeted site prioritisation and risk-informed environmental management.

The purpose of our paper is to develop a unified multi-agent architecture that automates end-to-end machine learning (ML) pipeline generation from datasets and natural-language (NL) goals, improving efficiency, robustness and explainability. A five-agent system is proposed to handle profiling, intent parsing, microservice recommendation, Directed Acyclic Graph (DAG) construction and execution. It integrates code-grounded Retrieval-Augmented Generation (RAG) for microservice understanding, an explainable hybrid recommender combining multiple criteria, a self-healing mechanism using Large Language Model (LLM)-based error interpretation and adaptive learning from execution history. The approach is evaluated on 150 ML tasks across diverse scenarios. The system achieves an 84.7% end-to-end pipeline success rate, outperforming baseline methods. It demonstrates improved robustness through self-healing and reduces workflow development time compared to manual construction. The study introduces a novel integration of code-grounded RAG, explainable recommendation, self-healing execution and adaptive learning within a single architecture, showing that tightly coupled intelligent components can outperform isolated solutions.

Current LLM safety alignment techniques improve model robustness against adversarial attacks, but overlook whether and how LLMs can recover helpfulness when benign users clarify their intent. We introduce CarryOnBench, the first interactive benchmark that measures whether LLMs can revise their interpretation of user intent and recover utility, while remaining safe through multi-turn conversations. Starting from 398 seemingly harmful queries with benign underlying intents, we simulate 5,970 conversations by varying user follow-up sequences, evaluating 14 models on both intent-aligned utility and safety. CarryOnBench yields 1,866 different conversation flows of 4--12 turns, totaling 23,880 model responses. We design Ben-Util, a checklist-based metric that evaluates how well each model response fulfills the user's benign information need using atomic items. At turn one, models fulfill only 10.5--37.6% of the user's benign information need. When the same query includes the benign intent upfront, models fulfill 25.1--72.1%, confirming that models withhold information due to intent misinterpretation, not limited knowledge. With benign clarifications in multi-turn conversations, 13 of 14 models approach or exceed this single-turn baseline, yet recovery cost varies across models. We identify three failure modes invisible to single-turn evaluations: utility lock-in, where a model rarely updates despite clarification; unsafe recovery, where a model updates at disproportionate safety cost; and repetitive recovery, where a model recycles prior responses rather than providing new information. Moreover, conversations converge to similar harmfulness levels regardless of how conservative the model starts. These findings expose a gap that single-turn evaluations miss -- whether a model is appropriately cautious or simply unresponsive to clarified user intent.

Scientific research has long been human-led, driving new knowledge and transformative technologies through the continual revision of questions, methods and claims as evidence accumulates. Although large language model (LLM)-based agents are beginning to move beyond assisting predefined research workflows, none has yet demonstrated end-to-end autonomous discovery in a real physical system that produces a nontrivial result supported by experimental evidence. Here we introduce Qiushi Discovery Engine, an LLM-based agentic system for end-to-end autonomous scientific discovery on a real optical platform. Qiushi Engine combines nonlinear research phases, Meta-Trace memory and a dual-layer architecture to maintain adaptive and stable research trajectories across long-horizon investigations involving thousands of LLM-mediated reasoning, measurement and revision actions. It autonomously reproduces a published transmission-matrix experiment on a non-original platform and converts an abstract coherence-order theory into experimental observables, providing, to our knowledge, the first observation of this class of coherence-order structure. More importantly, in an open-ended study involving 145.9 million tokens, 3,242 LLM calls, 1,242 tool calls, 163 research notes and 44 scripts, Qiushi Engine proposes and experimentally validates optical bilinear interaction, a physical mechanism structurally analogous to a core operation in Transformer attention. This AI-discovered mechanism suggests a route towards high-speed, energy-efficient optical hardware for pairwise computation. To our knowledge, this is the first demonstration of an AI agentic system autonomously identifying and experimentally validating a nontrivial, previously unreported physical mechanism, marking a milestone for research-level autonomous agents.

Large-language-models (LLMs) demonstrate enormous utility in long-context tasks which require processing prompts that consist of tens to hundreds of thousands of tokens. However, existing LLM training libraries do not provide easy to use abstractions to optimize for long-context training, instead focusing on optimizations for models with large parameter counts through ZeRO-3/FSDP, Tensor and Pipeline parallelism. This forces users to rewrite LLM training libraries to incorporate compositions of various complex long-context optimizations, such as sequence-parallelism, to training pipelines; a process that requires in-depth expertise, reducing developer productivity. To tackle these challenges, we introduce AutoSP: the first automated solution to automatically optimize LLM training for longer-contexts. AutoSP compiles models and applies a targeted set of optimizations: automated sequence parallelism, and long-context aware activation-checkpointing, to drastically enhance LLM trainability at negligible cost to throughput. Our evaluation demonstrates AutoSP's capability on both NVIDIA and AMD hardware, increasing training contexts by upto 2.7$\times$ and 2.5$\times$ respectively over competitive hand-written baseline at negligible cost to runtime performance.

Fine-tuning Large Language Models (LLMs) on consumer-grade GPUs is highly cost-effective, yet constrained by limited GPU memory and slow PCIe interconnects. Pipeline parallelism combined with CPU offloading mitigates these hardware bottlenecks by reducing communication overhead. However, existing PP schedules suffer from an inherent limitation termed the weight binding issue. Binding uneven model stages (e.g., the LM head is large) to GPUs limits the pipeline's throughput to that of the GPU with the heaviest load, leading to severe pipeline bubbles. In this paper, we propose RoundPipe, a novel pipeline schedule that breaks the weight binding constraint on consumer GPU servers. RoundPipe treats GPUs as a pool of stateless execution workers and dynamically dispatches computation stages across devices in a round-robin manner, achieving a near-zero-bubble pipeline. To ensure training correctness and system efficiency, RoundPipe integrates a priority-aware transfer scheduling engine, a fine-grained distributed event-based synchronization protocol, and an automated layer partitioning algorithm. Evaluations on an 8$\times$ RTX 4090 server demonstrate that RoundPipe achieves 1.48--2.16$\times$ speedups over state-of-the-art baselines when fine-tuning 1.7B to 32B models. Remarkably, RoundPipe enables LoRA fine-tuning of the Qwen3-235B model with 31K sequence length on a single server. RoundPipe is publicly available as an open-source Python library with comprehensive documentation.

RLVR and OPD have become standard paradigms for post-training. We provide a unified analysis of these two paradigms in consolidating multiple expert capabilities into a single model, identifying capability loss in different ways: mixed RLVR suffers from inter-capability divergence cost, while the pipeline of first training experts and then performing OPD, though avoiding divergence, fails to fully absorb teacher capabilities due to large behavioral pattern gaps between teacher and student. We propose Co-Evolving Policy Distillation (CoPD), which encourages parallel training of experts and introduces OPD during each expert's ongoing RLVR training rather than after complete expert training, with experts serving as mutual teachers (making OPD bidirectional) to co-evolve. This enables more consistent behavioral patterns among experts while maintaining sufficient complementary knowledge throughout. Experiments validate that CoPD achieves all-in-one integration of text, image, and video reasoning capabilities, significantly outperforming strong baselines such as mixed RLVR and MOPD, and even surpassing domain-specific experts. The model parallel training pattern offered by CoPD may inspire a novel training scaling paradigm.

We present a framework for migrating production Large Language Model (LLM) based systems when the underlying model reaches end-of-life or requires replacement. The key contribution is a Bayesian statistical approach that calibrates automated evaluation metrics against human judgments, enabling confident model comparison even with limited manual evaluation data. We demonstrate this framework on a commercial question-answering system serving 5.3M monthly interactions across six global regions; evaluating correctness, refusal behavior, and stylistic adherence to successfully identify suitable replacement models. The framework is broadly applicable to any enterprise deploying LLM-based products, providing a principled, reproducible methodology for model migration that balances quality assurance with evaluation efficiency. This is a capability increasingly essential as the LLM ecosystem continues to evolve rapidly and organizations manage portfolios of AI-powered services across multiple models, regions, and use cases.

The Normalized Transformer, or nGPT (arXiv:2410.01131) achieves impressive training speedups and does not require weight decay or learning rate warmup. However, despite having hyperparameters that explicitly scale with model size, we observe that nGPT does not exhibit learning rate transfer across model dimension and token horizon. To rectify this, we combine numerical experiments with a principled use of alignment exponents (arXiv:2407.05872) to revisit and modify the $μ$P approach to hyperparameter transfer (arXiv:2011.14522). The result is a novel nGPT parameterization we call $ν$GPT. Through extensive empirical validation, we find $ν$GPT exhibits learning rate transfer across width, depth, and token horizon.

Due to hardware-software co-development in embedded systems, continuous integration (CI) builds frequently fail because of complex cross-compilation, board configurations, and toolchain constraints. Although CI build logs contain valuable diagnostic information, they are short-lived and difficult to reuse due to heterogeneous runners, toolchains, and log formats. To address these challenges, we present PhantomRun, a unified abstraction layer and publicly reusable dataset that standardizes the retrieval, storage, and reproduction of CI build logs and metadata. Across 4628 failing CI runs, we reconstructed 91.8% of builds and preserved execution outcomes in 98% of evaluated cases. PhantomRun provides two core capabilities: retrieving the build log of any commit and faithfully re-executing the corresponding build in a controlled environment. By exposing all build artifacts and metadata in a uniform, machine-readable format, PhantomRun enables reproducible and longitudinal studies of CI failures. An empirical evaluation shows that reproduced builds closely match their originals, typically differing only in timestamps or minor nondeterministic reordering, demonstrating the feasibility of large-scale historical CI reconstruction.

Continual learning agents with finite capacity must balance acquiring new knowledge with retaining the old. This requires controlled forgetting of knowledge that is no longer needed, freeing up capacity to learn. Weight decay, viewed as a mechanism for forgetting, can serve this role by gradually discarding information stored in the weights. However, a fixed scalar weight decay drives this forgetting uniformly over time and uniformly across all parameters, even when some encode stable knowledge while others track rapidly changing targets. We introduce Forgetting through Adaptive Decay (FADE), which adapts per-parameter weight decay rates online via approximate meta-gradient descent. We derive FADE for the online linear setting and apply it to the final layer of neural networks. Our empirical analysis shows that FADE automatically discovers distinct decay rates for different parameters, complements step-size adaptation, and consistently improves over fixed weight decay across online tracking and streaming classification problems.

Nonlinear models and optimization methods have successfully tackled a rapidly growing set of problems in recent years. Indeed, a relatively small toolbox of such models and methods can provide sufficient performance across a large landscape of tasks: deep learning alone has made significant recent contributions in scientific modelling, natural language processing, visual analysis, etc. A similar relationship exists between physical theories and phenomena, where many applications and observations emerge neatly from remarkably minimal foundations. It is natural to wonder if sparse unified frameworks could be built to steer discussion and discovery in the fields concerned with learning, optimization, and modelling. In this work, we posit and examine a possible outline for such a unified theory, interpreting the notion of ''learning'' in a broad sense. In particular, we pursue our goals by viewing learning as an inter-connected process on multiple levels: problem setup, choosing methods, and the analysis of their interplay via imposed optimisation dynamics. We begin by proposing a precise yet versatile definition for ''solvable'' problems. We then define the ''parametrised methods'' by which their solution(s) may be ''learned''. Our goal is to sketch a ''universal convergence theorem'', specifying how and when solvable problems become amenable to the methods chosen for them. We find these constructions reduce the study of learning down to remarkably few ideas and tools - many of which are simply adapted from existing ones in dynamical systems theory, geometry, and fundamental physics.

As Large Language Model (LLM) agents transition from single-session tools to persistent systems managing longitudinal healthcare journeys, their memory architectures face a critical challenge: reconciling two imperfect sources of truth. The patient's evolving self-report is current but prone to recall bias, while the Electronic Health Record (EHR) is medically validated but frequently stale. General-purpose agent memory systems optimize for coherence by overwriting older facts with the user's latest statement, a pattern that risks safety failures when applied to clinical data. We introduce a Dual-Stream Memory Architecture that strictly separates the patient narrative from the structured clinical record (FHIR), governed by a dedicated Reconciliation Engine that evaluates every extracted memory against the patient's FHIR profile and classifies discrepancies by type, severity, and the specific FHIR resources involved. We evaluate this architecture on 26 patients across 675 longitudinal wellness coaching sessions, using a hybrid dataset that interleaves real provider-patient transcripts with synthetic, FHIR-grounded clinical scenarios. In isolated testing, the engine detects 84.4% of designed clinical discrepancies with 86.7% safety-critical recall. By coupling extraction and reconciliation evaluation on the same data, we directly quantify a 13.6% error cascade, tracing the degradation to clinical details lost during memory extraction from unstructured conversation rather than to downstream classification errors. These findings establish that validating patient-reported memories against clinical records is both feasible and necessary for safe deployment of longitudinal health agents.

Diffusion large language models (dLLMs) offer parallel decoding and bidirectional context, but state-of-the-art dLLMs require billions of parameters for competitive performance. While existing distillation methods for dLLMs reduce inference steps within a single architecture, none address cross-architecture knowledge transfer, in which the teacher and student differ in architecture, attention mechanism, and tokenizer. We present TIDE, the first framework for cross-architecture dLLM distillation, comprising three modular components: (1) TIDAL, which jointly modulates distillation strength across training progress and diffusion timestep to account for the teacher's noise-dependent reliability; (2) CompDemo, which enriches the teacher's context via complementary mask splitting to improve predictions under heavy masking; and (3) Reverse CALM, a cross-tokenizer objective that inverts chunk-level likelihood matching, yielding bounded gradients and dual-end noise filtering. Distilling 8B dense and 16B MoE teachers into a 0.6B student via two heterogeneous pipelines outperforms the baseline by an average of 1.53 points across eight benchmarks, yielding notable gains in code generation, where HumanEval scores reach 48.78 compared to 32.3 for the AR baseline.

We introduce Hyper Input Convex Neural Networks (HyCNNs), a novel neural network architecture designed for learning convex functions. HyCNNs combine the principles of Maxout networks with input convex neural networks (ICNNs) to create a neural network that is always convex in the input, theoretically capable of leveraging depth, and performs reliable when trained at scale compared to ICNNs. Concretely, we prove that HyCNNs require exponentially fewer parameters than ICNNs to approximate quadratic functions up to a given precision. Throughout a series of synthetic experiments, we demonstrate that HyCNNs outperform existing ICNNs and MLPs in terms of predictive performance for convex regression and interpolation tasks. We further apply HyCNNs to learn high-dimensional optimal transport maps for synthetic examples and for single-cell RNA sequencing data, where they oftentimes outperform ICNN-based neural optimal transport methods and other baselines across a wide range of settings.

Small language models (SLMs) offer computational efficiency for scalable deployment, yet they often fall short of the reasoning power exhibited by their larger counterparts (LLMs). To mitigate this gap, current approaches invoke an LLM to generate tokens at points of reasoning divergence, but these external calls introduce substantial latency and costs. Alternatively, standard distillation is often hindered by the capacity limitation, as SLMs struggle to accurately mimic the LLM's complex generative distribution. We address this dilemma by identifying local sufficiency: at divergence points, the LLM's preferred token consistently resides within the SLM's top-K next-token predictions, even when failing to emerge as the SLM top-1 choice. We therefore propose SELECT TO THINK (S2T), which reframes the LLM's role from open-ended generation to selection among the SLM's proposals, simplifying the supervision signal to discrete candidate rankings. Leveraging this, we introduce S2T-LOCAL, which distills the selection logic into the SLM, empowering it to perform autonomous re-ranking without inference-time LLM dependency. Empirically, we demonstrate that a 1.5B SLM's top-8 candidates capture the 32B LLM's choice with 95% hit rate. Translating this potential into performance, S2T-LOCAL improves greedy decoding by 24.1% on average across benchmarks, effectively matching the efficacy of 8-path self-consistency while operating with single-trajectory efficiency.

The Alternating Direction Method of Multipliers (ADMM) is a widely used method for structured convex optimization, and its practical performance depends strongly on the choice of penalty and relaxation parameters. Motivated by settings such as Model Predictive Control (MPC), where one repeatedly solves related optimization problems with fixed structure and changing parameter values, we propose learning online updates of the relaxation parameter to improve performance on problem classes of interest. This choice is computationally attractive in OSQP-like architectures, since adapting relaxation does not trigger the matrix refactorizations associated with penalty updates. We establish convergence guarantees for ADMM with time-varying penalty and relaxation parameters under mild assumptions, and show on benchmark quadratic programs that the resulting learned policies improve both iteration count and wall-clock time over baseline OSQP.

We introduce the problem of adaptive self-organization in which the nodes of an anonymous, synchronous dynamic network must distributively change the collective distribution of their responses (or "colors") as a function of time-varying environmental signals, even when these signals are only perceived locally and the network topology changes adversarially. Specifically, a signal adversary may change the type of signal and which node(s) witness that signal arbitrarily between rounds. If a signal (or lack thereof) $s$ persists in the system for sufficiently long, the dynamic network must stabilize such that nodes' colors reach and remain in a distribution closely approximating $r(s)$, a goal distribution defined by the problem instance. We first prove that if nodes are deterministic, the only solvable instances of adaptive-self organization are those with homogeneous goal distributions, i.e., those where all nodes must stabilize with the same color. We then present a linear-time, logarithmic-memory, deterministic algorithm for this subclass of instances that works even when the multiplicity and location of signal witnesses change arbitrarily. When nodes know $n$, the number of nodes in the network, a small adaptation of this algorithm achieves a stronger convergence property in which adversarial edge and signal dynamics are entirely unable to disturb stabilized configurations. Finally, we present a randomized extension of these algorithms that solves arbitrary (i.e., not necessarily homogeneous) instances of adaptive self-organization with high probability when nodes know the goal distributions.

Today's AI assistants such as OpenClaw are designed to handle context effectively, making context learning an increasingly important capability for models. As these systems move beyond professional settings into everyday life, the nature of the contexts they must handle also shifts. Real-life contexts are often messy, fragmented, and deeply tied to personal and social experience, such as multi-party conversations, personal archives, and behavioral traces. Yet it remains unclear whether current frontier language models can reliably learn from such contexts and solve tasks grounded in them. To this end, we introduce CL-bench Life, a fully human-curated benchmark comprising 405 context-task pairs and 5,348 verification rubrics, covering common real-life scenarios. Solving tasks in CL-bench Life requires models to reason over complex, messy real-life contexts, calling for strong real-life context learning abilities that go far beyond those evaluated in existing benchmarks. We evaluate ten frontier LMs and find that real-life context learning remains highly challenging: even the best-performing model achieves only 19.3% task solving rate, while the average performance across models is only 13.8%. Models still struggle to reason over contexts such as messy group chat histories and fragmented behavioral records from everyday life. CL-bench Life provides a crucial testbed for advancing real-life context learning, and progress on it can enable more intelligent and reliable AI assistants in everyday life.

In Orabona and Pál [2016], we introduced the shifted KT potentials, to remove the $\ln \ln T$ factor in the parameter-free learning with expert bound. In this short technical note, I show that this is equivalent to changing the prior in the Krichevsky--Trofimov algorithm. Then, I show how to use the same idea to remove the $\ln \ln T$ factor in the data-independent bound for the Squint algorithm.

LLMs have achieved strong results on both function-level code synthesis and repository-level code modification, yet a capability that falls between these two extremes -- compositional code creation, i.e., building a complete, internally structured class from a specification -- remains underserved. Current evaluations are either confined to isolated functions or rely on manually curated class-level tasks that are expensive to scale and increasingly susceptible to data contamination. We introduce ClassEval-Pro, a benchmark of 300 class-level tasks spanning 11 domains, constructed through an automated three-stage pipeline that combines complexity enhancement, cross-domain class composition, and integration of real-world GitHub code contributed after January 2025. Every task is validated by an LLM Judge Ensemble and must pass test suites with over 90% line coverage. We evaluate five frontier LLMs under five generation strategies. The best model achieves only 45.6% class-level Pass@1, with a 17.7-point gap between the strongest and weakest models, confirming the benchmark's discriminative power. Strategy choice strongly interacts with model capability: structured approaches such as bottom-up improve weaker models by up to 9.4 percentage points, while compositional generation collapses to as low as 1.3%. Error analysis over 500 manually annotated failures reveals that logic errors (56.2%) and dependency errors (38.0%) dominate, identifying cross-method coordination as the core bottleneck.

Learning curves are a fundamental primitive in supervised learning, describing how an algorithm's performance improves with more data and providing a quantitative measure of its generalization ability. Formally, a learning curve plots the decay of an algorithm's error for a fixed underlying distribution as a function of the number of training samples. Prior work on revenue-maximizing learning algorithms, starting with the seminal work of Cole and Roughgarden [STOC, 2014], adopts a distribution-free perspective, which parallels the PAC learning framework in learning theory. This approach evaluates performance against the hardest possible sequence of valuation distributions, one for each sample size, effectively defining the upper envelope of learning curves over all possible distributions, thus leading to error bounds that do not capture the shape of the learning curves. In this work we initiate the study of learning curves for revenue maximization and provide a near-complete characterization of their rate of decay in the basic setting of a single item and a single buyer. In the absence of any restriction on the valuation distribution, we show that there exists a Bayes-consistent algorithm, meaning that its learning curve converges to zero for any arbitrary valuation distribution as the number of samples $n \to \infty$. However, this convergence must be arbitrarily slow, even if the optimal revenue is finite. In contrast, if the optimal revenue is achieved by a finite price, then the optimal rate of decay is roughly $1/\sqrt{n}$. Finally, for distributions supported on discrete sets of values, we show that learning curves decay almost exponentially fast, a rate unattainable under the PAC framework.

Can Neural Assemblies -- groups of neurons that fire together and strengthen through co-activation -- learn the direction of causal influence between variables? While established as a computationally general substrate for classification, parsing, and planning, neural assemblies have not yet been shown to internalize causal directionality. We demonstrate that the inherent operations of neural assemblies -- projection, local plasticity control, and sparse winner selection -- are sufficient for directional learning. We introduce DIRECT (DIRectional Edge Coupling/Training), a mechanism that co-activates source and target assemblies under an adaptive gain schedule to internalize directed relations. Unlike backpropagation-based methods, DIRECT relies solely on local plasticity, making the resulting causal claims auditable at the mechanism level. Our findings are verified through a dual-readout validation strategy: (i) synaptic-strength asymmetry, measuring the emergent weight gap between forward and reverse links, and (ii) functional propagation overlap, quantifying the reliability of directional signal flow. Across multiple domains, the framework achieves perfect structural recovery under a supervised, known-structure setting. These results establish neural assemblies as an auditable bridge between biologically plausible dynamics and formal causal models, offering an "explainable by design" framework where causal claims are traceable to specific neural winners and synaptic asymmetries.

Claw-style environments support multi-step workflows over local files, tools, and persistent workspace states. However, scalable development around these environments remains constrained by the absence of a systematic framework, especially one for synthesizing verifiable training data and integrating it with agent training and diagnostic evaluation. To address this challenge, we present ClawGym, a scalable framework that supports the full lifecycle of Claw-style personal agent development. Concretely, we construct ClawGym-SynData, a diverse dataset of 13.5K filtered tasks synthesized from persona-driven intents and skill-grounded operations, paired with realistic mock workspaces and hybrid verification mechanisms. We then train a family of capable Claw-style models, termed ClawGym-Agents, through supervised fine-tuning on black-box rollout trajectories, and further explore reinforcement learning via a lightweight pipeline that parallelizes rollouts across per-task sandboxes.To support reliable evaluation, we further construct ClawGym-Bench, a benchmark of 200 instances calibrated through automated filtering and human-LLM review. Relevant resources will be soon released at https://github.com/ClawGym.

Token serves as the fundamental unit of computation in modern autoregressive models, and generation length directly influences both inference cost and reasoning performance. Despite its importance, existing approaches lack fine-grained length modeling, operating primarily at the coarse-grained sequence level. We introduce the Length Value Model (LenVM), a token-level framework that models the remaining generation length. By formulating length modeling as a value estimation problem and assigning a constant negative reward to each generated token, LenVM predicts a bounded, discounted return that serves as a monotone proxy for the remaining generation horizon. This formulation yields supervision that is annotation-free, dense, unbiased, and scalable. Experiments on LLMs and VLMs demonstrate LenVM provides a highly effective signal at inference time. On the LIFEBench exact length matching task, applying LenVM to a 7B model improves the length score from 30.9 to 64.8, significantly outperforming frontier closed-source models. Furthermore, LenVM enables continuous control over the trade off between performance and efficiency. On GSM8K at a budget of 200 tokens, LenVM maintains 63% accuracy compared to 6 percent for token budget baseline. It also accurately predicts total generation length from the prompt boundary. Finally, LenVM's token-level values offer an interpretable view of generation dynamics, revealing how specific tokens shift reasoning toward shorter or longer regimes. Results demonstrate that LenVM supports a broad range of applications and token length can be effectively modeled as a token-level value signal, highlighting the potential of LenVM as a general framework for length modeling and as a length-specific value signal that could support future RL training. Code is available at https://github.com/eric-ai-lab/Length-Value-Model.

We prove pathwise convergence of the layerwise evolution of tokens in a finite-depth, finite-width transformer model with MultiLayer Perceptron (MLP) blocks to a continuous-time stochastic interacting particle system. We also identify the stochastic partial differential equation describing the evolution of the tokens' distribution in this limit and prove propagation of chaos when the number of such tokens is large. The bounds we establish are quantitative and the limits we consider commute. We further prove that the limiting stochastic model displays synchronization by noise and establish exponential dissipation of the interaction energy on average, provided that the common noise is sufficiently coercive relative to the deterministic self-attention drift. We finally characterize the activation functions satisfying the former condition.

The Hypencoder, proposed by Killingback et al., is a retrieval framework that replaces the fixed inner-product scoring function used in standard bi-encoders with a query-specific neural network (the $q$-net), whose weights are generated by a hypernetwork from the contextualized query embeddings. This design enables more expressive relevance estimation while preserving independent query and document encoding. In this work, we conduct a reproducibility study of the Hypencoder and extend the original analysis in three directions. Our reproduction confirms that the Hypencoder outperforms a similarly trained bi-encoder baseline on in-domain and out-of-domain benchmarks, and that the proposed efficient search algorithm substantially reduces query latency with minimal performance loss. On hard retrieval tasks, we find partial support: the Hypencoder outperforms the baseline on DL-Hard and FollowIR, but not on TREC TOT, where checkpoint incompatibility and fine-tuning sensitivity complicate full verification. Beyond reproduction, we investigate three extensions: (i)~integrating alternative pre-trained encoders into the Hypencoder framework, where we find that performance gains depend on the encoder and fine-tuning strategy; (ii)~comparing query latency against a Faiss-based bi-encoder pipeline, revealing that standard bi-encoder retrieval remains faster under both exhaustive and efficient search settings; and (iii)~evaluating adversarial robustness, where we find that the $q$-net's non-linear scoring does not provide a consistent robustness disadvantage over inner-product scoring. Our code is publicly available at https://github.com/arneeichholtz/Hypencoder-reprod.

Despite the operational importance of hot fixes, large-scale evidence on how they reshape routine maintenance workflows, particularly in the era of autonomous coding agents, remains limited. We analyse hot fixes present in over 61,000 GitHub repositories from the Hao-Li/AIDev dataset and find consistent patterns of urgency: reduced collaboration (typically a single contributor), smaller and more targeted changes (median 2-3 commits and files, with <10 line modifications), limited review (often fewer than two reviewers), and substantially fewer test file modifications than regular bug fixes, consistent with their urgency-driven character. Leveraging the same urgency contexts, we examine differences between human- and AI-agent-authored hot fixes, revealing over 10 distinct repair behaviours, thus offering insights into future human-automation collaboration for hot fixing. Our study is the first to empirically analyse hot fix code changes at scale using a repository-level operationalisation of urgency. The comparison of human and agentbehaviours delineates their distinct characteristics, providing a foundation for understanding hot fixing in real-world practice

This paper extends and explains the Multiple Additive Neural Networks (MANN) methodology, an enhancement to the traditional Gradient Boosting framework, utilizing nearly shallow neural networks instead of decision trees as base learners. This innovative approach leverages neural network architectures, notably Convolutional Neural Networks (CNNs) and Capsule Neural Networks, to extend its application to both structured data and unstructured data such as images and audio. For structured data the advantages of capsule neural networks as feature extractors are used and combined with MANN as a classifier. MANN's unique architecture promotes continuous learning and integrates advanced heuristics to combat overfitting, ensuring robustness and reducing sensitivity to hyperparameter settings like learning rate and iterations. Our empirical studies reveal that MANN surpasses traditional methods such as Extreme Gradient Boosting (XGB) in accuracy across well-known datasets. This research demonstrates MANN's superior precision and generalizability, making it a versatile tool for diverse data types and complex learning environments.

Mixture-of-Experts (MoE) models offer high capacity with efficient inference cost by activating a small subset of expert models per input. However, deploying MoE models requires all experts to reside in memory, creating a gap between the resource used by activated experts and the provisioned resources. This underutilization is further pronounced in multi-tenant scenarios. In this paper, we propose FaaSMoE, a multi-tenant MoE serving architecture built on Function-as-a-Service (FaaS) platforms. FaaSMoE decouples the control and execution planes of MoE by deploying experts as stateless FaaS functions, enabling on-demand and scale-to-zero expert invocation across tenants. FaaSMoE further supports configurable expert granularity within functions, trading off per-expert elasticity for reduced invocation overhead. We implement a prototype with an open-source edge-oriented FaaS platform and evaluate it using Qwen1.5-moe-2.7B under multi-tenant workloads. Compared to a full-model baseline, FaaSMoE uses less than one third of the resources, demonstrating a practical and resource-efficient path towards scalable MoE serving in a multi-tenant environment.

Patient portals now give individuals direct access to their electronic health records (EHRs), yet access alone does not ensure patients understand or act on the complex clinical information contained in these records. The ArchEHR-QA 2026 shared task addresses this challenge by focusing on grounded question answering over EHRs, and this paper presents the system developed by the HealthNLP_Retrievers team for this task. The proposed approach uses a multi-stage cascaded pipeline powered by the Gemini 2.5 Pro large language model to interpret patient-authored questions and retrieve relevant evidence from lengthy clinical notes. Our architecture comprises four integrated modules: (1) a few-shot query reformulation unit which summarizes verbose patient queries; (2) a heuristic-based evidence scorer which ranks clinical sentences to prioritize recall; (3) a grounded response generator which synthesizes professional-caliber answers restricted strictly to identified evidence; and (4) a high-precision many-to-many alignment framework which links generated answers to supporting clinical sentences. This cascaded approach achieved competitive results. Across the individual tracks, the system ranked 1st in question interpretation, 5th in answer generation, 7th in evidence identification, and 9th in answer-evidence alignment. These results show that integrating large language models within a structured multi-stage pipeline improves grounding, precision, and the professional quality of patient-oriented health communication. To support reproducibility, our source code is publicly available in our GitHub repository

Deep learning for cross-subject EEG decoding is hindered by high inter-subject variability, which introduces a severe domain shift between training and unseen test subjects. This survey presents a comprehensive review of deep learning methodologies specifically engineered to address this cross-subject generalization challenge. To ground this analysis, we formalize the cross-subject setting as a multi-source domain problem and delineate the rigorous, subject-independent evaluation protocols required for valid assessment. Central to this survey is a systematic taxonomy of the current literature into discrete methodological families, including feature alignment, adversarial learning, feature disentanglement, and contrastive learning. We conclude by examining three critical elements for advancing robust, real-world decoding: the theoretical limitations of current methodologies, the structural value of subject identity, and the emergence of EEG foundation models.

Large Language Models (LLMs) have become an integral part of many real-world workflows. However, LLMs consume a lot of energy, which becomes a large concern in the scale of the demand for these tools. As LLMs become integrated into different workflows, different applications have arisen to deal with the challenge of running inference for these tools. This raises another issue of choosing the runtime parameter values for these services in order to minimize the energy consumption. Oftentimes this requires deep knowledge of the application or traditional optimization methods that can take days to find optimal values. In this work, we created a human-in-the-loop flow with LLM-assisted runtime parameter optimization in order to solve this issue. With human-created, specific feedback prompting methods, chat-based LLMs can iteratively find energy-efficient inference parameters faster than traditional search methods. LLMs can also tailor their solutions to different hardware setups and easily take into account other system constraints. The enhanced prompt template was able to converge below the threshold at an average of 3.4 prompts compared to the baseline, which converged in an average of 5.2 prompts, and consistently achieved lower final energy per token. The enhanced prompt template also outperformed Sobol sampling in convergence speed.

In a continual learning setting, we require a model to be plastic enough to learn a new task and stable enough to not disturb previously learned capabilities. We argue that this dilemma has an architectural root. A finite network has limited representational and plastic resources, yet the required capacity depends on properties of the future task stream that are unknown: how many tasks will be encountered, and how much they overlap in feature space. Regularization-based methods preserve past knowledge within fixed-capacity architectures and therefore implicitly rely on an oracle architecture sized for this unknown future. When tasks are only weakly related, fixed architectures progressively run out of plastic resources; when tasks are few or strongly overlapping, models are often over-provisioned. Inspired by neurogenesis in biology, we propose NORACL to address the stability-plasticity dilemma by tackling the oracle architecture problem through neuronal growth. Starting from a compact network, NORACL grows only when needed by monitoring two complementary signals for representational and plasticity saturation. We evaluate NORACL against oracle-sized static baselines across varying task counts and geometries. Across all settings, NORACL achieves final average accuracies that are better than or on par with oracle-provisioned static baselines while using fewer parameters. Additionally, NORACL yields architectures with interpretable growth, i.e. dissimilar tasks predominantly expand feature-extraction layers, whereas tasks which rely on common features shift growth toward later feature-combination layers. Our analysis further explains why fixed-capacity networks lose plasticity as tasks accumulate, whereas NORACL creates fresh capacity for new tasks through growth. Together, these results show that adaptive neurogenesis pushes the stability-plasticity Pareto frontier of continual learning.

Automatic feature engineering is an effective approach for improving predictive performance in tabular learning. However, expand-and-reduce methods, such as OpenFE, become increasingly computationally expensive as the input dimensionality grows. This limitation arises primarily from the combinatorial explosion of candidate features generated through operator-feature combinations. To address this issue, we propose SCOPE-FE, a structured search space control framework that improves efficiency by reducing the candidate space prior to feature generation. SCOPE-FE jointly regulates two major sources of combinatorial growth: the operator space and feature-pair space. First, OperatorProbing estimates the dataset-specific utility of candidate operators and eliminates low-contribution operators in advance. Second, FeatureClustering employs spectral embedding and fuzzy c-means clustering to group structurally related features, thereby restricting candidate generation to relevant within-cluster combinations. In addition, we introduce ReliabilityScoring, which incorporates variance across subsamples to stabilize pruning decisions. Experiments on ten benchmark datasets demonstrate that SCOPE-FE substantially reduces feature engineering time while maintaining competitive predictive performance relative to existing baselines. The efficiency gains are particularly pronounced for high-dimensional datasets. These results indicate that structured control of the search space is an effective strategy for scalable automatic feature engineering. The code will be made publicly available upon acceptance.

Safety-aligned language models must refuse harmful requests without collapsing into broad over-refusal, but the training-time mechanisms behind this tradeoff remain unclear. Prior work characterizes refusal directions and jailbreak robustness, yet does not explain how dynamic adversarial fine-tuning changes refusal carriers across training. We present a measurement-driven mechanism study, not a new defense, on one 7B backbone under supervised fine-tuning (SFT) and R2D2-style dynamic adversarial fine-tuning. Our protocol aligns fixed-source HarmBench, StrongREJECT, and XSTest with a five-anchor refusal-geometry suite and causal interventions. R2D2 drives fixed-source HarmBench ASR to 0.000 at steps 50 and 100, then partially reopens to 0.035 at step 250 and 0.250 at step 500; SFT remains less robust, with ASR between 0.505 and 0.588 at the same anchors. On XSTest, R2D2 any-refusal is 1.000 early, then falls to 0.664 and 0.228. Geometrically, R2D2 preserves a late-layer admissible carrier through step 100 before relocating to an early-layer carrier, while effective rank remains near 1.23--1.27. Causal interventions indicate low-dimensional but utility-coupled control. These results support a reorganization account rather than a drift-only account, with evidence limited to one backbone and fixed-source attacks.

Deep learning models for 12-lead electrocardiogram (ECG) analysis achieve high diagnostic performance but lack the intuitive interpretability required for clinical integration. Standard feature attribution methods are limited by the inherent difficulty in mapping abstract waveform fluctuations to physical anatomical pathologies. To resolve this, we propose a cross-modal method that projects feature attributions from high-performance 12-lead ECG models onto the CineECG 3D anatomical space. Our study reveals that while models trained directly on CineECG signals suffer from reduced accuracy and incoherent attributions, the proposed mapping mechanism effectively recovers clinically relevant feature rankings. Validated against a ground-truth dataset of 20 cases annotated by domain experts, the mapped explanations yield a Dice score of 0.56, significantly outperforming the 0.47 baseline of standard 12-lead attributions. These findings indicate that cross-modal averaging mapping effectively filters attribution instability and improves the localization of pathological features, combining the diagnostic expressiveness of standard ECG with the intuitive clarity of anatomical visualization.

The scarcity of high-quality annotated medical data, particularly in mental health, poses a significant bottleneck for training robust machine learning models. Privacy regulations restrict data sharing, making synthetic data generation a promising alternative. The use of Large Language Models (LLMs) in a data augmentation pipeline could be leveraged as an alternative in this field. In the proposed methodology, DeepSeek-R1, OpenBioLLM-Llama3 and Qwen 3.5 are used to generate synthetic mental health evaluation reports conditioned on specific International Classification of Diseases, Tenth Revision (ICD-10) codes. Because naive text generation can lead to mode collapse or privacy breaches (memorization), a comprehensive evaluation framework is introduced. The generated diagnostic texts are assessed across three dimensions: semantic fidelity, lexical diversity, and privacy/plagiarism. The results demonstrate that all models can generate clinically coherent, diverse, and privacy-safe synthetic reports, significantly expanding the available training data for clinical natural language processing tasks without compromising patient confidentiality.

The Barcelona Zetascale Lab (BZL) project aims to strengthening Europe's capacity in the design and manufacture of RISC-V based high-performance computing chips. In this context, we present a holistic pre-silicon verification and validation (V&V) methodology targeting highly robust RISC-V chip designs. This paper provides an overview of BZL's V&V approach, which integrates three complementary platforms: (1) a UVM-based verification environment to thoroughly validate RTL functionality; (2) an FPGA-based validation platform that enables system-level pre-silicon hardware-software RTL validation; and (3) a CI/CD flow that continuously automates build, deployment, and tests across these domains. By embedding these platforms into an industrial-grade V&V loop and exploiting large-scale CPU and FPGA hardware infrastructures, the BZL project enables continuous evolution of reliable hardware development and software integration. We believe that the BZL's V&V flow represents a robust and scalable foundation for ensuring the pre-silicon functional correctness and system level validation of RISC-V chip designs, and can serve as a key enabler for strategic initiatives in Europe, such as EPI and DARE, and beyond.

FPGA-level emulation is a key step in pre-silicon chip design validation. However, emulating large-scale multi-core systems increasingly exceed the hardware resource capacity of a single FPGA, limiting the feasibility of full-system emulation. To address this challenge, we introduce EMiX, a scalable multi-FPGA framework that enables distributed emulation of multi-core RISC-V architectures beyond single-FPGA resource limits. EMiX systematically partitions a monolithic multi-core design into multiple components and deploys them across multiple interconnected FPGAs, effectively exploiting inter-FPGA interconnects to balance scalability and performance without requiring fundamental RTL redesign. We prototype EMiX with a 64-core architecture across eight interconnected Alveo U55c FPGAs (scalable on core and FPGA counts), successfully demonstrating full-system execution including Linux boot. EMiX will be released as an open-source platform.

AutoML, intended as the process of automating the application of machine learning to real-world problems, is a key step for AI popularisation. Most AutoML frameworks are not accounting for the potential lack of fairness in the training data and in the corresponding predictions. We introduce \textsc{FairMind}, a software prototype aiming to automatise fairness analysis at the dataset level. We achieve that by resorting to the assumptions of the \emph{standard fairness model}, recently proposed by Plečko and Bareinboim. This allows for a sound fairness evaluation in terms of causal effects, based on \emph{counterfactual} queries involving the target, possibly confounders and mediators, and the different values of an input feature we regard as \emph{protected}. After the necessary data preprocessing, the tool implements a closed-form computation of the effects. LLMs are consequently exploited to generate accurate reports on the fairness levels detected in the training dataset. We achieve that in a zero-shot setup and show by examples the expected advantages with respect to a direct analysis performed by the LLM. To favour applications, extensions to ordinal protected variable and continuous targets and novel decomposition results are also discussed.

Large Language Models (LLMs) have become widely used for Software Engineering (SE) tasks, spanning from function-level code generation to complex repository-level workflows. However, the high latency of autoregressive inference remains a significant bottleneck, hindering their deployment in interactive environments. While Speculative Decoding (SD) offers a promising technique for lossless acceleration, prior research on long-context repository-level tasks and complex agentic interactions remains limited. To bridge this gap, we present the first systematic empirical study to evaluate the effectiveness of SD in SE tasks. We systematically benchmark a comprehensive spectrum of strategies, encompassing both model-based and model-free methods, across representative generation, editing, and repair scenarios. Our empirical results indicate that SD demonstrates clear potential for accelerating inference, particularly for smaller models that achieve higher speedups than those of their larger counterparts. We find that the effectiveness of SD methods varies across different task scenarios. Model-based approaches are well-suited for code generation, whereas model-free methods are better adapted to repository-level repair and editing scenarios. Furthermore, we observe that the repetitiveness of SE tasks improves the performance of model-free methods. In contrast to natural language tasks, the higher predictability of SE tasks allows for more aggressive hyperparameters. Our findings are summarized as guidelines to help increase inference efficiency for SE scenarios.

The Airborne Collision Avoidance System Xu (ACAS-Xu) relies on large certified Look-Up Tables (LUTs) that encode the exact decision logic used in operation. Neural-network-based approximations have been proposed to reduce memory requirements, but they inherently introduce approximation errors and complicate formal verification. This paper presents a symbolic compression approach based on Binary Decision Diagrams (BDDs) that preserves the exact semantics of the ACAS-Xu LUTs. The resulting representation is canonical, deterministic, and fully equivalent to the original tables, enabling sound and exact reasoning over the complete decision logic. By expressing both the system behavior and domain-specific operational properties within a common Boolean framework, verification reduces to efficient BDD operations and emptiness checks, with precise counterexamples generated when properties are violated. We demonstrate that the proposed BDD-based representation significantly reduces memory usage, achieves predictable and low-latency execution, and can be deployed on embedded platforms. These results highlight BDDs as a compelling alternative for exact, verifiable, and embedded deployment of ACAS-Xu decision logic.

We provide a causal analysis of Binary Spiking Neural Networks (BSNNs) to explain their behavior. We formally define a BSNN and represent its spiking activity as a binary causal model. Thanks to this causal representation, we are able to explain the output of the network by leveraging logic-based methods. In particular, we show that we can successfully use a SAT as well as a SMT solver to compute abductive explanations from this binary causal model. To illustrate our approach, we trained the BSNN on the standard MNIST dataset and applied our SAT-based and SMT-based methods to finding abductive explanations of the network's classifications based on pixel-level features. We also compared the found explanations against SHAP, a popular method used in the area of explainable AI. We show that, unlike SHAP, our approach guarantees that a found explanation does not contain completely irrelevant features.

Context: Study screening in systematic literature reviews is costly, inconsistency-prone, and risk-asymmetric, since false negatives can compromise validity. Despite rapid uptake of Large Language Models (LLMs), there is limited evidence on how such models behave during the study screening phase, particularly regarding the choice of specific LLMs and their comparison with classical models. Objective: To assess LLM performance and variability in screening, quantify the impact of input metadata (abstract, title, keywords), and compare LLMs with classical classifiers under a shared protocol. Methods: We analyzed 12 LLMs from 4 providers (OpenAI, Google Gemini, Anthropic, Llama) and 4 classical models (Logistic Regression, Support Vector Classification, Random Forest, and Naive Bayes) on 2 real Systematic Literature Reviews (SLRs), totaling 518 papers. The experimental design investigated 3 critical dimensions: (i) LLMs performance variability, (ii) the impact of input feature composition (abstract, title, and keywords) on LLM performance, and (iii) the real gain of using LLMs instead of more traditional classification models. Results: LLMs exhibited substantial heterogeneity and residual non-determinism even at temperature zero. Abstract availability was decisive: removing it consistently degraded performance, while adding title and/or keywords to the abstract yielded no robust gains. Compared to classical models, performance differences were not consistent enough to support generalizable LLM superiority. Discussion: LLM adoption should be justified by operational and governance constraints (reproducibility, cost, metadata availability), supported by pilot validation and explicit reporting of variability and input configuration.

Reasoning in Large Language Models incurs significant inference-time compute, yet the token-level information structure of reasoning traces remains underexplored. We observe that reasoning tokens split into two functional types: low-entropy \textit{structural} tokens (recurring phrases that scaffold the reasoning process) and higher-entropy \textit{organic} tokens (problem-specific content that drives toward a solution). This asymmetry motivates a simple, model-agnostic compression pipeline: apply cross-word BPE merges on a model's own reasoning traces to derive \textit{supertokens} that capture frequent structural patterns, then teach the model to adopt them via supervised fine-tuning. Across three model families and five mathematical reasoning benchmarks, our approach shortens reasoning traces by 8.1\% on average with no statistically significant accuracy loss on any model--benchmark pair. Beyond compression, supertokens act as interpretable reasoning-move annotations (backtracking, verification, strategy shifts), exposing the model's high-level strategy at a glance. Analyzing transitions between structural categories reveals systematic differences between correct and incorrect traces: correct traces show productive recovery (backtracking followed by strategy shifts and verification), while incorrect traces are dominated by confusion cycles (repeated hedging and unresolved contradictions). These diagnostic signals suggest applications in reward shaping and early stopping for RL-based reasoning training.

We propose EdgeSpike, a co-designed spiking neural network (SNN) framework for autonomous low-power sensing in edge Internet of Things (IoT) architectures. EdgeSpike unifies (i) a hybrid surrogate-gradient and direct-encoding training pipeline, (ii) a hardware-aware neural architecture search (NAS) bounded by per-inference energy and memory budgets, (iii) an event-driven runtime targeting Intel Loihi 2, SpiNNaker 2, and commodity ARM Cortex-M microcontrollers with custom spike-sparse SIMD kernels, and (iv) a lightweight local plasticity rule enabling continual on-device adaptation without backpropagation. The framework is evaluated across five sensing tasks (keyword spotting, vibration-based machine fault detection, surface electromyography gesture recognition, 77 GHz radar human-activity classification, and structural-health acoustic-emission monitoring) on three hardware targets. EdgeSpike achieves a mean classification accuracy of 91.4%, within 1.2 percentage points (pp) of strong INT8 convolutional neural network (CNN) baselines (mean 92.6%), while reducing energy per inference by 18x to 47x on neuromorphic hardware (mean 31x) and by 4.6x to 7.9x on Cortex-M (mean 6.1x). End-to-end latency remains at or below 9.4 ms across all 15 task-hardware configurations. A seven-month, 64-node wireless field deployment confirms a 6.3x extension in projected battery lifetime (from 312 to 1978 days at 2 Wh per node) and bounded accuracy degradation under seasonal drift (0.7 pp with on-device adaptation versus 2.1 pp without). Hardware-aware NAS evaluates 8400 candidates and yields a 12-point Pareto front. EdgeSpike will be released as open source with reproducible training pipelines, hardware-portable runtimes, and benchmark suites.

Memory-augmented LLM agents offer an appealing shortcut to continual learning: rather than updating model parameters, they accumulate experience in external memory, seemingly sidestepping the stability-plasticity dilemma of parametric learning. We show that this challenge does not disappear but resurfaces at the memory level. Under a limited context window, old and new experiences compete during retrieval, relocating the continual-learning bottleneck from parameter updates to memory access. To study this phenomenon, we introduce a (k,v) framework that disentangles two fundamental design axes of external memory: how experience is represented and how it is organized for retrieval. Across sequential-task experiments in ALFWorld and BabyAI, we find that abstract procedural memories transfer more reliably than detailed trajectories, while negative transfer disproportionately harms the hard cases. Moreover, finer-grained memory organization is not universally beneficial: designs that yield strong forward transfer can simultaneously induce severe forgetting. Together, these results reveal that external memory does not resolve the continual-learning problem; it reshapes it into a problem of memory representation and retrieval design.

Chest X-ray interpretation is one of the most frequently performed diagnostic tasks in medicine and a primary target for AI development, yet current vision-language models are primarily trained on datasets of paired images and reports, not the cognitive processes and visual attention that underlie clinical reasoning. Here, we present CheXthought, a global, multimodal resource containing 103,592 chain-of-thought reasoning traces and 6,609,082 synchronized visual attention annotations across 50,312 multi-read chest X-rays from 501 radiologists in 71 countries. Our analysis reveals clinical reasoning patterns in how experts deploy distinct visual search strategies, integrate clinical context, and communicate uncertainty. We demonstrate the clinical utility of CheXthought across four dimensions. First, CheXthought reasoning significantly outperforms state-of-the-art vision-language model chain-of-thought in factual accuracy and spatial grounding. Second, visual attention data used as an inference-time hint recovers missed findings and significantly reduces hallucinations. Third, vision-language models trained on CheXthought data achieve significantly stronger pathology classification, visual faithfulness, temporal reasoning and uncertainty communication. Fourth, leveraging CheXthought's multi-reader annotations, we predict both human-human and human-AI disagreement directly from an image, enabling transparent communication of case difficulty, uncertainty and model reliability. These findings establish CheXthought as a resource for advancing multimodal clinical reasoning and the development of more transparent, interpretable vision-language models.

Developers and organizations are using Large Language Models (LLMs) to generate security-critical code more frequently than ever, including cryptographic solutions for their products. This study presents an empirical evaluation of cryptographic security in 240 Rust code samples for two crypto algorithms (AES-256-GCM and ChaCha20-Poly1305) generated by three LLMs (Gemini 2.5 Pro, GPT-4o, and DeepSeek Coder) using four different prompt strategies. For each successfully compiled code sample, CodeQL static analysis and our rule-based crypto-specific analyzer were used to detect vulnerabilities, which are also associated with Common Weakness Enumeration (CWE). The evaluation results revealed that only 23.3% of the generated code samples were successfully compiled. Among the compiled code, CodeQL produced only two false positives, while our rule-based crypto-specific analyzer identified vulnerabilities in 57% of the compiled samples with zero false positives. This demonstrates that general-purpose analysis tools are insufficient for code validation for the experimented crypto algorithms. The compilation success of the two algorithms varied significantly (AES-256-GCM 34.2% versus ChaCha20-Poly1305 12.5%), showing a gap in code generation capabilities. While model choice had no significant effect on compilation success, prompt strategy significantly influenced outcomes (P = 0.002), with chain-of-thought prompting performing 5 times worse than zero-shot. All three models exhibit systematic failures, including nonce reuse and API hallucinations.

Modern software systems are increasingly developed within rapid continuous integration and deployment (CI/CD) pipelines, where ensuring security prior to release presents significant technical and organizational challenges. Traditional static and dynamic analysis tools provide valuable structural and behavioral insights, yet they often operate in non-adaptive workflows and produce large volumes of warnings requiring manual triage. Feedback-driven fuzzing and search-based testing approaches have demonstrated the power of iterative input refinement guided by execution signals, while large language models (LLMs) have shown promise in automated test generation but frequently lack semantic grounding in program structure. This paper presents a systematic survey of adaptive and AI-augmented security testing research across five domains: (1) structural program analysis for vulnerability detection, (2) DevSecOps and continuous security testing, (3) feedback-driven fuzzing and search-based testing, (4) LLM-based automated test generation, and (5) emerging hybrid systems integrating program analysis with adaptive learning. We analyze fifty-five peer-reviewed studies drawn from a systematic search of four major databases yielding 22,088 raw records. Our analysis reveals a persistent disconnect between structural program representations (ASTs, CFGs, and CPGs) and adaptive testing mechanisms. We characterize this as structural-adaptive fragmentation: a systematic separation that neither paradigm individually addresses. No existing system incorporates human triage signals as feedback for refining structural models. We conclude by identifying five open research challenges and outlining a unified agenda for semantically grounded, feedback-driven, polyglot security testing frameworks.

Physics-informed neural networks (PINNs) approximate solutions of partial differential equations (PDEs) by embedding physical laws into the loss function. In parameterized PDE families, variations in coefficients or boundary/initial conditions define distinct tasks. This makes training individual PINNs for each task computationally prohibitive, while cross-task transfer can be sensitive to task heterogeneity. While meta-learning can reduce retraining cost, existing methods often rely on a single global initialization and may suffer from negative transfer, particularly under feature-scarce coordinate inputs and limited training-task availability. We propose the Learning-Affinity Adaptive Modular Physics-Informed Neural Network (LAM-PINN), a compositional framework that leverages task-specific learning dynamics. LAM-PINN combines PDE parameters with learning-affinity metrics from brief transfer sessions to construct a task representation and cluster tasks even with coordinate-only inputs. It decomposes the model into cluster-specialized subnetworks and a shared meta network, and learns routing weights to selectively reuse modules instead of relying on a single global initialization. Across three PDE benchmarks, LAM-PINN achieves an average 19.7-fold reduction in mean squared error (MSE) on unseen tasks using only 10% of the training iterations required by conventional PINNs. These results indicate its effectiveness for generalization to unseen configurations within bounded design spaces of parameterized PDE families in resource-constrained engineering settings.

Automated depression detection often relies on static aggregation of conversational signals, potentially obscuring clinically meaningful behavioral dynamics. We investigated whether entropy-driven temporal biomarkers improve depression detection beyond standard pooled features using the DAIC-WOZ corpus. Using 142 labeled participants, we reconstructed utterance-level acoustic trajectories and compared pooled temporal baselines, trajectory dynamics, Shannon entropy biomarkers, recurrence quantification, sample entropy, fractal complexity, and coupling biomarkers under leakage-aware validation. Static pooling achieved an AUC of 0.593, trajectory dynamics improved performance to 0.637, and entropy biomarkers produced the strongest statistically significant improvement over pooled baselines (AUC 0.646; nested cross-validated AUC 0.615; permutation p = 0.017). Entropy biomarkers outperformed recurrence, coupling, sample entropy, and fractalbased features, with several biomarkers stable across folds. These findings suggest depression-related signal may lie less in average acoustic levels than in entropy of conversational dynamics, supporting temporally informed digital phenotypes for mental-health assessment.

Autonomous AI agent ecosystems require stronger mechanisms for secure discovery, identity verification, capability attestation, and policy governance. Current deployments frequently lack (1) uniform agent discovery, (2) cryptographic agent authentication, (3) capability proofs that protect secrets, and (4) enforceable policy controls. This paper presents an implementation-oriented proof of concept for the Agent Name Service (ANS), a DNS-inspired trust layer for AI agent discovery and interoperability in Kubernetes, grounded in the ANS protocol specification~\cite{huang2025ans}. The implementation uses Decentralized Identifiers (DIDs), Verifiable Credentials (VCs), policy-as-code enforcement with Open Policy Agent (OPA), and Kubernetes-native integration patterns (CRDs, admission controls, service mesh integration). In a demo research environment (3-node cluster, 50-agent workflow simulation), we observe sub-10ms response in demonstrated service paths and full success for scripted demo deployment scenarios. We explicitly scope these findings as proof-of-concept evidence rather than production certification. We further provide a threat model, assumptions, and limitations to separate implemented evidence from protocol-defined and roadmap capabilities. The result is an evidence-grounded pathway from ANS protocol concepts to reproducible engineering practice for secure multi-agent systems.

We study gradient descent for rank-1 matrix factorization through a certificate-based viewpoint. The central object is a parameterized quadratic certificate $I(δ;\,\cdot)$ whose level sets shrink along the dynamics, thereby inducing a monotone state parameter $δ_t$. In the certified regime, this mechanism yields convergence to a global minimizer; in the post-critical regime, it forces trajectories toward a terminal balanced manifold. To explain the origin of these certificates, we formulate a state-dependent Lyapunov framework based on structural axioms. Within this framework, the scalar certificate is uniquely determined, and the same local Lagrange analysis constrains the signal and noise blocks of rank-1 extensions. Thus, the certificates arise from the monotonicity structure of the dynamics, rather than from ad hoc algebraic constructions. We also provide numerical evidence beyond the proved cases. For the 2-dimensional rank-1 approximation problem $X=\mathrm{diag}(1,σ)$ with $σ\in(0,1)$, the experiments are consistent with the existence of a $C^1$ admissible certificate branch. For the quartic-augmented scalar loss $\frac12(ab-1)^2+μ(ab-1)^4$, the same scalar certificate remains predictive for several values of $μ$ after choosing an empirical threshold. These experiments suggest that the state-dependent Lyapunov method may extend beyond the settings proved in this paper.

Recent advances in data-centric medical AI have produced highly accurate diagnostic systems, but the emphasis on data curation and performance metrics has not translated into widespread clinical adoption. We conjecture that this limited uptake stems from insufficient attention dedicated to the optimisation of fair performance across diverse patient populations and to workflow integration: performance biases can create regulatory barriers, and poorly integrated automation can disrupt clinical routines, degrade the quality of human-AI collaboration, and reduce clinicians' willingness to adopt AI tools. Prior work on workflow integration (e.g., Learning to Defer (L2D) and Learning to Complement (L2C)) and AI fairness has typically examined these challenges in isolation, overlooking their natural interdependence and the practical constraints of clinical environments, such as restricted clinician availability. We propose People-Centred Medical Image Analysis (PecMan), a human-AI framework that jointly optimises fairness, diagnostic accuracy, and workflow effectiveness through a dynamic gating mechanism that assigns cases to AI, clinicians, or both under clinician workload constraints. We also introduce the Fairness and Human-Centred AI (FairHAI) benchmark for evaluating trade-offs between accuracy, fairness, and clinician workload. Experiments using this benchmark show that PecMan consistently outperforms existing methods, paving the way for more trustworthy and clinically viable AI systems. Code will be available upon paper acceptance.

With the rapid growth of large language models for code generation, distinguishing between human-written and AI-generated code has become increasingly critical for academic integrity, hiring evaluations, and software security. We present our system for SemEval-2026 Task 13: Multilingual Machine-Generated Code Detection, participating in Subtask A (binary detection) and Subtask B (multi-class attribution across 10 LLM families). For Subtask A, we fine-tune UniXcoder-base with a multi-view training framework that promotes generator-invariant representations. The framework combines domain-specific structural prefixes, delexicalization with symmetric KL consistency loss, token dropout, and mixed-content augmentation. Our system achieves 0.993 macro F1 on validation and 0.845 macro F1 on the test set, which spans unseen languages and domains. For Subtask B, we show that severe class imbalance (88.4% human code, 221:1 majority-to-minority ratio) causes catastrophic minority-class failure under standard fine-tuning, with macro F1 collapsing to 0.086 despite 88.4% accuracy. A class-weighted extension trained for 3 epochs recovers macro F1 to 0.345 (+301% relative), confirming that multi-class attribution requires imbalance-aware training strategies.

We introduce a novel task of digital battery passport (DBP) conformance classification and introduce the first public benchmark for the task: BatteryPass-12K, created synthetically from real pilot samples. This is as the EU's battery regulation on DBPs comes into effect soon and there exists no public dataset. We evaluated 22 language models (LMs) in zero-shot inference, spanning small LMs (SLMs), mixture of experts (MoEs), and dense LLMs. We also conducted analysis, additional evaluations of few-shot inference and prompt-injection attacks to find that (1) Thinking models have the best performance (with GPT-5.4 scoring 0.98 (0.03) and 0.71 (0.22) on average as F1 (and confidence interval at 95%) on the validation and test sets, respectively), (2) few-shot examples improve performance significantly, (3) generally capable frontier models find the task challenging, (4) merely scaling model parameters does not necessarily lead to improved performance, as SLMs outperformed some LLMs, and (5) prompt-injection attacks degrade performance. We note that BatteryPass-12K, though limited to real pilot samples, may be useful for other known or emerging tasks in the battery domain, e.g. lifecycle reasoning. We publicly release the dataset under a permissive licence (CC-BY-4.0).

All current LLM serving systems place the GPU at the center, from production-level attention-FFN disaggregation to NVIDIA's Rubin GPU-LPU heterogeneous platform. Even academic PIM/PNM proposals still treat the GPU as the central hub for cross-device communication. Yet the GPU's compute-rich architecture is fundamentally mismatched with the memory-bound nature of decode-phase attention, inflating serving latency while wasting power and die area on idle compute units. The problem is compounded as reasoning and agentic workloads push context lengths toward one million tokens, making attention latency the primary user-facing bottleneck. To address these inefficiencies, we present AMMA, a multi-chiplet, memory-centric architecture for low-latency long-context attention. AMMA replaces GPU compute dies with HBM-PNM cubes, roughly doubling the available memory bandwidth to better serve memory-bound attention workloads. To translate this bandwidth into proportional performance gains, we introduce (i) a logic-die microarchitecture that fully exploits per-cube internal bandwidth for decode attention under a minimal power and area budget, (ii) a two-level hybrid parallelism scheme, and (iii) a reordered collective flow that reduces intra-chip die-to-die communication overhead. We further conduct a design-space exploration over per-cube compute power and intra-chip D2D link bandwidth, providing actionable guidance for hardware designers. Evaluations show that AMMA achieves 15.5X lower attention latency and 6.9X lower energy consumption compared with the NVIDIA H100.

Masked diffusion models (MDMs) generate discrete sequences by iterative denoising under an absorbing masking process. In standard masked diffusion, if a token remains masked after a reverse update, the model discards its clean-state prediction for that position. Thus, still-masked positions must be repeatedly inferred from the mask token alone. This design choice limits cross-step refinement. To address this limitation, this paper proposes a simple, yet effective, post-training adaptation for MDMs that conditions each denoising step on the model's own previous clean-state predictions. The resulting method, called Self-Conditioned Masked Diffusion Models (SCMDM), requires minimal architectural change, does not introduce a recurrent latent-state pathway, does not rely on an auxiliary reference model, and adds no extra denoiser evaluations during sampling. This is an important departure from partial self-conditioning approaches which requires expensive model training from scratch. In particular, the paper shows that partial self-conditioning, including the commonly used 50% dropout strategy for training self-conditioned models from scratch, is suboptimal in the post-training regime. Instead, once the model's self-generated clean-state estimates become informative, the specialization to refinement is preferable to mixing conditional and unconditional objectives. SCMDM is evaluated across multiple domains, demonstrating consistent improvement over vanilla MDM baselines, achieving nearly a 50% reduction in generative perplexity on OWT-trained models (42.89 to 23.72), alongside strong improvements in discretized image synthesis quality, small molecular generation, and enhanced fidelity in genomic distribution modeling.

Representational collapse, where embeddings become anisotropic and lose multi-scale structure, can erode downstream performance long before performance metrics react. We propose an online, topology-aware monitor for evolving neural representations that couples Modular Morse Homology Maintenance (MMHM) with a composite Collapse Index (CI). Instead of rebuilding complexes each epoch, we apply sparse edits at a fixed scale and maintain a discrete Morse matching, yielding fast, incremental updates. Across LLM fine-tuning and temporal KGE training, CI provides a low-latency early-warning signal suitable for in-training interventions. Code and experimental scripts will be released publicly

This paper presents a novel value-aware approach to product recommendation that simultaneously addresses the high dimensionality and sparsity of user-item data while explicitly incorporating the contribution of each product and user to overall sales revenue. The proposed framework encodes revenue contributions in the user-item matrix and computes customer similarity directly on this basis using suitable distance measures. This enables the segmentation of users according to the revenue-based similarity of their purchase baskets and supports recommendations aligned with profitability objectives. We compare conventional similarity metrics with a novel alternative tailored to high-dimensional contexts and propose three recommendation strategies based on revenue share, product popularity, and expected profit generation. The effectiveness of the proposed method is validated through simulation experiments and a real-world application using the UCI Online Retail dataset.

Harnesses have become a central determinant of coding-agent performance, shaping how models interact with repositories, tools, and execution environments. Yet automating harness engineering is hard: a heterogeneous action space, sparse and noisy evaluation signal, multi-million-token trajectories, and edits whose effect is hard to attribute to the next round's outcomes. We introduce Agentic Harness Engineering (AHE), a framework that automates harness-level evolution by instrumenting the three stages of any engineering loop (component editing, trajectory inspection, and decision making) with matched observability pillars: (1) component observability gives every editable harness component a file-level representation so the action space is explicit and revertible; (2) experience observability distills millions of raw trajectory tokens into a layered, drill-down evidence corpus that an evolving agent can actually consume; and (3) decision observability pairs every edit with a self-declared prediction, later verified against the next round's task-level outcomes. Together, these pillars turn every edit into a falsifiable contract, so harness evolution proceeds autonomously without collapsing into trial-and-error. Empirically, ten AHE iterations lift pass@1 on Terminal-Bench 2 from 69.7% to 77.0%, surpassing the human-designed harness Codex-CLI (71.9%) and the self-evolving baselines ACE and TF-GRPO. The frozen harness transfers without re-evolution: on SWE-bench-verified it tops aggregate success at 12% fewer tokens than the seed, and on Terminal-Bench 2 it yields +5.1 to +10.1pp cross-family gains across three alternate model families, indicating the evolved components encode general engineering experience rather than benchmark-specific tuning. These results position observability-driven evolution as a practical pathway to keep coding-agent harnesses continually improving.

Large Language Models (LLMs) have revolutionized the field of natural language processing. However, they exhibit some limitations, including a lack of reliability and transparency: they may hallucinate and fail to provide sources that support the generated output. Retrieval-Augmented Generation (RAG) was introduced to address such limitations in LLMs. One popular implementation, RAG-as-a-Service (RaaS), has shortcomings that hinder its adoption and accessibility. For instance, RaaS pricing is based on the number of submitted prompts, without considering whether the prompts are enriched by relevant chunks, i.e., text segments retrieved from a vector database, or the quality of the utilized chunks (i.e., their degree of relevance). This results in an opaque and less cost-effective payment model. We propose Chunk-as-a-Service (CaaS) as a transparent and cost-effective alternative. CaaS includes two variants: Open-Budget CaaS (OB-CaaS) and Limited-Budget CaaS (LB-CaaS), which is enabled by our ``Utility-Cost Online Selection Algorithm (UCOSA)''. UCOSA further extends the cost-effectiveness and the accessibility of the OB-CaaS variant by enriching, in an online manner, a subset of the submitted prompts based on budget constraints and utility-cost tradeoff. Our experiments demonstrate the efficacy of the proposed UCOSA compared to both offline and relevance-greedy selection baselines. In terms of the performance metric-the number of enriched prompts (NEP) multiplied by the Average Relevance (AR)-UCOSA outperforms random selection by approximately 52% and achieves around 75% of the performance of offline selection methods. Additionally, in terms of budget utilization, LB-CaaS and OB-CaaS achieve higher performance-to-budget ratios of 140% and 86%, respectively, compared to RaaS, indicating their superior efficiency.

Source code and its accompanying comments are complementary yet naturally aligned modalities-code encodes structural logic while comments capture developer intent. However, existing vulnerability detection methods mostly rely on single-modality code representations, overlooking the complementary semantic information embedded in comments and thus limiting their generalization across complex code structures and logical relationships. To address this, we propose MultiVul, a multimodal contrastive framework that aligns code and comment representations through dual similarity learning and consistency regularization, augmented with diverse code-text pairs to improve robustness. Experiments on widely adopted DiverseVul and Devign datasets across four large language models (LLMs) (i.e., DeepSeek-Coder-6.7B, Qwen2.5-Coder-7B, StarCoder2-7B, and CodeLlama-7B) show that MultiVul achieves up to 27.07% F1 improvement over prompting-based methods and 13.37% over code-only Fine-Tuning, while maintaining comparable inference efficiency.

Organizations routinely make strategic budget allocations under operational constraints, but often lack a principled way to assess whether realized allocations were close to the best feasible choices in hindsight. We present a retrospective auditing framework based on hindsight regret, defined as the opportunity cost of the realized allocation relative to a constraint-faithful benchmark under the same budget and stability guardrails. The framework estimates regime-specific spend--response functions from historical logs, computes feasible hindsight allocations via constrained optimization, and propagates uncertainty through Monte Carlo evaluation to produce regret distributions, expected lift, and probability-of-improvement summaries. This separates allocation inefficiency from uncertainty in the estimated response surfaces. Experiments on real marketing allocation logs show that the framework yields interpretable post-hoc diagnostics and reveals a practical trade-off between allocation flexibility and detectability: moderate feasible reallocations often capture most measurable gain, while larger shifts move into weak-support regions with higher uncertainty. The result is a practical method for auditing historical budget decisions when online experimentation is costly or infeasible.

In-memory computing (IMC) enables energy-efficient neural network inference by computing analog matrix-vector multiplications (MVM) in memory crossbar arrays. In this work we present a simulation framework for N-ary crossbar architectures that retrieves MVM results with minimal implementation assumptions. The XOR and MNIST classification tasks were successfully inferred using a simulated crossbar array of (4x4) 4-states magnetic tunnel junctions (MTJ). MNIST accuracy reached 94.48% (vs. 97.56% software baseline). The software-hardware performance gap was further reduced using PCA dimensionality reduction. We identified weight quantization as the primary error source, and studied its impact alongside systematic nonidealities and random noise. We find that cell-specific random noise is less detrimental than systematic errors due to averaging across the array. Finally, we demonstrate an optimal number of states per cell that balances quantization error against resistance state resolution to minimize total MVM error.

In response to a concern raised by Horty, this paper develops a two-tiered, preference-based semantic framework for modeling defeasible conditional obligations. The paper extends a Hansson-Lewis style preference semantics for dyadic deontic logic by incorporating a nonmonotonic reasoning mechanism that enables previously derived obligations to be withdrawn when new, potentially conflicting information comes in. The account is bi-preferential: two orderings--ideality and normality--on worlds are employed to address shortcomings in earlier approaches, with a separate ranking method for each. At the nonmonotonic layer, a number of postulates are considered, including antecedent strengthening, inclusion and no-drowning. A connection is established with so-called constrained input/output (I/O) logic--an existing standard for normative reasoning based on a different methodology.

We study the problem of fitting a description logic (DL) ontology to a given set of positive and negative examples that take the form of an ABox and a Boolean query. While previous work has investigated this problem for the expressive DLs ALC and ALCI, we here focus on the Horn DLs EL and ELI, as well as their extensions with the bottom concept. As the query language, we consider atomic queries (AQs), conjunctive queries (rooted CQs), and unions thereof (rooted UCQs). We provide characterization of the existence of a fitting ontology based on simulations, use them to develop decision procedures, and clarify the exact computational complexity. For AQs, the problem is in PTime for both EL and ELI. For rooted CQs and UCQ, it is Sigma_P^2-complete for EL and ExpTime-complete for ELI. Adding the bottom concept does not change any of these complexities. Interestingly, moving from ALC and ALCI to EL and ELI introduces additional technical challenges rather than simplifying the matter.

Speculative decoding enhances the inference efficiency of large language models (LLMs) by generating drafts using a small draft language model (DLM) and verifying them in batches with a large target language model (TLM). However, adaptive drafting inference on a mobile single-NPU-PIM system faces idle overhead in traditional operator-level synchronous execution and wasted computation in asynchronous execution due to fluctuations in draft length. This paper introduces AHASD, a task-level asynchronous mobile NPU-PIM heterogeneous architecture for speculative decoding. Notably, AHASD achieves parallel drafting on the PIM and verification on a single NPU through task-level DLM-TLM decoupling and specifically, it incorporates Entropy-History-Aware Drafting Control and Time-Aware Pre-Verification Control to dynamically manage adaptive drafting algorithm execution and pre-verification timing, suppressing invalid drafting based on low-confidence drafts. Additionally, AHASD integrates Attention Algorithm Units and Gated Task Scheduling Units within LPDDR5-PIM to enable attention link localization and sub-microsecond task switching on the PIM side. Experimental results for different LLMs and adaptive drafting algorithms show that AHASD achieves up to 4.2$\times$ in throughput and 5.6$\times$ in energy efficiency improvements over a GPU-only baseline, and 1.5$\times$ in throughput and 1.24$\times$ in energy efficiency gains over the state-of-the-art GPU+PIM baseline, with hardware overhead below 3% of the DRAM area.

Retrieval-Augmented Generation (RAG) models frequently produce answers grounded in parametric memory rather than the retrieved context, undermining the core promise of retrieval augmentation. A fundamental obstacle to fixing this unfaithfulness is the lack of training data that explicitly requires models to prefer context over internal knowledge. We introduce Faithfulness-QA, a large-scale dataset of 99,094 samples constructed through counterfactual entity substitution. Starting from two established extractive QA benchmarks--SQuAD and TriviaQA--we automatically identify answer-bearing named entities in each context, replace them with type-consistent alternatives drawn from a curated bank of 76,953 entities, and thereby manufacture controlled knowledge conflicts between context and parametric memory. Rigorous quality filtering ensures 100% pass rates across four automated checks on random 200-sample audits. We release the full dataset, the construction pipeline, and a typed entity bank covering eight named entity categories. Faithfulness-QA is designed as a training resource for attention-based faithfulness objectives and as an evaluation benchmark for measuring context-grounding behavior in RAG systems. Data and code are available at https://github.com/qzhangFDU/faithfulness-qa-dataset.

Overparameterization is central to the success of deep learning, yet the mechanisms by which it improves optimization remain incompletely understood. We analyze weight-space symmetries in neural networks and show that overparameterization introduces additional symmetries that benefit optimization in two distinct ways. First, we prove that these symmetries act as a form of diagonal preconditioning on the Hessian, enabling the existence of better-conditioned minima within each equivalence class of functionally identical solutions. Second, we show that overparameterization increases the probability mass of global minima near typical initializations, making these favorable solutions more reachable. Teacher-student network experiments validate our theoretical predictions: as width increases, the Hessian trace decreases, condition numbers improve, and convergence accelerates. Our analysis provides a unified framework for understanding overparameterization and width growth as a geometric transformation of the loss landscape.

This paper presents C8s, a confidential computing architecture for Kubernetes that provides cryptographically rooted confidentiality, integrity, and verifiability guarantees for Kubernetes clusters from infrastructure operators. These guarantees are cryptographically provable to any independent third party verifier. The architecture is built on hardware Trusted Execution Environments (TEEs), specifically AMD SEV-SNP, Intel TDX, and NVIDIA Confidential Computing support, to establish an attestation-rooted trust boundary around confidential VMs. This design is compatible with managed Kubernetes services such as Amazon EKS, Google GKE, and Microsoft AKS, where the control plane cannot be attested. Under this boundary, three groups gain guarantees that are absent from conventional deployments. Data and artifact owners can deploy sensitive workloads and proprietary artifacts on third-party infrastructure without risking exfiltration. Compute providers can offer execution services without revealing workloads to cloud operators. End users can submit requests that remain opaque to all parties except the attested TEE processing them. Representative workloads include AI inference, securing AI model weights, and training or fine-tuning on sensitive data.

Scientific publication compresses a branching, iterative research process into a linear narrative, discarding the majority of what was discovered along the way. This compilation imposes two structural costs: a Storytelling Tax, where failed experiments, rejected hypotheses, and the branching exploration process are discarded to fit a linear narrative; and an Engineering Tax, where the gap between reviewer-sufficient prose and agent-sufficient specification leaves critical implementation details unwritten. Tolerable for human readers, these costs become critical when AI agents must understand, reproduce, and extend published work. We introduce the Agent-Native Research Artifact (ARA), a protocol that replaces the narrative paper with a machine-executable research package structured around four layers: scientific logic, executable code with full specifications, an exploration graph that preserves the failures compilation discards, and evidence grounding every claim in raw outputs. Three mechanisms support the ecosystem: a Live Research Manager that captures decisions and dead ends during ordinary development; an ARA Compiler that translates legacy PDFs and repos into ARAs; and an ARA-native review system that automates objective checks so human reviewers can focus on significance, novelty, and taste. On PaperBench and RE-Bench, ARA raises question-answering accuracy from 72.4% to 93.7% and reproduction success from 57.4% to 64.4%. On RE-Bench's five open-ended extension tasks, preserved failure traces in ARA accelerate progress, but can also constrain a capable agent from stepping outside the prior-run box depending on the agent's capabilities.

Block-sequential continual learning demands that a single model both protect prior solutions from catastrophic forgetting and efficiently infer at inference time which prior solution matches the current input without task labels. We present Functional Task Networks (FTN), a parameter-isolation method inspired by structural and dynamical motifs found in the mammalian neocortex. Similar to mixture-of-experts, this method uses a high dimensional, self-organizing binary mask over a large population of small but deep networks, inspired by dendritic models of pyramidal neurons. The mask is produced by a three-stage procedure: (1) gradient descent on a continuous mask identifies task-relevant neurons, (2) a smoothing kernel biases the result toward spatial contiguity, (3) and k-winner-take-all binarizes the resulting group at a fixed capacity budget. Like mixture-of-experts, each neuron is an independent deep network, so disjoint masks give exactly disjoint gradient updates, providing structural guarantees against catastrophic forgetting. This three-stage procedure recovers the sub-network of a previously-trained task in a single gradient step, providing unsupervised task segmentation at inference time. We test it on three continual-learning benchmarks: (1) a synthetic multi-task classification/regression generator, (2) MNIST with shuffled class labels (pure concept shift), and (3) Permuted MNIST (domain shift). On all three, FTN with fine grained smoothing (FTN-Slow) results in nearly zero forgetting. FTN with a large kernel and only 2 iterations of smoothing (FTN-Fast) trades off some retention for increased speed. We show that the spatial organization mechanism reduces the effective mask search from the combinatorial top-k subset problem in O(C(H,K)) to the complexity of a near-linear scan in O(H) over compact cortical neighborhoods, which is parallelized by the gradient-based update.

Multimodal Retrieval-Augmented Generation (MRAG) addresses key limitations of Multimodal Large Language Models (MLLMs), such as hallucination and outdated knowledge. However, current MRAG systems struggle to distinguish whether retrieved multimodal data truly supports the semantic core of an answer or merely provides superficial relevance. Existing metrics often rely on heuristic position-based confidence, which fails to capture the informational density of multimodal entities. To address this, we propose Multi-modal Evidence Grounding (MEG), a semantic-aware metric that quantifies the contribution of retrieved evidence. Unlike standard confidence measures, MEG utilizes Semantic Certainty Anchoring, focusing on high-IDF information-bearing tokens that better capture the semantic core of the answer. Building on MEG, we introduce MEG-RAG, a framework that trains a multimodal reranker to align retrieved evidence with the semantic anchors of the ground truth. By prioritizing high-value content based on semantic grounding rather than token probability distributions, MEG-RAG improves the accuracy and multimodal consistency of generated outputs. Extensive experiments on the M$^2$RAG benchmark show that MEG-RAG consistently outperforms strong baselines and demonstrates robust generalization across different teacher models.

Modern machine learning optimizes for accuracy without explicit treatment of internal computational cost, even though physical and biological systems operate under intrinsic energy constraints. We evaluate energy-aware learning across 2,203 experiments spanning vision, text, neuromorphic, and physiological datasets with 10 seeds per configuration and factorial statistical analysis. Three findings emerge. First, architecture alone explains negligible variance in accuracy (partial eta^2 = 0.001), while the architecture x dataset interaction is large (partial eta^2 = 0.44, p < 0.001), demonstrating that optimal architecture depends critically on task modality and rejecting the assumption of a universal best architecture. Second, a controlled lambda-sweep across lambda in {0, 1e-5, 1e-4, 1e-3, 1e-2} validates a single-parameter energy-regularized objective L = L_CE + lambda * E(theta, x): across this range, internal activation energy decreases by approximately three orders of magnitude relative to the unregularized lambda=0 baseline, with negligible accuracy change (<0.5 percentage points) on both MNIST and Fashion-MNIST. Third, energy-first architectures inspired by an action-principle framework yield 5-33% within-modality training-efficiency gains over conventional baselines. These results emerge from a research program that interprets learning through a structural correspondence between the action functional in classical mechanics, free energy in statistical physics, and KL-regularized objectives in variational inference. We frame this correspondence as a design hypothesis, not a derivation.

LLM agents increasingly rely on reusable skills, capability packages that combine instructions, control flow, constraints, and tool calls. In most current agent systems, however, skills are still represented by text-heavy artifacts, including SKILL{.}md-style documents and structured records whose machine-usable evidence remains embedded largely in natural-language descriptions. This poses a challenge for skill-centered agent systems: managing skill collections and using skills to support agent both require reasoning over invocation interfaces, execution structure, and concrete side effects that are often entangled in a single textual surface. An explicit representation of skill knowledge may therefore help make these artifacts easier for machines to acquire and leverage. Drawing on Memory Organization Packets, Script Theory, and Conceptual Dependency from Schank and Abelson's classical work on linguistic knowledge representation, we introduce what is, to our knowledge, the first structured representation for agent skill artifacts that disentangles skill-level scheduling signals, scene-level execution structure, and logic-level action and resource-use evidence: the Scheduling-Structural-Logical (SSL) representation. We instantiate SSL with an LLM-based normalizer and evaluate it on a corpus of skills in two tasks, Skill Discovery and Risk Assessment, and superiorly outperform the text-only baselines: in Skill Discovery, SSL improves MRR from 0.573 to 0.707; in Risk Assessment, it improves macro F1 from 0.744 to 0.787. These findings reveal that explicit, source-grounded structure makes agent skills easier to search and review. They also suggest that SSL is best understood as a practical step toward more inspectable, reusable, and operationally actionable skill representations for agent systems, rather than as a finished standard or an end-to-end mechanism for managing and using skills.

We explore a central question in AI for mathematics: can AI systems produce original, nontrivial proofs for open research problems? Despite strong benchmark performance, producing genuinely novel proofs remains an outstanding challenge for LLMs. Through systematic experiments with frontier LLMs on research-level proof tasks, we identify seven failure modes that prevent reliable proof generation, including context contamination, citation hallucination, hand-waving on key steps and misallocation of proof effort, unstable proof plans, unfocused verification, problem modification and single-model bottleneck. We argue that the gap between benchmark success and research-level proving is primarily one of system design, due to those failure modes. We present QED, an open-source multi-agent proof system in which each architectural decision directly addresses a specific failure mode. Evaluated on five open problems in applied analysis and PDEs contributed by domain experts, QED produces correct proofs for three problems, each verified by the contributing experts as original and nontrivial. QED is released as open-source software at https://github.com/proofQED/QED.

The solution of sparse linear systems constitutes the dominant computational bottleneck in interior point methods (IPMs), frequently consuming over 70% of the total solution time. As optimization problems scale to millions of variables, direct solvers encounter prohibitive fill-in, excessive memory consumption, and limited parallel scalability. We present SDSL-Solver, a scalable distributed sparse linear solver framework designed for IPMs. SDSL-Solver employs Krylov subspace methods, combined with numerics-based sparse filtering and diagonal correction techniques that produce high-quality preconditioners. To accommodate diverse problem characteristics, SDSL-Solver offers two complementary distributed parallel methods: Block Jacobi for diagonally dominant matrices, and Bordered Block Diagonal (BBD) for general or ill-conditioned matrices requiring globally coupled preconditioning via Schur complement techniques. A preconditioner reuse strategy further amortizes construction costs across consecutive IPMs iterations. We evaluate SDSL-Solver on benchmark problems with matrix dimensions ranging from tens of thousands to over five million on multi-node clusters equipped with X86 processors. The experimental results show that under the Block Jacobi and BBD distributed methods, SDSL-Solver on a four-node configuration achieves average speedups of 6.23 times and 7.77 times, respectively, compared to PETSc running on the same number of nodes. Relative to the single-node PARDISO, the average speedups reach 97.54 times and 5.85 times, respectively.

We propose a new regularized optimal transport (OT) formulation, termed sliced-regularized optimal transport (SROT). Unlike entropic OT (EOT), which regularizes the transport plan toward an independent coupling, SROT regularizes it toward a smoothened sliced OT (SOT) plan. To the best of our knowledge, SROT is the first approach to leverage a version of SOT plan as a reference to improve classical OT. We provide a formal definition of SROT, derive its dual formulation, and provide a post-Bayesian interpretation of SROT. We then develop a Sinkhorn-style algorithm for efficient computation, retaining the same scalability advantages as EOT. By incorporating a scalable SOT plan as a prior, SROT yields more accurate approximations of the exact OT plan than EOT under the same level of regularization. Moreover, the resulting transport plan improves upon the reference SOT plan itself. We further introduce the corresponding OT divergence induced by SROT, named SROT divergence, and analyze its topological and computational properties. Finally, we validate our approach through experiments on synthetic datasets and color transfer tasks, demonstrating that SROT is better than both EOT and SOT in approximating exact OT. Additional experiments on gradient flows further highlight the advantages of SROT divergence.

Foundation models of brain activity promise a new frontier for in silico neuroscience by emulating neural responses to complex stimuli across tasks and modalities. A natural next step is to ask whether these models can also be used in reverse. Can we recover a stimulus or its properties from synthetic brain activity? We study this question in a proof-of-concept setting using TRIBEv2. We pair the brain emulator with large language models (LLMs) that generate news headlines from linguistic parameters such as valence, arousal, and dominance. We then use simulation-based inference to learn a probabilistic mapping from brain maps to latent stimulus parameters. Our results show that these parameters can be recovered from predicted brain maps, validating the quality of neural encodings. They also show that LLMs can serve as controllable stimulus generators for simulated experiments. Together, these findings provide a step toward decoding and inverse design with foundation brain models.

Multiobjective optimization remains challenging for many scientific and engineering problems due to the need to balance convergence, diversity, and computational efficiency across high-dimensional objective landscapes. This work presents the Multiobjective Animorphic Ensemble Optimization (MAEO) framework, a parallelizable ensemble strategy that unifies state-of-the-art evolutionary algorithms within an island-based architecture, overcoming the limitations of relying on a single optimizer, as implied by the No Free Lunch theorem. MAEO uses a parameter-free hypervolume indicator for island performance assessment and a strict Pareto-rank-based individual scoring formulation that incorporates crowding distance and nadir-point proximity to ensure consistent selection pressure within each front. The framework is initiated using four algorithms (NSGA-III, CTAEA, AGEMOEA2, SPEA2) and evaluated through extensive benchmarking on 12 DTLZ/ZDT functions under 36 dimensionality settings using Wilcoxon signed-rank tests with both hypervolume and inverse generational distance metrics. Results show that MAEO achieves balanced convergence-diversity performance, outperforming or matching some of the leading multiobjective optimization algorithms across different benchmark problems. To demonstrate practical applicability, MAEO is applied to the equilibrium-cycle optimization of a small modular nuclear reactor. Eight discrete design variables (and three objectives (levelized cost of electricity, peak soluble boron concentration, fuel cycle length) are optimized under two safety constraints. The algorithm carried out roughly 40000 evaluations using computer simulations. MAEO identifies core designs that lower both the levelized cost of electricity and the peak boron concentration, while preserving fuel cycle length and meeting all safety constraints.

Relaxation of far-from-equilibrium quantum fluids, intimately related to the emergence of long-range order, is theoretically associated with the decay of a turbulent isotropic tangle of vortex lines. We observe and study such decaying quantum turbulence in a homogeneous 3D atomic Bose gas. Using matter-wave techniques to magnify the gas density distribution, and then imaging a thin slice of the magnified cloud, we observe imprints of randomly oriented vortex lines and measure the vortex line-length density $\mathcal{L}$. The observed decay of $\mathcal{L}$ agrees with the prediction for Vinen `ultraquantum' turbulence. Although our weakly interacting gases are highly compressible, their large-scale dynamics are consistent with the behavior of an incompressible hydrodynamic fluid, with the decay of $\mathcal{L}$ not depending on the strength of the interatomic interactions and being similar to that in the strongly interacting superfluid helium.

Multiferroic domain walls in functional oxides exhibit properties distinct from the bulk and are increasingly exploited as active elements in nanoelectronic and photonic devices. Deterministic control of domain populations has typically remained limited to local control, or removal with temperature. Here we demonstrate continuous, reversible manipulation of the ferroelastic domain structure in single-crystal LaAlO$_3$ using in-situ uniaxial strain. Combining atomic force microscopy, X-ray diffraction, and Raman spectroscopy with first-principles calculations we map the complete microscopic evolution of the twin domain population through the strain-driven transition from the rhombohedral $R\bar{3}c$ ground state toward the predicted orthorhombic $Fmmm$ phase. Applied strains below $0.5\%$ produce pronounced surface flattening and large-scale domain reorganisation, establishing uniaxial strain as a technically accessible control parameter for ferroelastic domain engineering. These results open a route to active, real-time programming of domain architectures in LaAlO$_3$-based heterostructures, with implications for strain-tunable superconducting interfaces, nanoscale phonon-polariton optics, and ultrafast lattice control.

In realistic nanoscale transport set-ups, electron-phonon coupling leads to the exchange of heat between phonon baths and electronic reservoirs with finite heat capacities. Such exchange affects the finite reservoir's temperature. However, this sensitivity of the finite reservoir temperature to the exchange of heat with the finite reservoir has remained unexplored for thermometry. Here, we fill this gap by combining current metrology techniques with a thermodynamic framework encompassing finite reservoirs. We focus on an experimentally realizable set-up with a quantum dot coupled to a finite reservoir and consider two distinct current-based strategies in the long time limit, namely monitoring quanta exchanged between the quantum dot and finite reservoir and the measurement of the total current flowing from the quantum dot into an infinite reservoir. A third strategy involves measurements of the quantum dot occupation. For a large but finite reservoir, we show that the Fisher information for all three strategies captures the finite reservoir's contribution to sensitivity through common factors. We also demonstrate that monitoring quanta exchanged between the system and finite reservoir in the long time limit achieves optimal precision. Finally, we provide an optimization analysis that explores how maximal precision can be achieved within each of the current-based strategies by tuning the gate voltage.

We investigate temperature quenches across the reentrant phase transition of the antiferromagnetic Ising model in a magnetic field and show that the strong direct and inverse Mpemba effects arise when quenches terminate in the paramagnetic phase. These anomalous relaxation phenomena originate from the selective excitation of the slowest relaxation mode, which in the paramagnetic phase is purely staggered. Consequently, quenches starting from the paramagnetic phase have zero overlap with the slow mode and exhibit a strong (inverse) Mpemba effect. Quenches from the antiferromagnetic phase excite the staggered mode and display a slow-relaxation tail. By varying the lattice coordination number we show that the strong Mpemba effect disappears in the absence of reentrance. Our results provide the first demonstration of the strong (inverse) Mpemba effect in the antiferromagnetic Ising model based on the pair-approximation, and establish a link between anomalous relaxation and the equilibrium phase behavior.

The nonlinear Hall effect (NLHE) connects crystalline symmetry to quantum geometry, offering a probe of band topology beyond linear transport. While most studies have focused on the Berry curvature dipole in low-symmetry crystals, mechanisms that directly probe Berry monopoles in higher-symmetry chiral lattices remain unexplored. Here, we report the observations of the NLHE in the chiral Weyl semimetal CoSi, a platform where the Berry curvature dipole is symmetry-forbidden. By employing focused ion beam-fabricated crossbar devices, we detect a robust second-harmonic Hall voltage under zero magnetic field, hosting all key signatures of the NLHE. Theoretical analysis attributes the nonlinear Hall conductivity to skew scattering of self-rotating electron wave packets, whose chirality is dictated by the underlying band topology, a process reminiscent of the classical Magnus effect. Furthermore, the NLHE signal exhibits a temperature-dependent sign reversal, and a strong, linearly field-dependent modulation that grows with carrier mobility, directly reflecting the topological Weyl nodes distribution near the Fermi level. These findings establish CoSi as a platform for Berry monopole-driven nonlinear transport, demonstrating a skew-scattering route to topological nonlinear Hall responses that bypasses conventional symmetry constraints.

Deep strong light-matter coupling represents an extreme non-perturbative regime of quantum electrodynamics, in which the interaction strength exceeds the bare frequencies of the uncoupled systems. The ground state features strong quantum correlations between photons and matter excitations, and new cavity-driven phase transitions are expected to occur. Whether a superradiant quantum phase transition, marked by spontaneous dipole ordering and photon condensation, is possible has remained a long-standing and controversial question. Such phenomena have been proposed to arise in exotic electronic systems hosting Dirac and Kane fermions, owing to the formal absence of an $A^2$ term in their low-energy Hamiltonian. Here we exploit the ultralow effective mass of Kane fermions to realise Landau polaritons in a bulk mercury cadmium telluride layer coupled to a Fabry-Perot resonator. Using thermally tunable carrier density, we continuously tune the coupling from the weak to the deep-strong regime, achieving a record normalised coupling ratio exceeding 1.6 above room temperature. The measured polariton spectra are in excellent agreement with a rigorous, gauge-invariant microscopic theory. Despite the nonlinear Landau level structure of relativistic Kane fermions, we show that a diamagnetic $A^2$ term naturally emerges and precludes a superradiant phase transition. These results resolve the long-standing controversy surrounding cavity quantum electrodynamics of relativistic-like matter systems, extend deep-strong-coupling physics to Kane fermions, and open new opportunities for polaritonic semiconductor devices operating in extreme light-matter coupling regimes.

We report on a new type of rectifier which is in full contact equilibrium and thus, if down-sized to the nanoscale, shows no drift even if exposed to elevated temperatures and/or extreme waiting times. This is in contrast to existing diodes which rely on frozen doping profiles and are hence non-equilibrium devices. Our rectifiers are related to Schottky diodes but employ "dopants" whose mobilities are high enough to follow the electrical field quickly but low enough to not compete with the electrons in terms of conductivities. In order to realize such a device based on mixed conductors, we use nanosized TiO2 films on Ru as a substrate which can store Li at the interface according to a job-sharing mechanism (Li-ions on the TiO2 side, electrons on the Ru side). The excellent functionality of this nanoionic device is demonstrated (e.g., current on-off ratio can exceed 6-7 orders of magnitude) and the additional advantages stressed (such as ease of preparation and tuning the characteristics electrochemically).

We prove that a class of classical lattice models on $\mathbb{Z}^d$ ($d \geq 2$) with on-site space $\mathbb{N}_0$ exhibits long-range checkerboard order at sufficiently high temperature. The model has a nearest-neighbour interaction $f : \mathbb{N}_0 \times \mathbb{N}_0 \to [0,\infty)$ satisfying four structural conditions, subsuming the recently introduced power-law model of Han--Huang--Komargodski--Lucas--Popov (arXiv:2503.22789) as a special case. The ordering mechanism is purely entropic: the checkerboard configurations are not energy minimisers, but are selected by the partial trace over occupation numbers in the $β\to 0$ limit. The proof uses Pirogov--Sinai theory and the key input is a Peierls bound.

For three-dimensional (3D) magnetic objects with linear size $L$ exceeding a few exchange lengths, the micromagnetic state exhibits pronounced informational sparsity: low-dimensional, high-gradient regions (e.g., domain walls) coexist with near-uniformly magnetized volumetric domains. Because standard micromagnetic simulation methods discretize the magnetization on near-uniform 3D grids with linear cell size $a$, they cannot take advantage of this sparsity. The computational problem scales as $\sim L^3$ and $\sim (1/a)^3$. In this Letter, we establish that direct tensor-train (TT) representations overcome these poor scalings by exploiting the spatial sparsity optimally, while preserving accuracy in a controlled way. Focusing on representative flux-closure configurations in soft-magnetic rectangular prisms, in the near-micrometer regime, we demonstrate that the parameter count of TT-compressed micromagnetic data scales approximately as $L^{1.8}$ and $(1/a)^{1.2}$. Hence the relative advantage over dense discretizations rapidly grows with the problem size and refinement level. These first results provide a strong motivation for future developments of micromagnetic solvers in TT format which could transcend the limitations of traditional simulators, with far reaching potential impacts on fundamental research and technology development.

Griffiths phases are typically associated with quenched disorder, while frustration gives rise to multistability and spin-glass behavior. Whether extended criticality can arise in other contexts remains an open question. Here, we show that synergistic interactions provide a distinct route to non-conventional critical phenomena. By combining spreading mechanisms that reinforce activity through complementary pathways, we uncover a broad distribution of relaxation rates, leading to Griffiths-like slow dynamics and extended criticality. We demonstrate that this mechanism is robust across networks and emerges both in systems with explicit higher-order interactions and in purely pairwise systems with nonlinear dynamics.

The Baik-Ben Arous-Peche (BBP) transition sets fundamental limits for detecting low-rank structure in noisy high-dimensional data and underlies a wide range of spectral methods in many fields from physics to statistics and data sciences. In standard settings, this transition is continuous, implying that signal recovery emerges gradually above a sharp threshold. We show that BBP transitions can instead be discontinuous in very general settings and provide a full theory of this phenomenon. When the eigenvalue density vanishes faster than linearly at the spectral edge, the overlap between the leading eigenvector and the signal jumps discontinuously at the critical point. We study this mechanism in deformed Gaussian and reweighted Wishart ensembles. We analyze in detail the finite-size effects, which play a central and qualitatively new role in the discontinuous BBP transition. Unlike the continuous BBP transition, we establish the existence of an extended pre-critical region where informative eigenvectors emerge well before the asymptotic threshold. The main consequence-and difference from the continuous BBP transition-is that signal recovery can occur at significantly lower signal-to-noise ratio and it is accompanied by strong sample-to-sample variability. Our results show the relevance and the novelty of the discontinuous BBP transition, and highlight the practical implications for signal detection.

Living systems routinely consume energy to achieve motility, often using intricate biomolecular machinery. In this work, we show that active droplets can sustain indefinite self-propulsion of a spherical colloid in an otherwise homogeneous, isotropic, and autonomous environment. Our proposed minimal mechanism consists of phase-separating proteins, enzymes passivating them, and complementary enzymes anchored to the colloid surface that reactivate the proteins. This passivation-activation cycle gives rise to a symmetry breaking - nucleation and stabilization of a condensate near the colloid surface, which in turn exerts a repulsive force on the colloid. We numerically demonstrate that this mechanism can propel micron-sized colloids at speeds of up to a hundred microns per second. This propulsion mode is strongly resistant to Brownian fluctuations and external forces, suggesting that propulsion mechanisms based on biomolecular condensates may offer a complementary, motor-free route to biological transport.

Quantum decoherence-the loss of quantum coherence due to interactions with an environment-plays a central role in quantum transport, and controlling this ubiquitous yet inevitable phenomenon is essential for practical quantum technologies. Despite its importance, the microscopic mechanisms of decoherence in infinite-size topological insulators remain poorly understood. Here, we develop a comprehensive theory that quantitatively investigates how quantum decoherence shapes the quantum spin Hall effect in macroscopic topological insulators, and reveal that decoherence-induced corrections scale quadratically with impurity density. Besides, we uncover a previously unidentified mechanism of the extrinsic spin Hall effect: a second-order skew-scattering process intrinsically tied to quantum decoherence-fundamentally distinct from, yet substantially stronger than, the conventional third-order skew-scattering mechanism. Furthermore, we predict a new scaling law in which the decoherence-induced spin Hall conductivity scales quadratically with the longitudinal conductivity, providing a clear experimental signature of decoherence effects. Our results establish the essential role of decoherence in quantum transport of topological insulators and reveal that macroscopic topological insulators offer a promising platform for next-generation spintronic applications.

Salts are an integral part of the environment for living systems and, therefore, understanding their effects on proteins and other biomolecules is of fundamental interest. Small-angle X-ray scattering (SAXS) of protein solutions can provide valuable information on salt effects, but extracting this information has been a significant challenge. For example, SAXS data of bovine serum albumin (BSA) at various salt concentrations were fit to three different spherical models. Here we combined the newly developed FMAPIq approach with explicit-solvent all-atom molecular dynamics simulations to show that the complex effects of salt on the SAXS of BSA originate from the interplay of ions and hydration water, leading to a general picture of protein-ion-water interactions.

We investigate the behavior of discrete interface growth models belonging to the Edwards--Wilkinson (EW) and Kardar--Parisi--Zhang (KPZ) universality classes, when defined on a complete graph, a topology commonly used to probe the infinite-dimensional limit of statistical mechanical systems. Our aim is to assess to what extent discrete lattice models reproduce the behavior of their corresponding continuum equations in this highly connected setting. After assessing the trivial behavior shown by some well known cases (like random deposition with surface relaxation or the etching model), we focus on two paradigmatic models associated with the KPZ universality class, the Restricted Solid-on-Solid (RSOS) and Ballistic Deposition (BD) models, and assess non-trivial behavior through several observables including the roughness, height fluctuations, power spectra, and two-time autocorrelation functions. Still, despite similarities with continuum equations, important differences arise in the fluctuations and long-time dynamics. In both discrete models the rescaled height fluctuations display a pronounced left tail, indicating the presence of lagging nodes. While the RSOS model exhibits a saturation roughness that decreases with system size, similarly to the EW and KPZ equations, the BD model exhibits a saturation roughness that increases with system size and an additional ultrafast growth regime, placing it outside the KPZ universality class on a complete graph.

We measured unusual critical currents as functions of temperature in the zero field and as functions of a magnetic field perpendicular to the substrate surface at a given temperature close to the critical temperature in thin-film long quasi-one-dimensional superconducting aluminum two-width structures consisting of narrow and wide wires with different critical temperatures. It is found that the experimental critical switching current as a function of the field at a given temperature, determined by the appearance of a dc voltage on a short section of the structure, is nonlocal (dependent on electron transport in the area containing the junction line between the narrow and wide wires). When current flows through the narrow and wide wires of the structure, the switching currents, experimental and calculated within the framework of the Ginzburg-Landau theory, differ radically from each other. A nonzero switching current exists in high fields greater than the maximum critical magnetic field in a quasi-one-dimensional superconducting wire. In the aluminum two-width structures studied here, the unusual measured switching current challenges description by known theories.

Discontinuous shear thickening (DST) in dense suspensions leads to flow instabilities that limit processing in many systems. While high-power ultrasound has been reported to reduce the apparent viscosity of such materials, the origin of this effect remains unclear. Here, we investigate dense adhesive cornstarch suspensions, where shear thickening arises from fragile, load-bearing force networks embedded in heterogeneous density-wave structures. Using a rheo-ultrasound setup, we show that ultrasound does not directly reduce viscosity but instead shifts the shear-thickening transition toward higher shear rates. This is evidenced by the collapse of stress probability distributions onto master curves, revealing a continuous evolution toward more fluid-like states without a sharp threshold. We interpret these results through a separation of time scales, in which the suspension behaves as an effectively immobile porous medium subjected to high-frequency interstitial flows. Fluidization then arises from a combination of boundary slip, bulk destabilization of force networks by drag-force fluctuations, and localized acoustic streaming. Beyond these mechanisms, we propose that ultrasound modifies the stability of force networks by introducing fluctuating hydrodynamic forces at the pore scale. As a result, larger stresses or shear rates are required to sustain jammed states, leading to a continuous renormalization of the DST transition. These findings provide a consistent physical picture of acoustic fluidization in adhesive suspensions and establish ultrasound as a powerful tool to control discontinuous shear thickening in confined flows.

The mixed alkali or mixed mobile ion effect in glasses manifests itself by strong nonlinear variations of ionic transport properties upon mixing of different types of mobile ions. We develop a theory for this effect based on thermally activated hopping transport in disordered site energy landscapes that consistently incorporates the statistical-mechanical and kinetic aspects of a mobile ion mixture. This includes a consideration of the joint probability density of site energy states, generalized Fermi distributions for mean site occupations, and cross-terms in the current response described by nondiagonal Onsager coefficients. The theory shows that a mixed alkali effect can arise even when the two ion species share identical site energy distributions. It suffices that sites have distinct energies when occupied by ions of different type. Taking into account that a mismatch energy is needed for ions of one type to occupy sites adapted to the other type, the mixed alkali effect becomes stronger. Spatial correlations between site energies are needed for the mobility of the majority ion to decrease stronger than exponential upon replacement by the minority ion. The theory agrees well with kinetic Monte Carlo simulations. Application to mixed alkali phosphate glasses yields good agreement with measured conductivity activation energies.

We report the formation of large-scale steady-state whirlpools in a GaAs-based two-dimensional electron liquid and demonstrate them by straightforward transport measurements. A whirlpool forming inside a circular cavity adjoining a wide conducting channel appears as a negative four-terminal resistance over a broad range of temperatures and cavity sizes. The effect scales with the Gurzhi length, in quantitative accord with the hydrodynamic analogy. Obtained results firmly establish this analogy and probe the limits of its applicability.

The third law of thermodynamics forbids cooling a physical system to absolute zero in a finite number of operational steps. Although this unattainability principle has been quantified for specific state-to-state transitions, a universal, dynamics-independent bound for implementing a state-agnostic reset map remains elusive. In this work, we unveil the fundamental limits of physical map implementation by deriving a trade-off relation based on geometric complexity. By analyzing continuous paths of maps on a geometric manifold, we prove that the geometric complexity of any classical stochastic map or quantum channel is bounded from below by its execution error. As a consequence, we show that achieving zero error in a state-reset operation requires a divergent geometric complexity -- a unified measure that naturally incorporates disparate physical resources, including infinite time, energetic cost, or control bandwidth. This unattainability principle holds universally across both classical and quantum regimes, establishing a strict geometric limit on the physical realization of reset operations in thermodynamic control and quantum computation.

Negatively charged boron vacancy (VB-) defects in hexagonal boron nitride (hBN) are promising for nanoscale-proximity quantum sensing. To evaluate their performance, it is important to characterize the spin coherence times T2* and T2. In this study, we realized sub-GHz Rabi oscillations of VB- using an isotopically enriched hBN thin film directly stamped onto a narrow gold wire. Using these strong microwave pulses, we performed Ramsey interference and Hahn echo measurements. The Ramsey interference signal showed Gaussian-like decay, yielding T2* = 13.8 ns. The Hahn echo measurement gave T2 = 108.7 ns and a stretch factor of α= 1.25. These results experimentally clarify the spin coherence properties of VB- and provide an effective method for evaluating the coherence of spin defects in van der Waals thin films with broad resonance linewidths.

We measured the critical temperatures and critical switching and retrapping currents of wide and narrow thin-film quasi-one-dimensional superconducting aluminum structures of the same thickness in zero magnetic field. For the first time, we found that the narrower the structure, the lower the critical temperature and critical current density in the structure. Probably, the influence of depairing centers that are on dirty longitudinal boundaries of the structure, is the stronger than the narrower the structure. It is found for the first time that, in most cases, the temperature-dependent switching critical current in both structures is approximated by two functions. At temperatures below the temperature corresponding to the bottom of the resistive N-S transition of structures, the switching critical current is described by the Kupriyanov-Lukichev theory. At temperatures close to the top of the N-S transition, the switching current is linear with temperature and coincides with the critical Josephson current. At these temperatures, Josephson SNS junctions are formed in structures.

We model a direct solid-state phase transition through a nucleation-and-growth process in which plates have simple, regular shapes - squares, cubes, or square-faced lamellae - and grow homothetically (self-similarly) until they either reach a randomly assigned maximum size or are stopped by impingement with previously formed plates. The reverse transformation is represented by the preferential disappearance of smaller plates, while larger plates are retained during an incomplete reversion. A subsequent direct transformation therefore produces a modified plate-size distribution, a memory effect that forms the main focus of this study. Building upon an earlier two-dimensional (2D) formulation, we extend the model to cubes (3D) and to lamellar plates (3DL) in order to examine how dimensionality affects transformation memory. We introduce a quantitative descriptor of memory, the size mass ratio, and find that memory is robust in all geometries but overall stronger in 2D than in 3D or 3DL. We provide growth snapshots, arrest-regrowth cycles, size distributions, and differential scanning calorimetry simulations, and we compute the Shannon size-entropy to quantify configurational diversity. Although motivated by the thermal memory effect in shape-memory alloys, the model more generally identifies a purely geometric mechanism for memory in first-order solid-solid transformations, highlighting the role of dimensionality and geometric blocking in controlling the strength of transformation memory.

We study theoretically the spin-orbit interaction of electrons confined in a tubular semiconductor nanowire, between an inner semiconductor core and an outer metallic extra shell. A band off-offset potential is present at the inner semiconductor-semiconductor interface and a more complex potential barrier at the outer metal-semiconductor contact. The cross section of the nanowire has a hexagonal geometry. We use a model derived from the k-dot-p method, and discuss the effects of the interface potentials on the strength of the spin-orbit coupling and on the localization of the wave functions within the semiconductor shell

Toxic cyanobacterial blooms are a growing environmental concern that affects freshwater ecosystems, drinking water supplies, and public health. The cyanobacterium Microcystis is among the most important bloom forming species. It often grows in large colonies, which enhances its flotation, reduces grazing, and improves nutrient regulation. Microcystis cells are held together by a matrix of extracellular polymeric substances (EPS), making colony mechanics crucial for bloom formation. However, an analysis of the biomechanical properties of cyanobacterial colonies, and how these properties relate to environmental conditions like nutrient availability, remains largely missing. Here, we use micropipette force sensors to quantify the linear and non-linear mechanical properties of individual colonies at single-cell resolution. Bulk shear rheology complements these measurements by probing macroscopic properties. The measured tensile strength and yield stress are broadly comparable to those of bacterial biofilms and are far greater than the hydrodynamic stresses typically found in wind-mixed lakes. This implies that cyanobacterial colonies are highly resistant to fragmentation by natural mixing processes. We also show that low nutrient availability, particularly low phosphorus, produced stronger colonies, suggesting structural changes in the EPS. Overall, our results establish mechanical testing as a tool for a more complete and physically grounded understanding of cyanobacterial colony formation.

Using the path-integral formalism, we show that photons possess a nontrivial quantum metric in momentum space. We derive the semiclassical action and equations of motion by taking into account the quantum metric. In media with a spatially varying refractive index $n(\mathbf{x})$, the quantum metric induces a shift in the trajectory of light at second order in derivatives of $n$, which may be regarded as a nonlinear Hall effect of light. The quantum metric also gives rise to corrections to gravitational lensing in curved spacetime at the nonlinear order in wavelength. This gravitational nonlinear Hall effect results from the interplay between the geometry of position space and that of momentum space.

Isolated quantum systems typically approach thermal equilibrium as described by the Eigenstate Thermalization Hypothesis (ETH). Going beyond this involves either higher order correlators (full thermalization) or the formation of state designs, i.e., the approach of moments of state ensembles after a projective measurement towards thermal equilibrium (deep thermalization). We compare these two extensions of ETH using extensive numerical studies within a paradigmatic model for chaotic many-body quantum dynamics. For this we find exponential relaxation for both extensions: For deep thermalization all moments relax with the same rate, which approximately equals the relaxation rate of the autocorrelation function captured by ETH. In contrast, higher order correlation functions in full thermalization approach equilibrium faster. This means that at higher orders full thermalization is faster than deep thermalization.

We have developed a novel BEDT-TTF-based organic conductor, $κ$-(BEDT-TTF)$_2 m$-HOOCC$_6$H$_4$SO$_3$ ($κ$-$m$-SBA), and propose it as a candidate for altermagnet. Tight-binding band calculations of $κ$-$m$-SBA provide a $t'/t$ of 1.01 at 100 K, indicating that the spin structure is closely aligned to an equilateral triangle ($t'/t= 1$). While most $κ$-type BEDT-TTF-based salts become spin liquids due to the spin frustration caused by the triangular lattice, $κ$-$m$-SBA surprisingly shows a weak ferromagnetic transition at $T_\mathrm{N} = 14$ K due to a canted antiferromagnetic (AFM) spin structure. Until recently, $κ$-(BEDT-TTF)$_2$Cu[N(CN)$_2$]Cl ($κ$-Cl) was the only $κ$-type organic conductor known to exhibit this order, and it is also recognized as the first candidate for altermagnetism in organic conductors. This was theoretically predicted by Naka et al. in 2019, who demonstrated that $κ$-type organic conductors can be candidates for altermagnetism if they display such order. Consequently, $κ$-$m$-SBA can be considered the second candidate for altermagnetism in organic conductors. Furthermore, numerical calculations demonstrate a characteristic of altermagnets in $κ$-$m$-SBA, namely spin splitting of energy bands.

Elastic wave propagation is intrinsically sensitive to the mechanical properties of the medium through which it travels. In soft elastomers, this makes guided elastic waves natural probes of viscoelastic and acoustoelastic behavior over a broad frequency range. In this work, we introduce a wave-based mechanical characterization method in which a thin elastomer strip acts as a waveguide supporting multiple in-plane guided modes. By combining stroboscopic measurements of monochromatic wave fields with a theoretical framework that couples frequency-dependent viscoelasticity and elongation-dependent acoustoelasticity, we extract complex-valued dispersion relations for guided modes under controlled static elongation. A dedicated numerical implementation allows these experimental dispersion curves to be quantitatively matched to theory, enabling identification of the material's rheological and hyperelastic parameters. Applied to several commercial silicone elastomers, the method yields mechanical parameters that are consistent with conventional plate-plate rheometry, while extending the accessible frequency range beyond that of conventional techniques. By exploiting the richness of guided-wave dispersion and the sensitivity of waves to both frequency and pre-stress, this approach provides a unified, broadband, and experimentally simple route to the mechanical characterization of soft elastomers.

We develop a coarse-grained theoretical description of the macroscopic emergent electric field generated by phonon-coupled lattice deformations in the breathing and rotational dynamics of a skyrmion lattice under microwave excitation. The analysis identifies the symmetry and dynamical conditions that yield rectified (dc) and oscillating (ac) electric fields, even in the absence of net translational motion of the skyrmion lattice, particularly in the dilute-lattice limit. Using experimentally measurable skyrmion profile parameters such as the equilibrium radius, domain-wall width, and dynamical resonance frequency of skyrmion lattice, the model further enables identification of harmonic components contributing to the observed macroscopic electrodynamic response in the long-wavelength phonon limit ($q \to 0$) and at finite phonon frequency, providing a unified framework for phonon-driven spin-charge-lattice coupling in topological magnets.

Room-temperature single-photon emission (SPE) resulting from a biexciton-exciton cascaded decay is demonstrated for the first time from chemically and photoelectrochemically etched site-controlled In0.14Ga0.86N quantum dots (QDs) embedded in vertical GaN nanowires. Diameter-dependent biexciton-exciton dynamics are analysed to determine the eligibility of QD as a single-photon emitter. The signal-to-noise ratio degrades with increasing QD diameter. Background noise photons pose a bottleneck to achieving SPE. This is also explained from a carrier dynamics perspective. Surface recombination contributes to inhomogeneous broadening at QD diameters larger than 35 nm. Below 35 nm, density-of-states-corrected Auger gradually becomes the principal biexciton-decay route with further reduction in QD diameter, thereby quenching the possibility of thermal broadening and setting a threshold for SPE. Below 9 nm, the Auger recombination rate becomes manyfold of other decay rates, causing multi-photon suppression via single Auger decay to form an exciton. Surface recombination probability of this exciton is minimized while biexciton state filling probability is maximized by reducing sidewall surface states through wet-treatment. These improve biexciton state preparation and enhance the single-photon purity of the exciton towards the exciton Bohr radius (3 nm) regime. Far away from this regime, higher-order autocorrelations to characterize quantum emission involving multi-photon events are discussed. This study establishes a generalized physical framework for predetermining SPE probability as a function of QD surface and geometry down to the exciton Bohr radius regime, with practical implementations. This work shows the pathway to design and develop next-generation semiconductor QDs for high-purity room-temperature SPE.

Magnon-polaritons provide a room-temperature platform for investigating nonlinear cavity quantum electrodynamics in the microwave domain, but experimentally observing controlled transitions among distinct nonlinear attractors remains challenging in conventional passive systems, where strong external driving is usually required. Here we report the observation of attractor transitions in an active magnon-polariton formed by a self-oscillating microwave cavity coupled to a yttrium iron garne (YIG) sphere. The feedback loop supplies an internal microwave drive, while Kerr frequency pulling and Suhl-mediated magnon-magnon scattering produce an enhanced effective nonlinearity. Stability analysis using experimentally calibrated parameters reveals a rich fixed-point (FP) landscape with multiple unstable-FP phases and a triple-point region. By tuning gain across these phases, we observe the first experimental evidence of explosive growth of bistability, followed by transitions to multifrequency limit cycles, comb-like/fractal spectra, and broadband chaotic dynamics at microwatt powers. Near a critical point, magnetic-field-triggered switching between nonlinear emission states produces spectral shifts up to 162 times the bare gyromagnetic response. By enabling low-power attractor transitions and attractor-switching-amplified spectral response, active magnon-polaritons open opportunities for nonlinear microwave signal generation, high-precision sensing, and neuromorphic computing.

We propose a noise-mitigation quantum simulation strategy for near-term quantum devices based on Quantum Circuit Learning (QCL), which is in particular effective for integrable quantum spin chains. The method trains a shallow variational circuit to approximate a deeper time-evolution circuit by learning the conserved charges and only a small amount of dynamical information in the system. Under realistic noise models, the learned circuit maintains both conserved quantities and dynamical observables significantly closer to their true values than the noisy simulation of the original circuit. This demonstrates QCL as an effective, physics-informed error mitigation strategy, producing shorter, more robust circuits without exponential sampling overhead.

We theoretically demonstrate that quantum fluctuations inherent to antiferromagnets can generate scalar spin chirality at zero temperature even in coplanar ordered magnets. In a kagome antiferromagnet with coplanar ground-state spin configurations, the quantum-fluctuation-induced scalar spin chirality is shown to arise at zero temperature when an effective time-reversal-like antiunitary symmetry is broken in the Hamiltonian describing fluctuations, and a magnetic point group of the classical ground state allows for its presence. The scalar spin chirality fluctuations are shown to grow further with increasing temperature by thermally excited magnons. These scalar spin chirality fluctuations can reach a magnitude comparable to the static one predicted for noncoplanar spin structures, highlighting their physical implications in coplanar spin systems.

Granular materials are inherently out-of-equilibrium systems due to energy dissipation through inelastic collisions and friction. When driven by mechanical agitation such as vibration, they exhibit rich collective behaviors including segregation, clustering, and spontaneous oscillations. Here, we report directional stepwise migration of particle clusters from one compartment to the next in a vertically vibrated granular system composed of small and large particles. To clarify the underlying mechanism, we directly measured how the flux of both particle species depends on the instantaneous particle populations. The measurements reveal an asymmetric interaction between particle species: the flux of small particles is enhanced by the presence of large particles, whereas that of large particles is suppressed by small particles. A minimal flux model incorporating these measured fluxes reproduces the observed directional dynamics and provides an experimentally grounded framework for collective transport in vibrated granular systems.

In non-Hermitian systems, spectra can be maximally boundary-sensitive, yet bulk physics need not be. Here we experimentally show that spectral moments provide boundary-robust bulk observables in finite non-Hermitian lattices, even when the spectra undergo dramatic geometry-dependent reshaping due to the skin effect. Using a unified acoustic platform with full spectral reconstruction and time-domain access, we probe one-, two- and three-dimensional lattices and demonstrate that spectral moments remain nearly invariant across distinct boundary geometries while the corresponding complex spectra differ strongly. To connect the thermodynamic theorem to realistic finite systems, we develop a loop-counting theory that identifies the physical origin of finite-size deviations in terms of missing boundary loops, quantitatively captures the corrections, and predicts a scaling law, which we verify experimentally. Beyond acoustic spectroscopy, we reveal a counterintuitive dynamical consequence of moment invariance: a dispersive-to-proliferative bulk transition governed by bulk moment structure rather than spectral boundary sensitivity. As a result, local bulk dynamics can remain stable (dispersive) even in a $\mathcal{PT}$-broken spectral regime, challenging the conventional expectation that $\mathcal{PT}$ breaking necessarily implies feedback-induced dynamical instability (proliferation) through exponentially amplifying spectral components. These results establish spectral moments as practical bulk descriptors for finite non-Hermitian matter and open a route to extracting and controlling intrinsic bulk behavior in realistic wave-based non-Hermitian devices.

Bosonic quantum error correction encodes a logical qubit in an oscillator, avoiding the hardware overhead of large qubit arrays. Among such encodings, Gottesman-Kitaev-Preskill (GKP) states are paticularly powerful because their phase-space grid structure protects against small displacement errors simultaneously in both conjugate quadratures. Here we provide the first protocol for preparing magnonic GKP states, which involves an ellipsoidal magnetic crystal effectively coupled to a superconducting qubit via a microwave cavity. The geometric anisotropy intrinsically squeezes the magnon mode, while the cavity-mediated qubit control realizes an effective conditional-displacement interaction. We show that two rounds of a conditional-displacement interaction and a qubit projective measurement yield three- and four-component magnonic GKP-like states. We also show how to realize single logical qubit gate operations, such as Pauli, Hadamard and phase gates, completing the logical Pauli basis of the approximate GKP code. Our results establish hybrid magnon-qubit systems as a promising platform for preparing bosonic code states, with applications in magnonic fault-tolerant quantum computation and quantum sensing.

{\it Jacob's ladder} -- a classic children's toy -- is a simple mechanical frame comprising rigid blocks connected by strings that shows curious unidirectional flipping waves. Nonetheless, its physical origin remains elusive. By combining experiment, numeral simulation, and theory, we show that understanding the underlying design principle of this toy requires diverse physical ideas. First, we conduct a water-tank experiment that excludes the domino-like mechanism, thus defying widespread expectations. Subsequently, we analytically demonstrate that the toy is bistable under gravity, thus implying its kink wave as a class of topological solitons. The waves are surprisingly reminiscent -- both experimentally and theoretically -- to those in the Kane--Lubensky topological chain, owing to the stiffening of zero modes by the pretension under gravity. However, a close examination based on the index theorem reveals that the similarity remains superficial and that the floppiness of the toy underlies the kink and antikink coexistence -- a forbidden mode in the topological chain. By analyzing a generalized asymmetric toy, we reveal that its symmetric connection renders it topologically singular, thus resulting in amusing motions. We demonstrate these ideas by experimentally observing a dramatic pair annihilation of kink and antikink waves.

Optoelectronic and nonlinear transport experiments probe the quantum geometric tensor of Bloch states, whose real and imaginary components -- the quantum metric and the Berry curvature -- are typically constrained by symmetry. Here, we show that geometric bending provides a route to engineer such responses in centrosymmetric ferromagnets. Curvature-induced strain gradients across the shell thickness break inversion symmetry and activate an orbital Rashba coupling. In the presence of in-plane magnetization and spin-orbit coupling, this generates spin textures with a nontrivial quantum geometry, leading to an intrinsic nonlinear anomalous Hall effect (NAHE) governed by the quantum metric and maximized when the magnetization aligns with the applied electric field. When geometric deformations further break twofold rotational symmetry around the out-of-plane axis, an additional NAHE emerges, maximal for magnetization perpendicular to the driving electric field and governed by the Berry curvature dipole, thus giving access to the imaginary component of the quantum geometric tensor. These results establish curved ferromagnetic shells as a platform for engineering anisotropic nonlinear transport and for selectively probing both components of the quantum geometric tensor.

Exceptional points (EPs), non-Hermitian degeneracies where both eigenvalues and eigenvectors coalesce, play a central role in the topology of non-Hermitian spectra. Recent advances have enabled the controlled creation and manipulation of EPs in a wide range of physical systems, raising the question of what new band topology emerges when EPs become mobile under cyclic modulation. Here we show that mobile EPs generate momentum-space switching domains that partition the Brillouin zone into regions with distinct band-switching behavior. Using a minimal two-band lattice model, we introduce a band-permutation invariant that determines whether eigenmodes exchange after one modulation cycle. The boundaries between switching regions arise from the projection of EP trajectories in an extended parameter space combining crystal momentum and the modulation parameter. As the modulation strength increases, the switching domains expand and eventually cover the entire Brillouin zone, resulting in global band switching. The predicted switching-domain structure is further demonstrated in a photonic crystal with lossy materials. These results open a new avenue within non-Hermitian topology by enabling the engineering of EP-driven phenomena through their controlled motion.

Viscoelasticity governs the locomotion strategies of deformable microorganisms, rendering it a fundamental mechanical property of microbial motility and an integral component in the design of envisioned microbots. Recent studies have shown that it can enable effective propulsion through non-reciprocal body deformations, even under time-reversible actuation. In this work, we investigate the dynamics of model microswimmers driven by reciprocal actuation, wherein the passive body exhibits viscoelastic deformability. We consider two linked-sphere designs, distinguished by the location of actuation: applied at one end (3-sphere design) or at the midpoint of the swimmer body (4-sphere design). Adopting Kelvin-Voigt deformability, we characterize the kinematic performance of both designs: the three-sphere swimmer possesses an optimal actuation frequency, while the four-sphere swimmer exhibits a critical frequency at which the locomotion direction reverses. We examine the swimmer's far-field hydrodynamic signature and find that resulting flow field is characterized by dominant dipolar and quadrupolar contributions, whose magnitudes are sensitive to the relative length of the actuator segment.

Recent experiments show that optically generated Rydberg excitons in cuprous oxide can neutralize charged impurities, strongly reducing stray electric fields and effectively purifying the crystal. Here, we develop a multichannel theory of Rydberg exciton-impurity scattering that resolves the competing roles of capture, elastic scattering, and inelastic transitions between excitonic states. We find that at high collision energies, as effective under conventional single-photon excitation, purification is reduced relative to Langevin capture. These collisions are accompanied by inelastic redistribution and dominant elastic scattering, including pronounced glory scattering, which suppress purification efficiency. We identify a quantum regime at ultralow collision energies favorable for purification, where only the s-wave contributes: capture is enhanced while elastic and inelastic channels are strongly suppressed. This regime can be accessed via degenerate two-photon excitation of even-parity Rydberg excitons with tunable recoil, additionally enabling the systematic exploration of exciton-impurity scattering over a wide range of collision energies beyond what is readily achievable in atomic counterparts in atomic gas experiments.

Quantum geometry encoded in the momentum space structure of electronic wavefunctions, governs charge dynamics through Berry curvature, enabling unconventional transport and optical responses. In topological semimetals, this geometry is sampled over Fermi pockets, suggesting electrical control by Fermi surface tuning, yet such control has remained largely limited to DC transport. Here we show that electrostatic gating of the 3D Dirac semimetal Cd3As2 reshapes Fermi pockets surrounding photoinduced Floquet Weyl nodes, enabling electrical control of terahertz (THz) emission chirality. Gate tuning selectively modulates the Berry curvature driven linearly polarized THz component by up to 60% and 49% at positive and negative bias, respectively, while the orthogonal linearly polarized photon-drag component remains unchanged. With the two orthogonal fields intrinsically phase-locked at \sfracπ{2} by the excitation geometry, the selective gate-tuned amplitude control enables the polarization tuning across the Poincaré sphere, achieving near-circular polarization (χ\approx-42°) at +10 V. These results establish Fermi surface tuning as a general route to programmable quantum geometric control of chiral terahertz emission.

We study prethermal time quasicrystalline (TQC) order in a quasiperiodically driven chain of non-interacting spin-1/2 particles. The drive consists of two parts, switched on and off periodically with frequency $ω_d$: (i) disordered Ising interactions, with exchange couplings chosen from a symmetric interval $[-J/2, J/2]$, allowing random antiferromagnetic or ferromagnetic nearest-neighbor couplings, together with a random transverse field; and (ii) a rotating transverse magnetic field with frequency $Ω$. The ratio $ω_d/Ω$ is chosen to be irrational, producing multiple incommensurate frequencies and yielding quasiperiodic dynamics beyond Floquet theory. Using exact diagonalization, we analyze the time autocorrelation function, dynamical structure factor, and entanglement entropy (EE). In the high-frequency regime, robust spectral peaks at incommensurate frequencies (not integer multiples of the fundamental drives) signal quasiperiodic time-translation symmetry breaking (QTTSB). The EE exhibits sublinear power-law growth followed by a prethermal plateau, indicating suppressed resonant heating due to an energy scale mismatch. The nonequilibrium lifetime increases rapidly with driving frequency. Unlike symmetric disorder sampling, an asymmetric distribution of the Ising exchange couplings induces collective spin rigidity, enhancing the system's resistance to heating. The TQC phase remains stable against next-nearest-neighbor (NNN) exchange perturbations and rotational imperfections, with robustness comparable to discrete time crystals (TCs) under periodic driving. Our results establish this quasiperiodically driven system as a platform for long-lived nonequilibrium temporal order, revealing the interplay of disorder, collective rigidity, and quasiperiodic driving.

Piezoionic hydrogels offer a route to mechanically driven bioelectronic interfaces, but their output is limited by rapid, symmetric ion redistribution that dissipates charge gradients. In biological electrocytes, efficient signal generation arises from the coupling of ion selectivity with spatial confinement that regulates transport. Here, we introduce a confinement-connectivity design strategy for piezoionic hydrogels, implemented through a supramolecular poly(vinyl alcohol)-glycerol-cucurbit[5]uril (PVA-glycerol-CB[5]) mesoporous network with a layered Negative-Neutral-Positive architecture that simultaneously increases pore fraction while reducing characteristic pore size. This architecture constrains ionic redistribution while maintaining a large mobile-ion reservoir, enabling deformation-driven charge separation. Compression generates peak outputs of ~180 mV and ~9 mA and elicits synchronized electromyographic responses in the mouse sciatic nerve without external power. These results establish confinement-connectivity coupling, rather than bulk conductivity, as a materials design framework in which coupling pore connectivity and confinement governs piezoionic transduction.

I study the impact of mixed contrarians on the opinion dynamics of an heterogenous population with conformists using Galam Majority Model. Activation of contrarians is a function of the ratio majority/minority in the local groups of discussion. Restricting the group size to 3, two types of contrarians are included in respective proportions $c_{3,0}$ for configurations with ratio 3 to 0 and $c_{2,1}$ for ratio 2 to 1. I then derive the explicit update Equation and obtained analytically the fixed points, their stability, and the resulting full two-dimensional landscape of the dynamics of opinion. Setting $c_{3,0} =c_{2,1} = c$ recovers the original results obtained with uniform contrarians. The findings allow for considering a wide spectrum of new disruptive strategies to secure either a majority/minority ending ensuring the opinion having the larger initial support to win, or a single attractor dynamics at fifty/fifty, which implies a random winner regardless of initial supports.

Multirate characterization of spin responses in nuclear magnetic resonance (NMR) is a promising approach to fingerprinting complex molecules in the presence of multiple relaxation mechanisms. Here we present experimental and theoretical investigations simultaneously accessing 8 relaxation rates describing the density matrix of two adjacent non-equivalent nuclear spins 1/2 ($^1$H and $^{\ 13}$C) belonging to a molecule in a liquid. The selected nuclear pair is stable with respect to chemical exchange. Some of the rates are obtained from conventional measurements of inversion recovery and nuclear Overhauser effect, while other, less conventional ones, are extracted from the relaxation initialized by the maximally entangled pseudo-pure Bell states (Bell PPSs) of the spin pair. The Bell PPSs are created using a hereby introduced method based on a detuned Hartmann-Hahn double resonance condition. Microscopic theory behind the measured relaxation rates is presented, and its consistency is demonstrated by several parameter-free tests. In particular, it is shown both theoretically and experimentally, that the eigenmodes of the off-diagonal relaxation of the two-spin density matrix can be selectively initialized using Bell PPSs. Our multirate analysis suggests that the measured off-diagonal relaxation is partly due to an unconventional mechanism arising from very weak $J$-couplings of the spin pair with fluctuating distant nuclear spins. Furthermore, we identify a dimensionless ratio of diagonal relaxation rates, which is determined exclusively by intra-pair magnetic dipolar interaction and hence possesses a universal value for a broad class of nuclear spin pairs. This value is consistent with both our experiments and other experiments reported in the literature.

Nanoscaled friction is a fundamental tribological phenomenon with complex behavior of its dynamical force. Here, we utilize the Prandtl-Tomlinson framework to investigate systematically the different means of control of the frictional force at the quantum and classical levels. It is found that the frictional dynamics can be controlled by the corrugation and characteristic length ratio parameters dependent upon properties of the nanoparticle-chain system. In addition to the stick-slip regime, other types of motion are uncovered, highlighting the richness of the frictional dynamics. The importance of Landau-Zener tunneling for the quantum motion is also analyzed. These findings provide valuable insights for interpreting experimental observations and controlling quantum frictional behavior in nanoscale systems.

Quantum integrable spin chains are known to possess a large family of long-range deformations generated by the local, boost and bilocal operators. Although these deformations are well-understood on the level of the pairwise commuting charges, the underlying quantum group structures had not yet been recognised. In this paper, we provide a quantum group-theoretical description for the family of long-range deformations of arbitrary homogeneous Yang-Baxter integrable spin chains up to first order in the deformation parameter. In particular, we show that the deformations are obtained via a twist of the algebraic structure of the underlying quantum group. This twisting results in a generally non-associative algebra that has a non-trivial Drinfeld associator. The Drinfeld associator is then shown to encode the information about the long-range interaction terms for the integrable spin chain. Moreover, the deformed quantum group is shown to contain a large perturbatively associative substructure, thus ensuring the perturbative integrability of the long-range model. The deformed quantum group provides explicit expressions for the Lax operators and R-matrices of the long-range deformed models, which manifestly satisfy the RLL relation and the Yang-Baxter equation up to first order in the deformation parameter. In order to derive the results, we introduce algebra elements that we call the algebraic charge densities. As a side result, we provide a conjecture for the explicit expressions of the undeformed charge densities in terms of these algebra elements.

Several recent methods have shown that it is possible to compute rate constants of very slow biomolecular processes using simulations where a time-dependent bias is added along one or several collective variables (CVs). We previously reported the exponential average time-dependent rate (EATR) method, which can improve upon these approaches by accounting for how efficiently the external biasing potential modifies the observed rate using a learned CV-quality factor $γ$. This results in more accurate rate estimates using the same data when biasing a suboptimal coordinate. However, as formulated EATR depended on the biasing potential varying over time to properly determine the biasing efficiency, which limits the method's applicability to quasi-static biasing schemes such as ``flooding'' or on-the-fly probability enhanced sampling (OPES). Here, we present the EATR-flooding approach, which generalizes our method by replacing the need for a time dependent bias by instead varying (stepping up) the strength of the biasing potential across multiple sets of simulations. We implement this approach as an open-source Python library, and demonstrate that this approach is accurate without substantial loss of efficiency compared to standard EATR for a coarse-grained protein system, and also show good performance on a fully atomistic cavity-ligand model. Two additional appealing features of EATR-flooding are an internal check for over-biasing and the fact that only a single $γ$ parameter is predicted for a given choice of CVs, as compared to our earlier results where $γ$ empirically depended on biasing rate. Finally, we believe EATR-flooding applies not only to OPES simulations but more generally to CV biasing enhanced sampling approaches, making it broadly useful.

We develop a framework to describe how quantum coherences between distinct charge sectors evolve under generic charge-conserving dynamics. Our framework captures the nonperturbative interactions between quantum coherences and hydrodynamic large deviations -- i.e., rare ``voids'' of low charge entropy. Conditional on surviving, the quantum coherence and its surrounding void form a collective polaron-like object. In one dimension, even at infinite temperature, we show that the lifetime of coherences is parametrically enhanced because they bind to voids. We use our framework to address two fundamental questions about generic quantum dynamics with a conserved charge. First, we argue that gapped Ruelle-Pollicott resonances are absent in the weak-noise limit, even in sectors of operator space that contain no hydrodynamic slow modes: instead, the spectral gap in all sectors vanishes nonperturbatively in the noise strength. Second, we compute the spacetime asymptotics of the dynamical single-particle Green's function, both in the weak-noise regime and in the absence of noise. In the noiseless case, we find that the void-coherence polaron undergoes subdiffusion, with an exponent we calculate. We support our general arguments with a microscopic derivation for random charge-conserving circuits, as well as numerical evidence from tensor-network simulations.

We study dissipative phase transitions in a system of two coupled fully-connected quantum Ising models interacting with an environment. The dynamics is governed by a Lindblad master equation combining coherent unitary evolution and incoherent dissipative processes, where the unitary part is described within a self-consistent mean-field framework effectively acting on the local Hilbert space of two coupled spins at each site. We analyze two fundamentally different classes of dissipators. In the first case, the jump operators are defined in the instantaneous eigenbasis of the mean-field Hamiltonian and satisfy a detailed-balance condition. In this setting, the relaxation dynamics depends strongly on the quench protocol: a parametric quench of the Hamiltonian leads to conventional relaxation, whereas a temperature quench gives rise to a dynamical phase transition characterized by nonanalytic behavior in time. Yet, in both cases, the system relaxes toward a steady state determined solely by the post-quench parameters and the bath temperature, which closely resembles a thermal Gibbs state of the mean-field Hamiltonian. As a result, the dissipative phase transition occurs at a critical point consistent with the corresponding equilibrium transition. In contrast, when the dissipators are realized via local spin raising and lowering operators, the steady state is genuinely nonequilibrium, leading to a significantly richer phase diagram. In particular, for sufficiently strong system-bath coupling, we observe a reentrant phase featuring a symmetry-broken region bounded by two continuous dissipative phase transitions. Our results evidence how the structure of dissipative processes controls the emergence of equilibrium-like versus genuinely nonequilibrium critical behavior in open quantum systems.

Nonlocal magic quantifies the irreducible nonstabilizerness of a bipartite quantum state after optimizing over local basis changes. We study nonlocal magic for pure fermionic Gaussian states, and derive a simple closed-form entanglement spectrum bound in terms of the singular values of the subsystem-restricted covariance matrix. We benchmark our result against simulated annealing over local Gaussian unitary transformations, which supports optimality along the full local Gaussian orbit. For states drawn from the Gaussian Haar ensemble, we show that the average nonlocal magic is extensive and determine its thermodynamic limit using random matrix theory for the appropriate circular unitary ensemble. We also study Gaussian ground states, focusing on the Kitaev chain, and find that nonlocal magic is suppressed deep in both trivial and topological phases and peaks near the critical points. Finally, we investigate Gaussian evolution via random circuits and in quenches with the XY chain. For random circuits, we find that nonlocal magic grows diffusively, while in the XY chain the XX limit reveals a striking separation between nonlocal magic and entanglement.

Entanglement and magic are fundamental resources that capture the complexity of quantum many-body systems. Non-local magic isolates the irreducible nonstabilizerness intrinsically tied to entanglement. However, evaluating this quantity generally requires a prohibitive minimization over the full Hilbert space, making it computationally inaccessible beyond a few qubits. Here, we overcome this bottleneck by suggesting a closed-form expression for the non-local stabilizer entropies of fermionic Gaussian states over local Gaussian unitaries, which can be evaluated in polynomial time directly from the eigenvalues of the reduced Majorana covariance matrix. We apply this framework to characterize fermionic non-local magic across diverse physical regimes: we derive an exact Page-like curve for typical random states, reveal logarithmic scaling at the quantum critical point of the XY model, and establish a quasiparticle picture for magic generation during out-of-equilibrium quantum quenches. Crucially, because our result relies solely on two-point correlation functions, it provides a scalable route for the experimental estimation of fermionic non-local magic in large-scale quantum processors via fermionic shadow tomography.

Multi-agent AI architectures whose dominant collective observable is a $k$-body spin correlator $O_k\equiv\langleφ^k\rangle$ over a $\mathbb{Z}_2$-symmetric order parameter exhibit composite-operator criticality with effective exponents $β_k = k/2$ and $γ_k = 2-k$, thereby producing a finite susceptibility for $k\geq2$ and a vanishing susceptibility for $k\geq3$. This is a qualitative departure from all pairwise-network universality classes. We derive these results for the first non-trivial case $k=3$ as presented in COGENT$^3$ (Salazar, 2026). The formation transition of COGENT$^3$ and comparable models, under controlled universality and mean-field arguments, reduces to an exactly solvable triadic Ising model. The minimal triad Hamiltonian admits an exact partition function on $\{-1,+1\}^3$, with crossover temperature $T^*=4(J+γw)/\ln 3$ and mean-field critical point $T_c=J+γw$ (gradient coupling cooperatively enhancing order). The formation correlator $Ψ_{\rm form}\equiv\langleφ_iφ_jφ_k\rangle$ scales as $(T_c-T)^{3/2}$. Its conjugate susceptibility vanishes at $T_c$, confirmed by an independent field-theoretic two-point function argument. A Mori-Zwanzig memory ansatz yields a continuously tunable dynamical exponent, completing the composite-operator scaling regime.

This article presents a systematic review of theoretical and experimental findings for bound states of two and several dissipative solitons in fiber lasers. The theoretical basis underlying the formation and stabilization of soliton molecules in the fibers, which is provided by the complex Ginzburg-Landau equations and bound states of such equations, is presented in necessary detail, which is followed by a detailed presentation of experimental findings, including very recent ones. In particular, included are the results for the multi-soliton bound states in the fibers, as well as for the bound states in the temporal and frequency domains, single-component (scalar) and two-component (vector), two- and multi-soliton modes, as well as for bound states of spatiotemporal dissipative solitons in the lasers based on multimode fibers.

A Killing tensor field on a Riemannian space corresponds to an integral of the geodesic flow polynomial in momenta. A Killing tensor field is called decomposable if it is a polynomial in Killing vector fields. In this paper, we first prove that the study of Killing tensor fields on symmetric spaces can be reduced to the case of compact irreducible ones. Then we introduce the class of top slot Killing tensor fields. We obtain an explicit and elegant description of such tensor fields and prove that the quadratic Killing tensor fields are spanned by the top-slot ones. We also show that quadratic Killing tensor fields on the quaternionic projective space and on the Cayley projective space are spanned by the indecomposable ones constructed in our earlier paper and the decomposable ones. This completes the classification of quadratic Killing tensor fields on Riemannian symmetric spaces of rank one.

The Amazon rainforest is considered one of the core tipping elements in the climate system with a potential tipping point from rainforest to savannah between 2 and 6 °C of global warming. However, ongoing deforestation constitutes an additional major threat to the Amazon rainforest that acts simultaneously to undermine the stability of the rainforest. Both effects could synergistically compound and lower the overall threshold in global warming and deforestation when tipping points may be crossed. Here, we quantify the safe operating space of the Amazon rainforest, which we define as the joint global warming and deforestation conditions where resilience of the system as a whole is preserved. Based on the underlying environmental data from a global climate model, we use a reduced complexity model and explicitly take into account the adaptive capacities of the forest as well as the atmospheric moisture recycling. We quantify that under current conditions of around 1.4 °C of global warming and around 17 % of deforestation, more than a third of the Amazon rainforest is at high risk of crossing critical thresholds. We therefore conclude that the Amazon rainforest may have already left its safe operating space. Furthermore, we find that the historic and projected deforestation pattern could be particularly detrimental. Our results support the need for ambitious climate action to hold the Paris climate target and also nature protection to end net deforestation.

Arabidopsis roots show oscillatory growth patterns on homogeneous agar surfaces, whereas other plants, such as maize, do not. Although several explanations have been proposed, a simple and general model that makes testable predictions across species has been lacking. Roots sense gravity and correct their growth direction towards the vertical. Motivated by recent evidence for a time delay in this gravitropic correction, we develop a minimal nonlinear model based on the delay hypothesis that predicts whether a root oscillates or grows vertically downwards. The model identifies a fourfold relation between the delay and time period, robust across different response functions. Analysing images of Arabidopsis, we find that the mode of the oscillatory arc length is not significantly different between inclined and vertical growth conditions. The quantitative agreement between the experimentally measured oscillatory arc length and the arc length estimated from estimated root growth speed and response delay supports this fourfold delay-period rule for delay-driven root oscillations. The simplicity of our model allows for a direct comparison with data from diverse plant species.

This paper develops the numerical inverse scattering transform (NIST) framework for the coupled modified Korteweg-de Vries (mKdV) equation based on its associated Riemann-Hilbert problem. The coupled system gives rise to a $3\times3$ matrix-valued Riemann-Hilbert problem, whose jump matrix and scattering data have a more involved structure than in the scalar case. This matrix setting makes the extension of NIST to the coupled system nontrivial, both in the direct scattering computation and in the numerical solution of the inverse problem. Within this framework, the scattering data are first computed by solving the matrix direct scattering problem using a Chebyshev collocation method with suitable mappings. The Deift-Zhou nonlinear steepest descent method is then used to analyze and deform the oscillatory Riemann-Hilbert problem. In particular, the phase function admits two stationary points symmetric about the origin, and the analysis leads to a division of the $(x,t)$-plane into three regions with corresponding contour deformations. Compared with traditional numerical methods, the NIST computes the solution directly at prescribed spatial and temporal points without relying on time-stepping. Numerical experiments illustrate the performance of the proposed NIST in long-time simulations and indicate that it captures the main asymptotic features of the coupled mKdV solutions.

Winds and jets are symbiotic when the accretion rate is low, according to black hole accretion theory. Both components are potentially important for active galactic nucleus (AGN) feedback, but previous works typically include only jets with free parameters. We perform hydrodynamical simulations of an isolated elliptical galaxy with both jets and winds included. The key features discriminating our simulations from others are that our simulations resolve the Bondi radius for reliable black hole accretion rate calculation and use parameters from GRMHD simulations. By selectively activating jets and winds, we examine their individual and combined effects. We find that effective AGN feedback, which is capable of generating strong turbulence and subsequently increasing central gas entropy and suppressing cool gas condensation and star formation, occurs only when both jets and winds operate simultaneously. The physical mechanism is the interaction between winds and jets: this interaction produces strong shear at their interface, leading to turbulence via the Kelvin-Helmholtz instability. In contrast, neither jets nor winds alone can generate strong turbulence due to the insufficient shear. The turbulence produced by wind-jet interaction is predominantly solenoidal in nature, giving rise to a broad energy spectrum approximately following a Kolmogorov-like power law and a dissipation rate $\sim 10^{-27}\,\mathrm{erg\,cm^{-3}\,s^{-1}}$ in the interstellar medium, consistent with observations. Our findings highlight the importance of simultaneously considering both jets and winds in studying the effects of AGN feedback in the evolution of elliptical galaxies.

Vector-borne diseases often infect multiple host species, increasing the likelihood of disease persistence due to the presence of multiple reservoirs. Vector biting patterns and feeding preferences can shift in response to selective pressures introduced by disease control interventions, altering the dynamics of transmission. In this paper, we develop a mathematical model that couples host diversity and adaptive vector behavior with vector-borne disease transmission dynamics, focusing on a system with two hosts and a vector population exhibiting preference for one host. We derive the basic reproduction number, $R_0$, a threshold that determines the existence of two equilibria in our model, and obtain conditions that can lead to the long-term persistence of the disease. Our analysis suggests that shortening the infectious period of the vector's preferred host is an effective control strategy. We also identified a threshold that determines whether shifting vector preference toward a non-preferred host amplifies or reduces the disease burden on the primary preferred host. Our results show that protective measures for the preferred host can trigger adaptive shifts in vector preferences, reducing disease prevalence in that host. However, this shift may lead to an increase in overall host prevalence.

We discovered generalized structures, named astrocytes due to their shape, that constitute a defined region characterizing regular behavior within the parameter plane (PP) of dynamical systems (DSs). Morphologically, they are characterized by a branch and a soma with several vertices (arms) and sometimes with multiple periodicities. A bunch of infinite astrocytes emerge through their branches from a region, in general, of low periodicity. Astrocytes are embedded in a quasiperiodic-chaotic scenario. The soma complexity (number of vertices) determines a kind of hierarchy of the astrocytes; moreover, bunches of subsequent structures from the astrocyte have been emphasized, revealing a self-similarity property. We conducted a detailed analysis in a Zeeman laser model, but we also observed astrocytes in many other DSs. The multiperiodicity exhibited by the astrocytes in their soma gives rise to harlequin dress-like patterns and tri-, quad-, and quint-critical points, which indicate the coexistence of different higher-order periodicities. In the concave borders of the soma, a doubling cascade of quint-points emerges as a bifurcation in the PP, defining regions of ordered sequences of higher periodicity in the route to chaos.

We study a generalisation of the set-theoretic Yang-Baxter equation and investigate the connection between its solutions and matrix refactorisation problems. We refer to such solutions as scalene Yang-Baxter maps. Moreover, we construct scalene Yang-Baxter maps associated with integrable equations of KdV and NLS type.

We investigate, both analytically and numerically, the scattering of quasi-one-dimensional quantum droplets from Pöschl-Teller potential wells and barriers. For attractive wells, we find a sharp transition between complete reflection and transmission at a critical incident velocity for both small and large flat-top droplets. The scattering interactions differ: small, soliton-like droplets form a spatially symmetric trapped mode at the critical velocity, showing their compressibility and coherence characteristics, while large droplets develop a spatially asymmetric trapped state, revealing incompressibility and internal structure. The critical velocity depends non-monotonically on atom number: it rises in the small, compressible-droplet regime, falls in the incompressible, flat-top regime, and turns at the crossover point. We also show that the reflectionless well generates a $π$-phase shift, strongly altering droplet-droplet collisions relative to free space. The persistence of a confined mode after collisions between trapped and incident droplets depends sensitively on their relative phase. For the repulsive barrier, we identify regimes of complete reflection, partial return, and full transmission, depending on incident velocity, barrier height, and particle number. Our predictions match direct numerical simulations in all cases.

Differential-difference matrix Lax representations (Lax pairs), gauge transformations, and discrete Miura-type transformations (MTs) belong to the main tools in the theory of (nonlinear) integrable differential-difference equations. For a given equation, two matrix Lax representations (MLRs) are said to be gauge equivalent if one of them can be obtained from the other by applying a matrix gauge transformation. Generalizing and extending several previous works on MLRs and MTs, we present new results on the following problems: - When and how can one simplify a given MLR by means of gauge transformations? - How can one use MLRs and gauge transformations for constructing MTs? - A MLR is called fake if it is gauge equivalent to a trivial MLR. How to determine whether a given MLR is not fake? We consider the general (1+1)-dimensional evolutionary differential-difference case when a MLR can depend on any shifts of dependent variables and can be non-autonomous. As applications and illustrations of the presented general theory, we construct several new two-component integrable equations (with new MLRs) connected by new MTs to known integrable equations from the papers [S. Konstantinou-Rizos, A.V. Mikhailov, P. Xenitidis, J. Math. Phys. 2015], [E. Mansfield, G. Mari Beffa, Jing Ping Wang, Found. Comput. Math. 2013]), including non-autonomous examples.

This paper investigates the asymptotic behavior of high-order vector rogue wave (RW) solutions of the coupled nonlinear Schrödinger (CNLS) equation in the presence of multiple large internal parameters. We report several new high-order RW patterns in the CNLS system, including double-sector, double-heart, and mixed sector-heart configurations. The main novelty is that each RW pattern contains two distinct regions in which two different fundamental first-order RWs coexist simultaneously, potentially appearing as bright (eye-shaped) versus four-petaled or dark (anti-eye-shaped) forms. These two regions are respectively associated with the simple root structures of two different Adler--Moser polynomials: each region consists of well-separated first-order RWs in one-to-one correspondence with the simple roots of the associated polynomial. In addition, by tuning certain free parameters, the two regions of the RW pattern can be shifted to arbitrary locations in the $ (x,t) $-plane. This flexibility, together with the rich simple-root structures of Adler--Moser polynomials, enables the systematic generation of a much broader family of structured RW patterns in the CNLS equation.

We investigate topological vector potentials underlying the phases of nonlinear waves by performing Dirac's magnetic monopole theory in an extended complex plane, taking into account self-steepening effects while ignoring the usual cubic nonlinearities. We uncover that the simple poles and third-order poles of the density function constitute virtual monopole fields with higher charges $\pm3/2$ and $\pm5/2$, respectively. These results are in sharp contrast to the previous findings, where the simple zeros of the density function yield charges $\pm1/2$. We choose scalar and vector rogue waves as well as bright solitons to demonstrate the Dirac monopole potentials. These results confirm a series of quantized magnetic charges for virtual monopoles underlying nonlinear waves, and reveal new relations between poles of density functions and topological charges.

In this paper, we study generating series enumerating polygonal angulations of closed oriented surfaces of fixed genus, focusing on $b$-angulations with $b = 3$ or $b = 2ν$, $ν\geq 2$. Based on Toda integrability, we establish new structural results in the cases $b = 3$ and $b = 4$. Furthermore, via the Hodge--GUE correspondence, we derive a fine structure in the $b = 2ν$ case, which implies a conjectural statement of Gharakhloo--Latimer.

We present a minimal agent-based model of interacting agents characterized by their wealth to study taxation and inequality in a non-conservative economy. Wealth evolves through an extremal stochastic replacement process in which the poorest agent has its wealth replaced by a new random value, financed through a collective taxation mechanism. We explore taxation regimes ranging from regressive to progressive schemes and tune the overall redistribution strength. Under regressive taxation, the system self-organizes into two distinct stationary phases when changing the total tax collected: a non-ergodic, high-inequality regime characterized by wealth condensation in a subset of agents that permanently escape replacement, and a more homogeneous ergodic phase in which all agents participate in the dynamics. Increasing taxes drives an abrupt transition between these phases. The transition is discontinuous and exhibits hysteresis and bistability, consistently detected through the Gini index, the Top $1\%$ wealth share, the entropy, and the Binder cumulant. In contrast, neutral and progressive taxation suppress persistent wealth concentration, preventing the emergence of strongly unequal states and eliminating hysteretic behavior. These results show that minimal stochastic redistribution mechanisms alone can produce discontinuous transitions, metastability, and non-ergodicity, demonstrating that taxation structure can determine the emergence and stability of macroscopic inequality.

The Quantum Lattice Boltzmann Method (QLBM) has emerged as one of the most promising quantum computing approaches for the numerical simulation of problems in computational fluid dynamics (CFD). The dynamics is formulated in terms of mesoscopic particle distribution functions governed by a discrete Boltzmann transport equation, comprising local streaming and collision operations. In this work, the resulting macroscopic behavior corresponds to the advection-diffusion equation, which we adopt as a canonical model problem for transport phenomena. Building upon recent progress in QLBM implementations, we advance towards more realistic problem settings that better reflect conventional CFD requirements. We address, for the first time, transport under the action of non uniform velocity fields on quantum hardware. We implement our demonstration using IonQ's trapped-ion systems including Forte generation systems and a 64-qubit Barium development system similar to the forthcoming IonQ Tempo line. We identify the density readout and subsequent reloading of the fluid density as a potential bottleneck of the current algorithm and discuss several approaches to mitigate this bottleneck. We identify the use of MPS shadow tomography as a promising method to efficiently scale the readout to large system with complex density distributions. Lastly, we introduce and simulate a novel method to implement wall boundaries for advection-diffusion in QLBM, and discuss the prospects of scaling to higher-complexity problems.

Sensitive biomarker detection in physiological fluids is often limited by Debye screening, which suppresses electrostatic signals at sensor surfaces. Here we report a sensing approach based on flexoelectric resonance in silicon nanowire field-effect transistors. An applied radiofrequency field induces strain gradients in the nanowires, generating flexoelectric polarization that is amplified at resonant frequencies. This effect enhances the sensitivity of conductance measurements to small surface charge variations associated with biomolecular binding. Using C-reactive protein as a model biomarker, we observe an order-of-magnitude improvement in detection sensitivity compared to conventional operation, with a 62% conductance increase versus 30% without radiofrequency modulation. The high-frequency field also perturbs the electrical double layer, reducing Debye screening in high-ionic-strength environments. These combined effects enable direct biomarker detection without sample dilution. This work establishes flexoelectric resonance as a general strategy for improving nanoscale biosensing performance in physiologically relevant conditions.

Semiconductor-based plasmonic nanostructures support localized surface plasmon modes in the infrared region. Unlike metallic nanostructures, they support both free electrons and holes, requiring a two-fluid hydrodynamic Drude equation (HDE) to accurately capture spatial dispersion effects and low-frequency acoustic plasmon modes that cannot be described by single-fluid models. In this work, a volume integral equation (VIE)-based solver is proposed for the analysis of electromagnetic scattering from semiconductor nanostructures. The proposed approach couples the VIE, formulated in terms of the electric flux density and the free-electron and hole polarization currents, with the two-fluid HDE. The coupled system is discretized using a tetrahedral mesh and solved efficiently using a two-level iterative solver. In contrast to finite-element-based methods, the proposed VIE-based approach does not require domain-wide meshing and inherently satisfies the radiation condition, thereby eliminating artificial absorbing boundaries. Numerical results for InSb-type semiconductor nanostructures demonstrate the accuracy and efficiency of the proposed VIE-based solver and its ability to capture unique optical phenomena, such as acoustic plasmon resonances and the blueshift of localized surface plasmon resonances, that cannot be described by the single-fluid HDE or classical Drude-based models.

To couple many independent modes from free space to on chip, the key challenge is not enhancing the many necessary coupling rates (scattering-matrix elements) between targeted mode pairs. Instead, the key is to avoid additional cross-couplings to undesired modes, due to the presence of multiple simultaneously satisfied phase-matching conditions. With this principle, we identify scaling laws for the maximum number of high-efficiency multi-mode couplings that may be achievable for a given refractive index and design region, which are strongly supported by extensive numerical inverse-design experiments in 2D (one-dimensional coupler patterns, scattering in 2D). For such couplers, typical mode counts of 5--10 appear achievable. Three-dimensional couplers (patterned across two dimensions) can be markedly better, with tens of Fourier components in a single-layer device offering the possibility of high-efficiency coupling of hundreds to thousands of modes in relatively compact form factors. Numerical simulations of such a device, without any parameter optimization, predict efficiencies on the order of 5\% for 100 modes -- a collective order-of-magnitude improvement over previous designs.

Although intersections are the most complex parts of the roadway network, pedestrian crashes at non-intersection locations are disproportionately frequent, highlighting a serious traffic safety concern. This study investigates non-intersection crashes involving pedestrians using a crash database (2017-2021) collected from Louisiana State. As the risk of pedestrian crashes tends to vary with distance from the intersection, the research team utilized a unique framework "distance to intersection" to capture the differences in crash patterns at non-intersection locations. The study identified that around 50% of non-intersection pedestrian crashes occurred within 198 ft. of the intersection. In the next step, the collected 3,135 pedestrian crashes at non-intersection locations during the study period were subdivided into three zones: D1 zone designates crashes occurring within 150 ft. of an intersection (1,277 crashes), D2 zone designates crashes occurring within 151 ft. to 435 ft. of an intersection (1,060 crashes) and D3 zone designates crashes occurring at 435 ft. or higher from an intersection (798 crashes). To explore the complex interaction of multiple factors, an intuitive data mining technique, Association Rules Mining was used. A total of the top 60 interesting association rules (20 for each zone) were identified by the algorithm (based on lift and support measures). In addition, a total of 124 rules were explored based on Lift Increase Criterion (LIC) measure. The findings of this research provide critical insights into pedestrian crash involvement at non-intersection locations and the variation in crash patterns according to the "distance to intersection". Based on the findings, some of the targeted problem-specific countermeasures are also recommended to address the crash patterns at non-intersection locations.

We apply a systematic inverse design approach to discover foundry-compliant, multilayer grating couplers that can efficiently couple a number of independent waves from free space to on-chip propagating modes. For visible- and near-infrared couplers, we find that minimum feature sizes are by far the most important constraint to tailor the design algorithms around. If, additionally, one forces the optimization to be robust to over- and under-etch errors, the resulting designs exhibit stable optimal efficiencies in the presence of other imperfections (critical dimension variations, overlay mismatch, and sidewall angle variation). The foundry-compliant designs exhibit moderate efficiency penalties as feature sizes increase, but no change to simple underlying scaling laws with respect to requisite numbers of layers and layer thicknesses. These results establish a practical, generalizable framework for high-efficiency multimode coupling within the constraints of modern semiconductor foundries.

Optical clocks have achieved remarkable estimated fractional frequency uncertainties reaching the $10^{-18}$ level and below, enabling applications in fundamental physics, general relativity, and geodesy. However, the challenge of verifying the international consistency of optical clocks remains critical as efforts intensify toward redefining the SI second based on an optical transition or transitions. We report on a two-month international clock comparison campaign involving seven optical clocks in four national metrology institutes (INRIM, LNE-OP, NPL, and PTB) connected via the optical fiber network established in Europe. The campaign resulted in optical frequency ratios with uncertainties ranging from $7.7\times10^{-18}$ to $6.1\times10^{-17}$. Among the results, the $^{171}$Yb$^+$(E3) clocks at NPL and PTB demonstrated agreement within an uncertainty of $7.7\times10^{-18}$, marking the first international verification of two independently developed optical clocks below one part in $10^{17}$. The operation of the $^{199}$Hg clock at LNE-OP (formerly LNE-SYRTE) resulted in frequency ratios with improved uncertainties with $^{171}$Yb$^+$(E3), $^{171}$Yb, and $^{87}$Sr optical clocks. These results provide input for the redefinition of the second and underscore how fiber-linked clock networks can advance metrology and scientific applications.

Clustering is a fundamental collective phenomenon in agent-based models (ABMs) of opinion dynamics. To study clustering in systems with co-evolving social and opinion variables, we derive stochastic partial differential equation (SPDE) models that describe the evolution of clusters on a reduced state space. We consider two settings: one in which opinions do not affect social interactions, and another one in which a feedback mechanism couples the two. Our approach extends reduced PDE modelling to a stochastic framework, which is essential for capturing long-term cluster behaviour. Numerical experiments demonstrate that the proposed reduced SPDEs substantially decrease computational cost compared to full-state SPDE models, such as the Dean-Kawasaki equation, while still accurately reproducing the clustering behaviour of the underlying ABM. As a result, these reduced models provide an efficient tool for studying systems with large populations, including those arising in the analysis of real-world data: in particular, we provide an application related to the large-scale General Social Survey (GSS), which comprises opinion and social data of the US population since 1972.

The colonisation of extraterrestrial planets requires sustainable food production independent of Earth-based supplies. Due to the high costs and complicated logistics of food transport, in-situ cultivation will be essential. Growing plants directly in regolith offers a practical approach to achieve sustainable long-term human habitation beyond Earth. In this study, Laser-Induced Breakdown Spectroscopy (LIBS) technique was employed for bioimaging of broccoli (Brassica oleracea) and salad (Lactuca sativa) plants grown in Lunar regolith simulant and control substrate. For this purpose, the potential of the 2090 nm laser wavelength for bioimaging of plant tissue was studied compared to the conventional 1064 nm. The signal-to-noise ratio (SNR), total emissivity ($ε_{\mathrm{tot}}$), and Mg II / Mg I intensity ratio (ionisation degree) were all higher when using the 2090 nm laser wavelength compared to 1064 nm. These findings indicate that the 2090 nm laser produces a hotter and more efficiently ionised plasma, supporting its feasibility for bioimaging of plant tissues. Additionally, bioimaging with both laser wavelengths confirmed higher uptake of key plant nutrients such as magnesium (Mg) and calcium (Ca) from Lunar regolith simulant. These results support the potential of LIBS as a diagnostic tool for plant growth monitoring in extraterrestrial environments.

Primates exhibit a robust deviation from canonical allometric scaling: at fixed body mass, their lifespans exceed those of non-primate mammals by factors of two to three. A rhesus macaque (8 kg) lives 25-40 years, whereas a cat of similar mass rarely exceeds 18 years. This statistically significant clade-level excess cannot be explained by standard metabolic or ecological models. We provide a thermodynamic explanation within the Principle of Biological Time Equivalence (PBTE), where lifespan is determined by a finite cycle budget governed by entropy production. We show that primates reduce entropy production per physiological cycle through increased neural energy allocation. The neural power fraction acts as a control parameter, extending the effective lifetime cycle count. Three mechanisms, predictive regulation, enhanced repair, and behavioral buffering, jointly suppress dissipation. This yields a quantitative neuro-metabolic multiplier that explains primate longevity and provides testable predictions linking brain energetics, entropy production, and lifespan.

In heterogeneous network systems such as ecological and social networks, structural stability depends on how connectivity changes under node removal, as different removal sequences can trigger distinct modes of systemic collapse. While robustness to random failures and targeted attacks has been extensively studied, most analyses have focused on connectivity loss or degree distribution, rather than on how scale-invariant organization emerges and evolves with system size. Here we examine how scale-free structure evolves under progressive degree-dependent node removal, systematically varying the hub-protection strength $θ$. Starting from scale-free networks, we apply the recently developed finite-size scaling (FSS) analysis to node-removed networks and compare the results with those from Kullback-Leibler (KL) divergence-based classification. We find that under random ($θ=0$) and hub-protecting removal ($θ>0$), the two criteria largely agree, whereas under hub-preferential removal ($θ<0$), networks may appear scale-free according to the KL criterion while failing the FSS test of scaling collapse. This discrepancy indicates that similarity to a reference degree distribution does not guarantee the persistence of scale-invariant organization across system sizes. The two diagnostics thus probe complementary aspects of network structure, and their joint use provides a more complete characterization of structural degradation.

Predictive simulation of electrochemical interfaces requires atomistic models that capture reactive bond rearrangements, long-range electrostatics, and charge distributions reflecting the electronic distinctness of electrode and electrolyte. Existing charge-aware machine-learned interatomic potentials (MLIPs) built on global charge equilibration (QEq) settle electrode and electrolyte at a common electrochemical potential, leaving no room for the interfacial gradient that the double layer requires and admitting spurious charge transfer between electronically disconnected regions. Per-fragment charge equilibration is the established remedy in classical molecular dynamics, but reliance on predefined molecular topology has confined it to non-reactive systems. We lift this restriction by making fragment identification itself a differentiable function of atomic geometry, yielding soft fragment-constrained charge equilibration (Soft-FQEq) -- a solver layer that restores fragment-resolved charge conservation in reactive MLIPs. The layer consumes four scalar MLP readouts from a shared atomic-feature network -- per-atom electronegativity, source charge, short-range energy, and a soft bond connectivity -- and returns equilibrated charges together with per-fragment chemical potentials. We implement Soft-FQEq as an extension of the hippynn framework on a HIP-NN feature network and train it on DFT energies, forces, and DDEC6 charges for IrO2/H2O/Na+/ClO4- interfaces. The trained model recovers a clear electrode-to-electrolyte gradient in the per-atom electrochemical potential. With the same trained weights but the fragment-constrained solver replaced by global QEq at inference, this gradient collapses to an essentially uniform profile, directly showing that the gradient cannot be sustained within global QEq while the fragment formulation recovers it.

Capturing ultrafast transient phenomena conventionally requires streak cameras or computational imaging based on compressed sensing, which lead to complex and costly systems. In this Letter, we demonstrate, to the best of our knowledge, the first fully passive single-shot ultrafast imaging architecture assembled entirely from off-the-shelf, low-cost components. A commercial microlens array combined with a stack of standard microscope cover glasses maps temporal information into multiple spatial channels, and a consumer-grade CMOS image sensor records all delayed replicas within a single camera exposure. The proposed system has a total hardware cost below US\$500 and captures the evolution of a picosecond laser pulse with a temporal sampling interval of 1.46~ps, an effective frame rate of 685~Gfps, and a sequence depth of ten frames. The temporal fidelity of the system is verified by recovering the expected Gaussian pulse profile, and the spatial resolution is characterized through a point-source measurement with a point spread function of 1.86 and 1.62 pixels full width at half maximum along the horizontal and vertical directions, respectively. The proposed architecture presents an alternative approach to single-shot ultrafast imaging with a simple, low-cost, computation-free, and fully passive design.

High-order harmonic generation (HHG) in solids - the frequency up-conversion of an optical signal - is governed by symmetries. At terahertz (THz) frequencies, HHG is a key technology to access high frequency spectral windows that are usually difficult to cover using conventional solid state laser technologies. This effect has been recently exploited in graphene where HHG has been demonstrated, albeit only at odd multiples of the driving frequency owing to its inherent centro-symmetry. In topological insulators (TIs), the combination of spin-orbit interaction and time-reversal symmetry create an insulating bulk state with an inverted band order, inseparably connected with conducting surface states. TIs have been predicted to support unconventional high harmonic generation from the bulk and topological surface, which are usually difficult to be distinguished. However, no experimental results have been provided, so far. Here, we exploit the strong optical field amplification provided by an array of single or double split ring resonators, with embedded Bi2Se3 or (InxBi(1-x))2Se3/ Bi2Se3 van der Waals heterostructures, to achieve up-conversion in the 6.4 (even) - 9.7 (odd) THz frequency range. This results from bulk centro-symmetry (odd states) and symmetry breaking in the topological surface states (odd and even).

A pygmy shrew (\textit{Suncus etruscus}, ${\approx}2$\,g) sustains a resting heart rate near $1{,}000$\,beats\,min$^{-1}$ and dies within two years; an African elephant (${\approx}4{,}000$\,kg) beats at $28$\,beats\,min$^{-1}$ and lives seven decades. Their chronological lifespans differ by a factor of 35, yet each accumulates close to $10^9$ cardiac cycles before death -- a near-constancy first noted by Rubner~(1908) and quantified by Lindstedt and Calder~(1981)~\cite{lindstedt1981}, but never subjected to multi-clade statistical testing, phylogenetic correction, or explicit falsifiability criteria with a large modern dataset. We address this gap with a curated 230-species vertebrate dataset spanning non-primate placentals ($n=43$), primates ($n=18$), marsupials and monotremes ($n=19$), duty-cycle-corrected bats ($n=31$), dive-corrected cetaceans ($n=12$), birds ($n=78$), and Arrhenius-corrected ectotherms ($n=26$), and subject the log-invariant $\ell = \log_{10}(N^{\!\star})$ -- where $N^{\!\star} = f_H\,L\times 525{,}960$ cardiac cycles -- to four independent tests.

Degradation of near surface nitrogen vacancy (NV) centers in diamond under optical illumination has restricted their deployment in applications such as scanning NV magnetomety, particularly under harsh environment such as low temperatures and vacuum. Previously, alumina passivation of planar diamond samples has been shown to reduce the degradation of near surface ensemble NV centers in vacuum. Here, we expand this study to incorporate photonic nanostructures by analyzing the single photon emission characteristics of NV centers embedded in an array of alumina-coated diamond nanopillars in high vacuum and low temperature (6K, high vacuum) environments under non-resonant (522 nm) laser exposure. We find that, in contrast to the oxygen-terminated diamond nanopillars, NV centers in the alumina-coated nanopillars demonstrate negligible change in the single photon purity and brightness over the course of laser exposure in vacuum. At low temperature, NV centers under alumina termination demonstrate stable single photon emission, whereas under oxygen termination the single photon purity degrades under high intensity laser exposure. Alumina surface passivation is therefore shown as a viable path toward the realization of robust NV-diamond based nanoscale sensing under non-ambient atmospheric environments, including using diamond scanning probes.

The excitation of plasma wakefields driven by chirped laser pulses is investigated using a reduced relativistic fluid Poisson model supported by fully relativistic particle in cell (PIC) simulations. The study considers exponential, linear, quadratic, and unchirped phase-modulated laser drivers propagating in an underdense plasma. Numerical solutions of the governing equations demonstrate that exponential chirping produces enhanced wakefield amplitudes compared to polynomial and unchirped cases due to nonlinear phase variation across the pulse envelope. The analytical predictions are validated using quasi cylindrical PIC simulations performed under identical plasma and laser parameters. The simulations reveal strong chirp dependent wakefield modification, with positively chirped pulses generating peak accelerating fields exceeding 58 GV per m, accompanied by pronounced density compression and enhanced electron momentum gain. These results demonstrate that exponential chirping provides an effective mechanism for controlling wakefield strength and improving plasma based particle acceleration.

We present YOSO (You Only Shot Once), a single-frame phase retrieval framework for digital in-line holographic microscopy (DIHM) in which supervised deep learning is used to numerically generate an additional hologram corresponding to different defocus distance, creating a so-called multi-height dataset, which is then conventionally processed with a well-established Gerchberg-Saxton (GS) algorithm. YOSO is trained on computer-generated data derived from natural images, enabling strong generalization. The selected multi-scale ResNet architecture enables rapid training in under two hours on a mid-range workstation, which is done only once, enabling efficient inference thereafter. We further show that YOSO network can process inputs of varying spatial dimensions, allowing training on small inputs and direct inference on full-sized holograms while bypassing patch-and-stitch procedure. A further advantage of YOSO is its physics-consistent hologram padding, which replaces conventional zero or edge-value padding with a physically grounded approach compatible with the GS framework. The YOSO framework is tested on various systems (lens-based and lensless DIHM) and diverse samples: a resolution test target, adherent and suspended biological cells, and a mouse brain slice. The results show that YOSO is compatible with 3D objects and correctly recovers defocused object wave features, enabling holographic postprocessing such as numerical refocusing. The results of this work are available publicly as software for end-to-end implementation.

Photoacoustic imaging is the leading technique for deep tissue optical imaging, allowing single-shot imaging at depths. However, its resolution may be limited by acoustic aberrations, caused by natural unknown heterogeneities in the tissue speed of sound. In recent years, reflection-matrix based scattering-compensation techniques have been successfully employed in ultrasound, optics, and seismology, to computationally correct such distortions. However, they have not been adapted to photoacoustic imaging since they rely on multiple acquisitions under different controlled excitations, such as input plane-wave illuminations, which do not result in signal changes in photoacoustics. Here, we introduce a framework that enables the direct application of the state-of-the-art reflection-matrix based aberration correction techniques to photoacoustic imaging of dynamic targets. Specifically, we show that a covariance matrix analysis of a conventional set of photoacoustic frames of dynamic targets, such as flowing red blood cells in blood vessels, yields a virtual reflection-matrix that is mathematically analogous to a pulse-echo reflection-matrix, and lends itself to direct processing by conventional reflection-matrix based scattering-compensation algorithms. We validate and demonstrate the approach for photoacoustic aberration correction of vessel-mimicking targets containing flowing absorbers in both simulations and experiments.

Multispectral polarisation imaging has a broad range of applications, from biological cell imaging to agricultural remote surveying. For such applications, especially involving lightweight unmanned aerial vehicles like drones, it is necessary to have compact, single-shot, efficient optical systems. We present a metasurface design that diffractively separates a scene into spectral and polarimetric measurements with a single optical component, operating for 532 nm and 700 nm in a single-shot imaging system. The polarisation imaging performance of the design is shown to be robust to both spectral and angular bandwidths, and multispectral polarimetry is demonstrated experimentally.

The polarization state of light plays a central role in strong-field light--matter interactions and is widely used to probe electronic structure in solids via high-order harmonic generation (HHG). In particular, helicity-resolved HHG has been interpreted as a fingerprint of crystal symmetry and topology. Here, we demonstrate deterministic and continuous control of harmonic helicity in solids using polarization-crafted beams, formed by two orthogonally polarized pulses with a controlled time delay. By tuning this delay, the polarization state of individual harmonics can be driven from linear to circular, independent of the material under investigation. We show that this behavior is robust across systems with distinct symmetry and topology, and originates from the sub-cycle modulation of the light--matter interaction mediated by the dipole coupling. Furthermore, the orthogonal configuration allows to break the dynamical symmetry of the light-matter interaction which is manifested in the generation of otherwise forbidden harmonics under standard selection rules.. These results establish harmonic helicity as a field-controlled observable rather than a direct material fingerprint.

We propose an OAM sorter based on a novel optical element that we refer to as a wavefront twister. It is a generalization of the conventional wavefront rotators such as the Dove prism. However, unlike a Dove prism, which simply rotates a wavefront, the rotation generated by a wavefront twister varies linearly with radial position, resulting in the twisting of the wavefront. We demonstrate that the wavefront twister, followed by a lens, maps each OAM mode to an annulus of distinct radius at the back focal plane of the lens with negligible inter-modal overlap and preserves the circular symmetry. Thus, the proposed wavefront twister offers a scalable scheme for high-dimensional OAM mode sorting, with important consequences for the practical realization of OAM-based applications.

Crowds have long held a paradoxical place in the human imagination, feared for their destructive potential yet essential for collective expression. This tension was tragically manifested in the 1919 Jallianwala Bagh massacre, when British colonial troops opened fire on a peaceful gathering in Amritsar, India. Although officially 379 deaths were recorded, eyewitnesses and historians have long challenged this figure. With this study, we critically revisit the events through the lens of the specific role of the crowd as a phenomenon, both regarding the physical and the socio-psychological dynamics. We show that even under conservative physical assumptions - moderate shooting cadence, crowd-shielding, and constrained escape routes - our agent-based simulations consistently yield fatality estimates well above the official death count. On the socio-psychological front, we explore how early 20th-century discourses, influenced by Le Bon's theory of crowd psychology, constructed the crowd as an inherently irrational and threatening entity, thus providing a rationale for the application of excessive force. Our findings show that acknowledging the socio-cultural construction of crowds as a relevant factor in how state power engage with and respond to collective gatherings brings to light contemporary parallels and the risks posed by their rhetorical framing. Furthermore, this study highlights the importance of interdisciplinary modelling for both historical accountability and current crowd safety, particularly in an era of growing political unrest, surveillance, and militarised crowd policing.

Fabricating brightly fluorescent layers with nanometric thickness and digitally controlled lateral structuration remains a challenge for next-generation photonic devices, optical calibration standards, and biocompatible interfaces. Here, we introduce Nano-Bead Emitters (NBEs), hydrogel nanoparticles covalently functionalized with fluorophores, as a universal, water-processable ink platform for fabricating programmable nanometric fluorescent architectures. By immobilizing fluorophores within a charged nanohydrogel scaffold, the platform entirely decouples film morphology from dye solubility. This molecule-independent strategy enables spectrally distinct, inherently water-insoluble dyes to be processed using a single, standardized aqueous ink formulation. Combined with laser-induced forward transfer (LIFT) printing, this additive approach yields highly uniform fluorescent layers (~7 nm thickness, sub-nanometric roughness). This structural invariance produces complex multicolor patterns sharing identical thickness and surface morphology across all spectral channels, a critical requirement for quantitative optical calibration. Furthermore, LIFT printing provides programmable, layer-by-layer control over fluorescence intensity via successive deposition cycles, yielding precisely tunable brightness without aggregation-caused quenching. This maskless technique enables rapid, high-fidelity printing of both monochromatic and multicolor patterns over macroscopic areas with absolute spatial resolution. Finally, these universally compatible NBE inks stably deposit onto diverse substrates (glass, polymers, semiconductors, metasurfaces), effectively bridging scalable manufacturing with high-performance integrated photonic systems.

This paper presents a physically-informed fuzzy clustering of vertical sounding ionograms for automatically separating the ionogram into tracks suitable for further interpretation and determining their optimal number. The model is designed for use not only in conditions where the number of tracks is known, but also in disturbed ionospheric conditions where the number of tracks is preliminary unknown. The method is based on an expectation-maximization algorithm, used for clustering, and on parametrically specified distributions of distances from points to parametrically specified curves. The curves used as track models are close to model tracks in the parabolic ionospheric layer model. The resulting model of each track has six parameters: three standard ones (the critical frequency, the lower boundary of the layer, and its half-width), and three additional ones to take into account possible underlying layer effects. By sequentially increasing the number of tracks and optimizing their parameters, the model finds the optimal number of tracks on the ionogram by minimizing the modified Bayesian information criterion. The Sequential Least Squares Quadratic Programming algorithm is used to find the parameters of a single track. The width of each single track is assumed to be unknown constant found during fitting process. To improve the quality of ionogram clustering, automatic adaptive noise filtering is performed before clustering. This filtering is based on a combination of the DBSCAN and Gaussian Mixture algorithms. Also, to improve clustering quality on an ionosonde without hardware separation of the ordinary and extraordinary components, a preliminary approximate removal of points belonging to the extraordinary mode is performed.

Biological neurons perform arithmetic computations - including additive integration and divisive gain modulation - through synaptic conductance changes and shunting inhibition, enabling context-dependent information processing that far exceeds simple threshold-and-fire models. Replicating these capabilities in compact hardware remains a fundamental challenge for neuromorphic engineering. Here, we demonstrate artificial neuron circuits based on Ovonic threshold switches (OTS) that physically implement three arithmetic operations: SUM, PARALLEL, and DIVISION. The SUM and PARALLEL neurons exploit MOSFET-controlled dendritic conductances, producing output firing rates that collapse onto invariant curves as a function of combined inputs - satisfying the canonical criteria for neuronal addition. The DIVISION neuron leverages a JFET-based shunting pathway, inspired by GABA_A-mediated inhibition in the cortex, to achieve divisive gain modulation well described by a Hill-type function (R2 ~ 0.95, Hill exponent n ~ 1.3), consistent with nonlinear normalization observed in visual and olfactory circuits. Applying the DIVISION neuron to pixel-wise image normalization under non-uniform illumination recovers obscured visual content, mirroring contrast normalization in the visual cortex. Compared to CMOS-based division implementations, the proposed approach offers improvements in energy efficiency and scalability exceeding an order of magnitude, establishing a viable path toward compact, brain-inspired analog computing.

Accurately linking microscopic energy deposition from ionizing radiation to emergent biological outcomes remains a central challenge in radiobiological modelling, particularly when stochastic damage induction, cell-cycle dynamics, and spatial organisation within irradiated tissues must be treated explicitly and consistently across scales. To address this, we introduce a stochastic agent-based radiobiological modelling framework for simulating biological response to particle irradiation, developed as an explicit single-cell extension of the Generalized Stochastic Microdosimetric Model (GSM2). Each cell is represented as an autonomous agent whose internal state, including DNA lesion counts, cell-cycle phase, and oxygenation level, evolves according to a continuous-time Markov chain driven by GSM2 transition rates. Radiation-induced damage induction, repair, misrepair, cell-cycle progression, proliferation, and migration are treated as competing stochastic events resolved through a next-event, event-driven algorithm, which provides computationally efficient scaling with system size while preserving full single-cell resolution. The framework is applied to three-dimensional tumour spheroids irradiated with 1H and 12C ions across a range of energies and dose rates. We characterise the spatiotemporal evolution of cell-cycle phase composition and spheroid volume following irradiation, and examine the dependence of cell survival on dose rate over four orders of magnitude. Several empirically established trends in biological response, including the dose-rate dependence of cell survival, its attenuation at high LET, and the inverse dose rate effect in split-dose irradiation, emerge from the model through the explicit coupling of particle arrivals, damage accumulation, and repair kinetics, without recourse to empirical correction factors as typically done.

Tunable flat optics are essential for advancing compact photonic devices. Here we show a numerical study of a reflective magneto-optical metasurface with a dynamically tunable focal length. The structure comprises bismuth iron garnet nanodisks in a Gires-Tournois resonator configuration. The magneto-optical properties of the garnet modulate the reflected phase response via an external magnetic field, allowing focusing at different focal lengths. Full-wave simulations demonstrate that the metasurface exhibits distinct focusing characteristics depending on the applied magnetic field direction for a fixed right circularly polarized incident wave at 1.550 μm. Specifically, switching the external field from +0.2 T to -0.2 T changes the focal length by a factor of approximately two (from 7.16 mm to 13.76 mm). These findings demonstrate that magneto-optical metasurfaces offer a flexible, viable approach for non-mechanical, tunable focusing in compact reflective optical components.

Significance: Jacobians, or spatially resolved sensitivity profiles, are central to image reconstruction in model-based optical tomography of biological tissue. Although Monte Carlo (MC) simulations are the gold standard for modeling light transport in turbid media, methodology for frequency- and time-domain Jacobians remains incomplete. Aim: This work extends MC to directly compute absorption and scattering Jacobians for frequency-domain (amplitude and phase) and time-domain (intensity and mean time-of-flight) measurements and prism-terminated optical fiber detectors. Approach: Jacobians are derived in the perturbation MC framework and implemented in the high-performance, open-source Monte Carlo eXtreme (MCX) simulator. Results are validated against the diffusion approximation (DA) solved using the finite element method in neonatal head models. MC with split voxels on curved surfaces is extended to Jacobian computation. The detector model is implemented in post-processing and compared with isotropic reception at surface. Results: MC- and DA-derived Jacobians show excellent agreement only in high-scattering regimes, highlighting the importance of MC for low-scattering domains. The detector model reduces surface sensitivity and marginally increases sensitivity to deeper tissues at short (< 2 cm) source-detector separations. Conclusion: A complete theoretical framework and MC software for computing frequency- and time-domain Jacobians is provided. Realistic detector modeling is encouraged for short-separation channels.

Understanding the boundary between classical and nonclassical phenomena is important for both fundamental researches in quantum optics and applications in quantum information. One of the most interesting research directions in this field is exploring nonclassical effects with classical light. In this paper, we will show that it is possible to observe antibunching with thermal light in a Hanbury Brown-Twiss interferometer by treating single-photon detectors as photon-number-resolving detectors to perform photon-number projection measurements. Both temporal and spatial antibunching is observed via the correlation of two detectors detecting one and zero photon, respectively. By comparing the measured results of thermal and laser light, it is found that the observed antibunching arises from the combined effect of photon statistics of thermal light and photon-number projection measurement.The classical and nonclassical nature of the observed antibunching is analyzed. The results are helpful to understand the connection between classical and nonclassical correlation and may find applications in multiphoton interference and quantum imaging.

The simple form of the optical theorem of scattering theory, $σ_{\rm tot}^{\rm pw} = (4π/k)\,\Im f(0)$, is valid for an incident plane wave or for a wave packet whose Fourier components possess azimuthal symmetry about the incident wave vector $\vec{k}$. Previous work has shown that this expression can break down for structured beams of light which possess orbital angular momentum (OAM), despite the fact that there is clearly no violation of unitarity, and the relevant modifications have been worked out for the case of massless photons. We present a form of the optical theorem involving neutron OAM states for the case of the scattering of massive nonrelativistic particles. We apply this form to Ryndin's theorem on the application of time reversal symmetry to forward scattering, indicate how the statement of the null condition for T violation in forward scattering is modified, and show that this effect is negligible compared with other sources of systematic error in neutron optics transmission experiments.

Choosing the optimal observable to model dynamical systems for which we do not know the driving equations is nearly always an ad hoc art. Takens' Delay Embedding Theorem guarantees a diffeomorphism between delay-coordinate vectors built from generic scalar observables and the underlying invariant attractor, but is agnostic to optimal observable choice, and formal bounds on reconstruction quality across observables are not known. Here we prove that, under modest technical conditions, the Kolmogorov-Sinai entropy of an observable predicts its reconstruction error of the underlying dynamics in chaotic, ergodic systems. Using the Oseledets Multiplicative Ergodic Theorem, we show that the tangent bundles of reconstructed manifolds admit an invariant Oseledets filtration diffeomorphically related across admissible observables, with Lyapunov exponents controlling the propagation of perturbations. We bound reconstruction error by a quantity monotonically related to the sum of positive Lyapunov exponents and, by the Ruelle inequality, the Kolmogorov-Sinai entropy. We validate this empirically on the Lorenz-63 attractor, the Hastings-Powell food chain, and a tetracosane molecular-dynamics trajectory, recovering Spearman rank correlations between $h^{KS,UB}$ and reconstruction RMSE up to $ρ=+0.89$ ($p=5.5\times 10^{-8}$) for the realistic tetracosane case, sharpening to $ρ=+0.97$ under added measurement noise. This provides a rigorous foundation for observable selection in chaotic systems, applicable as an a priori data-selection criterion for any data-driven modeling pipeline.

We analytically derive the all-order strong-field S-matrix series incorporating intermediate-state Coulomb-Volkov corrections (ICSFA). Focusing on rescattering processes described by the second-order term, we systematically investigate the impact of intermediate-state Coulomb interactions on above-threshold ionization (ATI) spectra of atomic hydrogen in linearly polarized laser fields. Crucially, ICSFA spectra demonstrate superior agreement with the results obtained by numerically solving the time-dependent Schrödinger equation compared to the standard strong-field approximation (SFA) and final-state Coulomb-corrected SFA (FCSFA). Our analysis reveals that intermediate-state Coulomb corrections enhance the yield of the third- and fourth-return-recollision trajectories while modifying interference patterns in the energy spectrum. The observed enhancement of the multi-return-recollision trajectories can be attributed to modifications of the ionization yield and scattering cross-section, which are induced by intermediate-state Coulomb effects. These effects are equivalent to the so-called Coulomb focusing effect.

Exceptional points (EP) featuring enhanced responsivity and rich dynamics have attracted extensive attentions in device developments and sensing applications. However, it remains debated whether employing EP systems is beneficial in practical sensing applications. Here, we demonstrate that a nonlinear EP in our microwave-frequency acoustic-wave oscillator improves longwave infrared (LWIR) detection under practical conditions. By phase tuning the nonlinear gain, our detector can be operated at different conditions with respect to the nonlinear EP. Compared with operation away from EP, our detector at EP shows a 33-fold improvement in responsivity and an 8.75-fold extension of 3-dB bandwidth. We observe a 6-fold enhancement in signal-to-noise ratio at an input modulation frequency of 6.2 kHz. At the incident LWIR wavelength of 9.6 um, our detector at EP exhibits a noise equivalent power (NEP) of 310 pW*Hz^-1/2 at input frequency of 10 kHz, yielding a figure of merit, product of NEP and time constant, of 9.87*10^-3 pW*Hz^-3/2, a 10-fold improvement over operation away from EP. Our integrated acoustic devices offer a versatile platform for exploring noise dynamics and developing practical sensors that exploit non-Hermitian nonlinearities.

Fostering coordinated pro-environmental behaviors at scale is a key challenge for climate mitigation. Individual actions only generate meaningful impact when they diffuse widely and become socially coordinated, yet monitoring such processes remains difficult with traditional survey-based tools alone. In this study, we examine whether large-scale online climate discourse is associated with differences in offline pro-environmental behavior across European regions. We combine geolocated Twitter data from the Climate Change Twitter Dataset (2017-2019) with survey-based measures from the 2019 Special Eurobarometer, focusing on the regional density of climate-related tweets and the average number of self-reported pro-environmental actions. We find a strong positive association between tweet density and pro-environmental behavior that remains robust to socio-economic controls, alternative spatial aggregations, and a wide range of robustness checks. To move beyond aggregate volume, we further decompose online discourse using Natural Language Processing tools that capture distinct social dimensions. While knowledge exchange shows no clear relationship with offline behavior, the prevalence of activism- and social support-related expressions is negatively associated with pro-environmental actions. Overall, our results suggest that online climate discourse can serve as an informative, attention-related signal of regional differences in pro-environmental behavior, but that different forms of online engagement relate to offline action in markedly different ways. More broadly, the study highlights the potential of integrating large-scale digital traces with survey data to investigate collective behavior in socio-environmental systems, while remaining explicitly observational in scope.

To accommodate classical communication systems with progressively increasing transmission rates, quantum access networks (QAN) have undergone systematic and protocol-level optimizations in recent years, where quantum passive optical network (QPON) architectures are gaining significant attention due to their simple structure. It is challenging for the previous QAN based on active protocols or Stokes operator coding protocols to achieve high-speed linear modulation with high extinction ratio and stability under practical conditions. In this work, we propose and experimentally demonstrate a downstream fully passive quantum access network protocol using passive state preparation (PSP) with free-space and single-mode fiber hybrid channels, and the final key generation rate is up to a record-breaking 19.48 Mbps per quantum network unit. The proposed PSP-QPON scheme extends the scope of PSP-CVQKD from point-to-point to point-to-multi-point networks, which enables high-key-rate, high-stability, and low-resource-consumption implementation. Moreover, the network channel in this experiment is fully compatible with access networks in classical optical communications, which allows integration with existing optical infrastructure without the need for additional modifications, providing a promising solution for local area network quantum access network at home or a mobile terminal.

Label-free identification and real-time tracking of biochemical substances became critical for molecular diagnostics and chemical analysis, yet conventional resonant terahertz metasurface sensing relies on a single resonance, limiting spectral selectivity and dynamic capability. Here, we suggest multiresonant membrane metasurfaces and implement them for simultaneous static molecular fingerprint retrieval and dynamic reaction monitoring within a single pixel. We consider a membrane metasurface supporting multiple quasi-bound states in the continuum designed at target frequencies and enabling the tailoring of the field enhancement and frequency-selective interaction with target analytes. As a proof-of-concept, we achieve label-free detection of the dual fingerprint absorption features of pefloxacin at 0.78 THz and 0.99 THz, and real-time tracking of vitamin C oxidation and denaturation under ambient conditions. The kinetic profiles extracted from the THz amplitude evolution show excellent agreement with nonlinear reaction models, demonstrating quantitative biochemical tracking capabilities. Our results establish a versatile and scalable THz photonic platform that unifies static fingerprint identification and dynamic reaction monitoring, paving the way toward integrated on-chip biochemical analytics and multifunctional metasurface sensors.

Epidemics have shaped human history, often with devastating consequences, motivating the development of mathematical models to understand and control their dynamics. Among the many aspects of epidemic behavior, the conditions that lead to epidemic extinction stand out as a central-if not the fundamental-question in epidemic modeling. In this work, we study epidemic extinction in a continuous SIRS (Susceptible-Infected-Recovered-Susceptible) model governed by a system of ordinary differential equations (ODEs). The model includes vaccination as a time-dependent process and considers the reinfection of recovered individuals through waning immunity. We analyze how different parameter regimes -- particularly infection, recovery, and immunity loss rates -- affect the persistence or extinction of the epidemic. Special attention is given to the limitations of continuous population models, in which the infected fraction can fall below the equivalent of a single individual, leading to nonphysical outcomes such as unrealistically long persistence or artificial secondary peaks. By comparing the continuous SIRS dynamics with expected real-world thresholds for extinction, we highlight the importance of incorporating stochasticity or discrete effects to accurately describe epidemic fade-out.

Twisted bilayer graphene aligned with hexagonal boron nitride (TBG/hBN) hosts rich topological and correlated quantum phases, such as (fractional) Chern insulators, whose character is dictated by the topology of the moiré flat band. This topology is highly sensitive to several material parameters in the continuum model, yet a systematic understanding of their combined influence has been lacking. Here, we present a comprehensive study of topological phase transitions in TBG/hBN by varying the interlayer hopping strengths ($w_0, w_1$) and hBN-induced staggered potential, both with and without the hBN moiré potential. We map out Chern number phase diagrams across a broad, experimentally relevant parameter space, revealing a progressive enrichment of the topological landscape including multiple high-Chern number ($C$ = 3, 4, and 5) states. Each transition is linked to distinct band-inversion mechanisms at generic $C_3$-symmetric k points, high-symmetry momenta, or parabolic touchings, clearly reflecting in the evolution of the Berry curvature. Our results offer theoretical insights that help interpret existing experimental observations, elucidate the mechanisms driving these topological phase transitions and facilitate the exploration of topological states in TBG/hBN and related moiré systems.

Continuous-variable quantum key distribution (CVQKD) using passive state preparation (PSP) offers low-cost, high-rate secure communication. However, the existing PSP-CVQKD scheme with a transmitted local oscillator has high photon leakage noise and poor stability, making it unsuitable for high-loss transmission. In this work, for the first time, we propose and implement a local local oscillator (LLO) CVQKD system using a self-referenced (SR) PSP scheme, and give a theoretical proof of the equivalence of the PSP and GMCS protocol using temporal-mode theory. By employing the novel self-referenced pilot scheme to achieve high-precision time-varying frequency and phase compensation algorithms, we significantly improve the system' s signal-to-noise ratio and stability. The system achieves a record-high asymptotic secret key rate of 10.34 Mbps over a free-space channel with up to 23.5 dB loss, while maintaining low excess noise and robust performance under turbulent conditions. This work establishes the feasibility of SR-LLO CVQKD, providing a practical pathway toward secure, high-rate quantum communication in realistic environments.

Thermal plasma properties play a critical role in plasma simulations and plasma-related applications. However, their strong nonlinear dependence on temperature, pressure, and gas composition makes accurate and efficient evaluation challenging. In this work, an operator learning-based model, termed DeepPropNet, is proposed for fast prediction of thermodynamic and transport properties of thermal plasmas. Two architectures are developed, including a single-property model (S-DeepPropNet) and a Mixture of Experts (MoE)-based multi-property model (MoE-DeepPropNet). The proposed models learn the nonlinear mapping from plasma operating conditions to physical properties based on high-fidelity datasets. The MoE architecture enables efficient multi-property prediction within a unified framework. Predictions are performed for binary SF6-N2 and ternary C4F7N-CO2-O2 mixtures. The results show that the proposed models achieve high accuracy, with relative L2 errors on the order of 10-3 to 10-2, while maintaining strong generalization capability under unseen conditions. The applicability of DeepPropNet is further demonstrated by coupling with finite volume method (FVM) and physics-informed neural networks (PINNs). The results indicate that DeepPropNet provides an efficient and scalable approach for plasma property prediction and plasma simulations.

We demonstrate stable operation of a thin-film lithium tantalate (TFLT) modulator at very high operating temperatures. We show that the electro-optic modulation and bandwidth of the TFLT modulators are not affected by high-temperature operation, and both waveguide and resonant modulators are DC-bias stable even at 120°C. At higher temperatures, we even observe 10% reduction of the Vπ of the modulator. Our results position TFLT modulators as a strong candidate for uncooled operation in co-packaged optics.

Beta-gallium oxide (β-Ga2O3) holds enormous potential for medium voltage range power electronic applications. This work reports VBr > 10 kV/Ron,sp = 43 mΩ*cm2 class edge terminated vertical heterojunction diodes (HJDs) with e-beam/sputtered nickel oxide (NiOx) stack on epitaxial (011) β-Ga2O3. The power figure of merit (PFOM) of the HJD exceeds 2.3 GW/cm2. The extracted parallel plane breakdown field is > 5.3 MV/cm, which is the highest reported electric field for thick (011) β-Ga2O3 epitaxial drift layer.

If sufficient training data are available, neural networks are attractive for representing missing physics in simulations, such as sub-grid scales in the coarse-mesh particle-turbulence system we consider. Physical constraints are known to both increase performance and reduce the need for data; we use the complete physics represented in the discretized governing equations as a constraint. Two-way coupled particles in two-dimensional turbulence provide a sufficiently complex system to assess effectiveness for various training data, all constructed from well-resolved simulations, in cases intentionally degraded to assess robustness. Surprisingly, using the full space-time data actually hinders model effectiveness. Instead, training that targets only spectra -- hence, neglecting phase information -- provides better closures, which is related to the well-known success of non-dissipative discretizations for simulating turbulence. It is found that some of the missing physics that lead to preferential particle concentration errors are fundamentally stochastic on the coarse mesh and therefore uncorrectable by the basic approach; a learning formulation is introduced for a Langevin-type closure to correct this. Most importantly, training just for particle kinetic energy -- without any direct input from the flow field -- also yields effective sub-grid-scale stress models. This holds true even if noise is added to the particle data, if only a sub-sample of particles are used, or if only one component of the particle velocity is used. In sum, these results show a path for inferring sub-grid-scale physics based just on particle data from experiments.

On-chip silicon photonic coupled cavity arrays (CCA) are a promising platform for quantum simulators, with access to high Quality (Q) factor resonators, tunability, and foundry compatibility. Furthermore, scalable two-dimensional (2D) silicon photonic CCAs allow for simulation of rich physical phenomena via Hamiltonian engineering. However, complete reconstruction of the Hamiltonian is limited by access to cavities in the bulk, with current approaches relying on imaging scattered light from bulk resonators. These approaches often require additional scatterers to be built in, limiting scalability, while also being hampered by imaging technology in the near-infrared range. Instead of these approaches, Hamiltonian tomography algorithms that require homodyne boundary measurements have been demonstrated in literature, however measurements of complex scattering measurements along a CCA boundary have not been shown. Here, we experimentally demonstrate an on-chip homodyne measurement setup along a single boundary of a $3\times 3$ silicon photonic racetrack resonator array and reconstruct the system's edge scattering matrix.

We demonstrate a micro-Fabry-Pérot cavity array through laser etching of high-surface-quality mirrors onto a single fused silica substrate. A cavity finesse of $4750\pm 200$ was achieved with a simple array design with $500~μm$ cavity length, $100~μm$ diameter micromirrors and $300~μm$ transverse separation. Arrays with up to 12 cavities were simultaneously tested for single mode operation, and absolute frequency measurements correlated strongly with the etched depth as measured by profilometry. Simultaneous measurements of the absolute resonant frequency for neighboring cavities showed a factor of 5 common-mode cavity drift reduction. Arrays of such cavities can be employed in chip-scale cavity QED networks (current cooperativity estimates are at the border of strong coupling for $^{87}$Rb atoms, $C=1$) as well as for precise laser stabilization at nearby wavelengths on a chip.

The standard architecture for a high-peak-power femtosecond laser is chirped pulse amplification using diffraction gratings for compression; the damage threshold of the compression gratings limits current lasers to multi-petawatt peak power. Plasma gratings have orders-of-magnitude higher damage tolerance than conventional optics, so plasma gratings with sufficiently high optical quality could allow the construction of ultra-high-power femtosecond lasers. Here, we present experimental measurements of the angular dispersion, angular bandwidth, and diffraction angles of ionization-based plasma transmission gratings and show that both the dispersive and the diffractive properties of these gratings are in close agreement with optical theory and simulations. Gratings with a period of 10.2 microns are found to have an angular dispersion of approximately 0.005 degrees/nm. The dispersion and bandwidth of these gratings suggest plausible designs for a plasma-grating-based compressor and indicate a pathway to compact lasers with petawatt to exawatt peak power.

Wave-driven desalination systems are an innovative solution to the global freshwater crisis, leveraging the complementary characteristics of seawater reverse osmosis and wave energy converters. However, the high costs of this system pose a significant barrier to widespread adoption. Optimization can help these systems reach a more competitive levelized cost of water, but the highly coupled nature of the system necessitates a multidisciplinary design optimization approach. This paper presents a holistic, multidisciplinary design optimization framework for wave-driven desalination system design, integrating models for wave energy converter hydrodynamics, power take-off transmission, seawater reverse osmosis constraints, and economic analysis. This study demonstrates the impact of multidisciplinary design optimization for wave-driven desalination systems, resulting in a 69.5% reduction in levelized cost of water compared to a nominal design. We demonstrate that multidisciplinary design optimization outperforms sequential design approaches, yielding lower levelized costs of water and substantially different optimal designs. The multidisciplinary design optimization results suggest major design changes compared to designs found in the literature. Notably, smaller wave energy converters and larger pistons, along with smaller accumulators and larger seawater reverse osmosis plant installations, are preferred. These design trends are consistent across a range of sea states, suggesting potential generalizability beyond a single location. This study demonstrates the importance of holistic modeling and co-design for wave-driven desalination systems and establishes an effective optimization framework for future studies to build upon.

Graph neural network architectures are advantageous for predicting core-electron binding energies which depend on local bond environment effects, as the number of message passing layers defines the topological (bond) radius of the model's receptive field. This provides an interpretable connection between the model's architecture and the definition of locality in the considered environment. Here we present a graph neural network model for predicting carbon 1s core-electron binding energies in organic molecules. The model is trained with multiconfiguration pair-density functional theory on 8637 carbon atoms in 2116 molecules with 4-16 atoms and evaluated against 570 experimental values in 113 different molecules containing 3-45 atoms. Previous work benchmarked a mean absolute error of 0.27 eV to experiment for the training data level of theory [J. Phys. Chem. A 2025, 129, 36, 8419-8431] and the present model demonstrates an experimental evaluation error of 0.33 eV with good size transferability to larger systems. By examining the effect of the number of message passing layers on the performance, we show that two chemically informed node features, the atomic binding energy and environment electronegativity, encode molecule-specific information when normalized across the graph and capture beyond nearest-neighbor environment effects outside the receptive field. A case study on the 45 atom avobenzone tautomers demonstrates the model's ability for instant and precise analysis of complex molecules. Finally, the model's E(3)-equivariance is shown to out-perform an invariant model on non-equilibrium geometries from a methanol C-O bond stretch. The software and data are provided by the open-source AugerNet package at https://doi.org/10.5281/zenodo.19689244.

Cell invasion and spatial pattern formation are two distinct manifestations of cellular self-organisation in development, regeneration, and disease. Here, we develop and analyse a unified theoretical framework that links these two seemingly different behaviours within a single mechanistic model for adhesion-mediated self-organisation in growing cell populations. Using a multiscale analysis, we show that the balance between cell-cell adhesion, self-diffusion, and proliferation controls the emergence of distinct collective dynamics. We find that for weak adhesion, tissues invade through stable monotone fronts. As adhesion increases, invasion slows, fronts become unstable, leading to aggregates and spatial patterns emerging behind the advancing edge. In two spatial dimensions, these instabilities generate fingering morphologies reminiscent of dysregulated invasion in cancer. Crucially, we show that density-dependent regulation of adhesion suppresses these instabilities and restores cohesive tissue expansion. Together, our results identify adhesion strength and its regulation as key determinants of whether tissues invade cohesively or fragment into patterns, and provide a unified framework for understanding collective migration, morphogenesis, and dysregulated growth.

Ghost imaging uses two light beams correlated in the transverse position, time, or frequency to create an image of a spatial, temporal, or spectral object. We propose a scheme of time-to-space ghost imaging for creating a spatial image of a temporal object, enabled by two spatio-temporally correlated light beams. Assuming a spatio-temporal Gaussian Schell model for the description of the source, we obtain analytical expressions for the point-spread function of the system and its temporal resolution. We show how the required source of partially coherent light can be realized by a combination of a diffraction grating and a spatial light modulator. As follows from our analysis, the temporal resolution of a time-to-space imaging system is determined by the duration of the laser pulses used and the transverse coherence length imposed by the spatial light modulator, does not depend on the resolution time of the photodetectors, and can reach the sub-picosecond range.

This study presents a versatile ontology and a useful codification scheme for describing all kinds of engineered living materials (ELMs). The different components of the ontology, namely: families according to the taxonomy for ELMs, industrial applications and synthesis or processing methods, are systematically organized, enumerated, classified, codified and explained. The methodic application of the ontology to a set of 100 relevant examples of ELMs helps to demonstrate its utility and adaptability to many different types of ELMs with a wide range of industrial applications and obtained through numerous synthesis and processing methods. This proves that the developed ontology and codification schemes, with the glossary provided to support its implementation and application, can serve as a comprehensive classification tool for the emergent field of ELMs. Furthermore, the usability of the ELMs ontology and codification by a generative artificial intelligence (AI) is explored and validated by different means, checking that both natural language and the codification are understandable for describing ELMs, verifying that the generative AI adequately codifies examples of ELMs according to the ontology, and validating the synergic applicability of the ontology and codification with generative AI tools for illustrating novel ELMs and supporting their conceptual design. This study is expected to provide a universal language to facilitate communication in the ELMs field and to foster the discovery of new ELMs and related innovations, hoping it may accelerate scientific and technological discoveries.

In pedestrian dynamics, the internal drive that propels individuals toward their goals is typically captured by a single, fixed parameter, the desired walking speed. This simplification overlooks that motivation fluctuates in response to changing spatial and social conditions within a crowd. This paper proposes a dynamic motivation model grounded in expectancy-value theory from psychology, in which each agent's motivation evolves over time depending on proximity to the goal, relative position among other pedestrians, and individual goal importance. The resulting motivation modulates multiple movement parameters simultaneously, including walking speed, gap-closing behavior, and interpersonal spacing. The model is evaluated in simulated pre-bottleneck waiting scenarios using paired statistical comparisons across multiple random seeds and population sizes, and compared with trajectory data from the CROMA concert-entry bottleneck experiments under low- and high-motivation framings. Simulations show that the dynamic model produces structured heterogeneity in the crowd: agents self-organize into differentiated positions near the bottleneck, with those closer to the front occupying less space, a pattern absent in the static baseline but clearly present in the experimental data. These findings suggest that motivation in crowds should be understood not as a uniform increase in urgency, but as a mechanism that reorganizes competitive positioning along spatial and social axes. Future work should extend the framework to open-door throughput scenarios, larger populations, and richer social interactions such as group cohesion and cooperative strategies.

This paper investigates the influence of interfaces on the performance of finite-sized mechanical metamaterial structures for vibration damping applications. The metamaterial structures are designed in a sandwich configuration in which two homogeneous plates are connected to a metamaterial array. We test four different arrays that are obtained from the same metamaterial by differently cutting the metamaterial's unit cell at the metamaterial/plate interface. When the four unit cells are periodically repeated in space, they create the same infinitely large metamaterial with an identical mechanical response. In finite-sized structures, however, the different interfaces between the metamaterial array and the plates~--~called ``material interfaces''~--~and between the metamaterial and the air~--~called ``free interfaces''~--~strongly affect the specimen's vibration transmission characteristics. Using experimental measurements and validated finite-element (FE) models, we demonstrate a significant influence of the different types of interfaces on the global responses and local displacement fields of the structures. We also demonstrate the presence of a vibroacoustic coupling in the structures which also depends on the type of metamaterial/plate interfaces. Furthermore, we explore optimization strategies for enhancing the vibration damping performance of the metamaterial structures considering not only the metamaterial array but also the adjacent structures (the homogeneous plates). A comparison with benchmark cases illustrates the optimization potential that the interfaces' design offers for the vibration damping capability of finite-sized metamaterial structures. We show that optimizing the type of targeted interfaces can shift a metamaterial's response from underperforming to significantly outperforming compared to classical solutions for noise and vibration damping in civil engineering.

The coupling between light and strain gradients shows two kinds of effects: light enhanced flexoelectricity (photoflexoelectricity) and gradient enhanced photovoltage (flexophotovoltage). Although these effects originate from fundamentally different physical mechanisms (one is light enhanced electromechanical coupling, the other is a bulk photovoltaic effect), in this article we show that dynamic flexoelectric measurements of semiconductors under illumination intrinsically contain contributions from both. To allow disentangling them, we have developed a general theoretical framework for their combined response in oscillating systems, demonstrating that the two contributions can be unambiguously separated through their distinct frequency and phase dependencies. We have validated these predictions using oscillating cantilever measurements on centrosymmetric perovskite semiconductors (SrTiO3 and methylammonium lead bromide, MAPbBr3), obtaining selfconsistent values for the coefficients both effects which are in excellent agreement with independent static measurements. Our results establish a general protocol for disentangling both light strain gradient couplings using only oscillatory measurements, and clarify the interpretation of flexoelectric measurements under illumination.

Laser annealing offers a promising route to back end of the line fabrication of ferroelectric thin film transistors based on hafnium-zirconium oxide (HZO). Due to the wide band gap of this material, previous reports have studied the crystallization of HZO using ultraviolet or infrared light. In contrast, we monitor its crystallization in a Si3N4/TiN/Hf0.5Zr0.5O2 thin film heterostructure upon irradiation with visible nanosecond laser pulses. This geometry mimics the structure of CMOS devices and harnesses the absorption of TiN in the visible regime to generate the heat necessary for the transformation. Through a series of local in situ measurements using a modified transmission electron microscope, we quantify the relationship between the HZO film thickness, critical laser energy density and the ferroelectric HZO phase fraction, finding a sharp threshold behavior in the laser pulse energy necessary to crystallize HZO. The optimal condition of irradiating an 8-nm HZO film with a single laser pulse with an energy density of 177 mJ/cm2 is found to produce 86% of the ferroelectric orthorhombic phase. Heat transfer dynamics within the heterostructure during laser annealing are revealed by finite element simulations, where the partial melting of the silicon nitride substrate is found to play an important role limiting the temperature to 1900 °C. This finding as well as the observed laser pulse energy threshold behavior support a kinetic crystallization pathway involving the tetragonal phase. More generally, these findings show how laser-driven phase engineering can lead to scalable design and enhanced performance of ferroelectric materials in advanced electronic applications.

Surface confined electronic states provide a fertile ground for discovering emergent phenomena that have no counterpart in the bulk, offering new routes to manipulate correlations, symmetry breaking, and dimensionality at the atomic scale. Here, we show that charge density wave (CDW) symmetry breaking can yield a surface states in 1T-TiSe2. Micro angle resolved photoemission spectroscopy resolves a sharp, two dimensional surface resonant state (SRS) that emerges within the CDW reconstructed low energy spectrum. The SRS exhibits notable temperature dependence and its spectral weight collapses around 160 K, while CDW transition temperature TCDW is commonly reported as 202 K. Slab DFT+U calculations reproduce a surface localized resonance when CDW folding brings valence and conduction states into near degeneracy, suggesting a correlation tuned, surface selective origin. These results point to a form of correlation-tuned surface resonance in a layered CDW compound and suggest a framework for engineering low dimensional quantum states in van der Waals materials via symmetry breaking and electronic structure tuning.

We consider a two-dimensional wave system containing a subwavelength hole, such as an aperture in an interface supporting surface electromagnetic or acoustic waves, or an island in a fluid surface sustaining gravity-capillary waves. Recent studies have revealed the emergence of pronounced wave vortices around such structures, termed type-II vortices, in contrast to conventional (type-I) vortices associated with phase singularities and intensity nulls. A striking natural manifestation of type-II vortices occurs in ocean tides around islands such as New Zealand, Madagascar, and Iceland, where the tidal phase increases by $\pm 2π$ around the island. Although this phenomenon is usually associated with the Coriolis effect from the rotation of the Earth, here we demonstrate the controlled generation of type-II vortices using a minimal and tunable setup: a dipole-oscillating subwavelength hole and a single incident plane wave. Using laboratory gravity-capillary waves and an oscillating subwavelength `island', we directly measure the resulting phase structure, topological charge, and wave angular momentum. We show that the emergence and handedness of the vortices can be precisely controlled via the relative phase between the dipolar source and the incident wave. Our results offer a simple and versatile mechanism for engineering subwavelength wave vortices, with potential applications in a variety of two-dimensional wave systems.

Traditional iterative reconstruction methods are accurate but computationally expensive, limiting their use in high-throughput and real-time ptychography. Recent deep learning approaches improve speed, but often predict phase as a Euclidean scalar despite its $2π$ periodicity, which can introduce wrapping artifacts, discontinuities at $\pmπ$, and a mismatch between the loss and the underlying signal geometry. We present a deep learning framework for ptychographic reconstruction that models phase on the unit circle using cosine and sine components. Phase error is optimized with a differentiable geodesic loss, which avoids branch-cut discontinuities and provides bounded gradients. The network further incorporates saturation-aware dual-gain input scaling, parallel encoder branches, and three decoders for amplitude, cosine, and sine prediction, together with a composite loss that promotes circular consistency and structural fidelity. Experiments on synthetic and experimental datasets show consistent improvements in both amplitude and phase reconstruction over existing deep learning methods. Frequency-domain analysis further shows better preservation of mid- and high-frequency phase content. The proposed method also provides substantial speedup over iterative solvers while maintaining physically consistent reconstructions.

The rise of machine learning and additive manufacturing has enabled the design of architected materials with tailored properties that surpass those of natural materials. Inverse design offers a data-efficient alternative to trial-and-error methods, yet most existing approaches depend on either large datasets or scarce high-fidelity data from simulations and experiments. These requirements pose a particular challenge for architected materials with nonlinear mechanical responses, where capturing complex deformation modes requires expensive evaluations. To address this, a Multi-Fidelity Bayesian Optimisation (MFBO) framework for the inverse design of cellular composites that directly targets their full nonlinear response is introduced. By integrating information from multiple fidelity sources and scalarising the response using a similarity score, the framework enables efficient exploration of the design space while reducing reliance on costly evaluations. As a proof of concept, the method is applied to spinodoid cellular composites using finite element models, validated with compression tests on short carbon-fibre reinforced PET-G composites. Four target responses were considered, with three multi-fidelity strategies benchmarked against a standard single-fidelity approach. Across all cases, MFBO achieved higher similarity scores and consistently recovered the targeted responses, outperforming the single-fidelity baseline under the same evaluation budget, while also successfully recovering all targeted responses. These results demonstrate the effectiveness of MFBO for inverse design of stochastic architected materials, where high-quality data is scarce but lower-cost proxies exist. By efficiently navigating complex design spaces, MFBO enables the creation of cellular composites with precisely tailored nonlinear mechanical behaviour.

Synthetic complex-frequency excitations have emerged as a powerful tool for loss compensation and resolution enhancement. We show that, ideally, these excitations allow for the complete offsetting of intrinsic damping over long evolution times, governed by a universal inverse-time scaling law for residual damping under Nth-order synthetic illumination. However, in realistic experimental settings, the achievable virtual gain is fundamentally restricted by the finite spectral measurement range, which introduces unwanted temporal artifacts and disrupts this ideal scaling. We demonstrate that the conventional rectangular spectral window creates a slowly decaying temporal kernel (1/t) that leaks unwanted early-time signals into the late-time regime, thereby masking the targeted response. To mitigate this constraint, we introduce a Hann-window filtering technique that yields a faster decaying temporal kernel (1/t)^3. This simple spectral engineering dramatically suppresses spurious contributions and extends the usable lifetime of the synthetic waveform. Experimental validation using coupled plasmonic resonators demonstrates that Hann-window filtering improves the loss-offsetting efficiency by nearly a factor of three compared with the standard rectangular window. Our results reveal the fundamental temporal limits of synthetic complex-frequency waves and provide a practical strategy to achieve long-lived, high-SNR loss compensation in nanophotonic systems.

Topology describes properties of physical systems that remain constant under continuous deformations. For infinite vector waves, global topological invariants in position space are typically associated with periodic patterns. We demonstrate that even for aperiodic Helmholtz-decomposable wave fields, possessing only the wave's intrinsic periodicity, a topological invariant can be found in momentum space. This invariant, the linking number, represents a Berry phase. By utilizing electromagnetic and hydrodynamic surface waves, we confirm the robustness of the linking number against deformations, and experimentally observe discrete transitions between distinct topological sectors. The linking number captures the topology of vector wave fields across both continuous and discrete momentum spaces. Our work introduces a unified topological framework for vector wave fields, enabling their classification via a global invariant.

At laser wavelengths longer than the prominent 1 μm range of high-power ytterbium-doped lasers, nonlinear phase shifts produced in nonlinear media for spectral broadening and subsequent pulse compression decrease drastically. Consequently, at the 2 μm wavelength range, the threshold of the applicable peak power for pulse compression in gas-filled multipass cells increases. The common approach of choosing a Herriott multipass cell configuration close to the concentric resonator does not necessarily lead to the highest total nonlinear phase shift, due to a restriction of the total number of reflections on the cell mirrors of a given size. Therefore, an analytical approach is presented here to maximize the nonlinear phase shift for a given set of mirrors, considering lossless and dispersionless propagation. Furthermore, to achieve pulse compression with gas-filled multipass cells for relatively low peak powers at wavelengths around 2 μm, we developed a high-pressure gas cell and demonstrated experimentally pulse compression in the negative- and positive dispersion regimes, with achieved pulse durations of around 40 fs and 55 fs respectively.

Higher energy density and longer lifetime are the requirements for next-generation lithium-ion batteries. A promising anode material is silicon, which offers high specific capacity, but its significant volume change during lithiation and delithiation enormously reduces battery lifetime. A physical understanding of the processes degrading the battery is key to mitigate this effect and advance in the field. We develop a physics-based model to describe degradation during battery cycling under various protocols and storage conditions, with varying check-up (CU) frequencies. The model can disentangle basic degradation mechanisms, such as the growth of the Solid-Electrolyte Interphase (SEI), from silicon mechanisms, such as particle cracking, SEI growth on cracks, and loss of active material (LAM). We investigate the impact of CUs on the observed storage degradation and the reason behind the increased degradation in batteries, including silicon in the anode. Additionally, we relate the observed degradation to operating conditions, enabling future optimization of battery use and design.

Lattice metamaterials enable lightweight, multifunctional structures, yet homogenization-based evaluation of their effective properties remains computationally expensive. Neural surrogates offer speed but often lack the accuracy and stability required for engineering-grade simulations. We introduce GMT, a Geometric Multigrid Transformer -- a neural solver with high numerical fidelity for fast and reliable lattice homogenization. GMT achieves architectural alignment with Geometric Multigrid (GMG) by restructuring Point Transformer V3 to operate across sparse GMG hierarchies, capturing long-range dependencies and cross-level interactions essential for multigrid convergence. To enforce physical consistency, GMT incorporates physics-aware positional encoding for strict enforcement of periodicity and predicts both the finest-level solution and multi-level residual corrections. These predictions deliver a spectrally-aligned initialization, enabling end-to-end training under physics-informed and solver-aware losses and requiring only a single GMG V-cycle refinement to reach convergence. This fusion of neural prediction and numerical rigor achieves relative residual errors of $10^{-5}$ with a $160\times$ speedup over state-of-the-art GPU-based solvers at equivalent accuracy -- particularly at high resolutions (e.g. $512^3$), where traditional methods become most costly. We validate GMT across mechanical and thermal domains, demonstrate robust generalization to unseen geometries and non-periodic settings, and showcase scalability to high resolutions -- enabling real-time design iteration, multi-scale simulations, high-throughput material discovery, and inverse design.

Dual-comb spectroscopy enables broadband, high-resolution measurements with microsecond temporal resolution, but extending this capability to the 10~12.5 $μ$m molecular fingerprint region remains technically challenging, particularly for transient radical kinetics. Here, we demonstrate microsecond-resolved dual-comb spectroscopy in this spectral range using electro-optic combs and difference-frequency generation in an orientation-patterned gallium phosphide crystal. Operation near a turning-point quasi-phase-matching condition at approximately 140 $^\circ$C reduces the wavelength sensitivity of the nonlinear conversion, enabling robust tuning of the idler comb over 83 cm$^{-1}$, corresponding to approximately 1.2 $μ$m near 12 $μ$m, by adjusting only the signal-comb center wavelength while keeping the pump wavelength and crystal temperature fixed. As a demonstration, we perform high-resolution, microsecond-resolved spectroscopy of transient chlorine monoxide (ClO) near 12 $μ$m. Time-resolved dual-comb spectra capture the temporal evolution of ClO produced by the Cl + O$_3$ reaction with a temporal resolution of 1.5 $μ$s, enabling quantitative determination of the ClO formation rate coefficient. These results establish this dual-comb platform as a promising tool for quantitative, microsecond-resolved studies of short-lived radicals, particularly atmospherically relevant halogen oxides.

Optical anisotropy in crystalline solids enables direction-dependent light-matter interactions and underpins a variety of advanced photonic functionalities. In this context, Vanadium dioxide (VO2) represents a prototypical material that undergoes a reversible MIT near 67°C, accompanied by pronounced electronic, structural, and optical modifications. The MIT not only dramatically modifies the VO2 electrical conductivity but also reshapes its anisotropic optical response, making VO2 an exceptional platform for dynamically tunable photonic and optoelectronic devices. In this work, we investigate how the intrinsic crystalline anisotropy of VO2 induces a hyperbolic optical behavior in the metallic rutile phase. We study two epitaxial VO2 thin films of different thicknesses grown on (110) oriented MgF2 substrates. Broadband polarized spectroscopic measurements, spanning the infrared to UV spectral range, are employed to independently investigate the optical response in both the monoclinic and rutile phases. From these measurements, we extract the optical conductivity and the dielectric function, revealing a pronounced anisotropy in the rutile metallic phase, with an enhanced free-carrier response along the rutile c axis. Our data show that, within a narrow near-infrared spectral window, the real parts of the dielectric tensor components along the two principal axes acquire opposite signs, indicating the emergence of a hyperbolic type-II dispersion. The hyperbolic response is quantitatively evaluated through the quality factor and the degree of dielectric anisotropy, enabling a systematic assessment of VO2 as a thermally switchable, hyperbolic optical medium. These findings expand the understanding of anisotropy-driven optical phenomena in phase-change materials and highlight VO2 thin films as a promising platform for tunable and reconfigurable photonic applications.

We propose Geometric-Configuration Modulation (GM), a novel AO-free FSO paradigm utilizing multi-source geometric configuration encoding and active correlative decoding. GM demonstrates exceptional resistance to strong atmospheric turbulence ($D/r_{0}\sim 5$) over a 1.2 m link in preliminary experiments.

Conservation laws are conventionally discretized through floating-point flux evaluation, with invariants obtained by cancellation of approximate interface contributions and admissible weak solutions selected by reconstruction and Riemann solvers. Here we introduce an operator-level formulation in which conservative dynamics is realized as an exact discrete interaction rule on a quantized state space. The update is defined by an antisymmetric integer-transfer operator, which enforces conservation exactly at the arithmetic level and eliminates round-off drift from the primitive evolution \cite{highamAccuracyStabilityNumerical2002}. For scalar laws, monotone order-preserving transfers select admissible shock structures within the primitive update, rather than through flux reconstruction. Numerical experiments show that the interaction rule preserves high-frequency transport near the Nyquist limit and maintains sharply localized discontinuities in Burgers dynamics. The same construction extends to multidimensional problems and systems of conservation laws through oriented, vector-valued integer transfers. These results indicate that conservative dynamics admits an exact discrete realization in which both invariance and entropy selection are encoded at the operator level, rather than arising from approximate flux cancellation.

Challenge-led R and D programs increasingly assemble heterogeneous people, organizations, funders, projects, and technical outputs around defined missions. Yet program evaluation often describes these systems through project lists, output counts, or retrospective case narratives. This article develops a typed network framework for representing R and D program architecture directly. We model programs as networks of people, places, and things: researchers, program directors, institutions, funders, publications, patents, projects, and citations. Applied to ARPA-E project impact sheets from the agency's first decade, the framework reconstructs 23 program-induced networks and an agency-level composed network. We show that R and D programs have an analysable topology: a typed arrangement of people, institutions, funders, projects, publications, patents, and citations that can be reconstructed, compared, and monitored. The analysis shows that programs can be compared by their local structural patterns, that cross-program overlap is concentrated more in recurring institutions than in individual researchers, and that program fingerprints differ across thematic areas. The article contributes to network science by extending topological analysis to R and D program systems, a class of governed, typed, and output-generating networks that has not been systematically represented in existing innovation-network work.

Supersonic turbulence is vital to astrophysical and high-speed engineering flows, yet its energy transfer mechanisms remain poorly understood. We present high-resolution ($1024^3$) direct numerical simulations (DNS) of forced compressible turbulence across a range of turbulent Mach numbers ($M_t = 0.2$ to $3.0$). Using the GPU-accelerated solver \texttt{DHARA} with a seventh-order, low-dissipation Targeted Essentially Non-Oscillatory (TENO) scheme, we resolve both fine-scale eddies and sharp shock fronts. Our results reveal a fundamental shift in the energy cascade in the supersonic regime. As $M_t$ increases, the rotational kinetic energy spectrum steepens from a Kolmogorov-like $k^{-5/3}$ scaling toward a Burgers-like $k^{-2}$ scaling. Conversely, the compressive energy spectrum becomes shallower, deviating from Burgers scaling. We show that these spectral modifications are driven by a dominant cross-scale transfer of energy from solenoidal to compressive modes within the inertial range, alongside significant contributions from pressure dilatation. Scaling laws for the root-mean-square compressive velocity ($U_C$) and compressive energy flux ($Π_C$) are found to mirror classical Burgers turbulence. Finally, we show that while energy injection rates depend on forcing type rather than Mach number, increased $M_t$ leads to decreased rotational dissipation and increased compressive dissipation and pressure dilatation. These findings elucidate intermodal energy cascade mechanisms, advancing our understanding of energy transfers in supersonic turbulence.

Understanding how human mobility and information propagation influence the course of an epidemic remains a key challenge in digital epidemiology. In this work, we develop a new awareness-based, event-triggered epidemic model embedded within a networked Cyber-Physical System (CPS). In our framework, disease transmission and the dissemination of epidemic-related information evolve together on two interconnected layers. In detail, the physical layer models disease spread through human movement between two types of locations - residences and transfer stations - forming a bipartite metapopulation network. This structure captures the rendezvous effect, which reflects how gatherings in shared locations contribute to infection spread. The cyber layer represents the flow of information through digital communication networks. We introduce an event-triggered migration regulation mechanism, whereby individuals adapt their movement patterns based on local awareness thresholds, leading to a decentralized control process embedded within the network. Using a microscopic Markov chain approach (MMCA), we derive the epidemic threshold analytically and validate our results through extensive Monte Carlo simulations. Our findings show that event-triggered migration effectively suppresses the overall spread of the disease and lowers infection peaks - especially in heterogeneous populations and densely connected gathering points. These results demonstrate the potential of CPS-based epidemic models to enable real-time, awareness-driven interventions and to inform the design of decentralized control strategies that leverage digital communication dynamics.

The current thermo-industrial civilization is critically dependent on fossil fuel energy sources. An intuitive model capturing the interplay between economic activity, physical power consumption, depletion and energy quality is presented.

Random lasing exploits multiple scattering to provide optical feedback without conventional resonant cavities, enabling simplified architectures that are readily integrated into compact photonic platforms such as wearable sensors and lab-on-chip devices. However, the same disorder that enables cavity-free lasing also makes it challenging to control and tune the emission properties. Here, an electrically reconfigurable random-lasing platform based on dielectrophoretic assembly of chaining silver nanowires suspended in a dye gain medium is reported. An applied electric field across patterned quadrupole electrodes induces nanowire chaining and programmable alignment, enabling real-time reconfiguration of the disorder landscape. Based on the electrically driven disorder state transitions, tunable random-lasing characteristics, including reduced lasing thresholds, modulation of emission intensity, and control of the polarization state have been demonstrated. Simulations further indicate that chaining enhances scattering relative to absorption, providing more efficient radiative feedback, and the orientation of the nanowire network governs the polarization dependence of the system. These results establish a route to actively modulate random lasing through controllable disorder and point toward adaptive, reconfigurable photonic light sources and sensing systems.

Music genre classification shapes how listeners discover music, how platforms design recommendations, and how sociologists study cultural taste. Yet existing genre labels are inconsistent in granularity: they exaggerate boundaries between overlapping categories and hide sociologically important heterogeneity within broad labels. Cultural sociologists have long theorized that genres vary along two independent dimensions, boundary strength and internal differentiation, but existing empirical work has relied on fixed label sets, leaving these dimensions without quantitative operationalization from actual consumption behavior data. Here we propose a two-dimensional framework that extracts music communities bottom-up from playlist co-occurrence networks and characterizes each along two axes: external closure $B(C)$, measuring boundary strength relative to a random null, and internal differentiation $D(C)$, measuring organized internal subdivision. We validate the framework on two independent datasets across platforms, cultural contexts, and time periods, confirming that $B(C)$ and $D(C)$ are statistically independent and that each captures a distinct structural property. The framework reveals genre structures invisible to fixed labels: single labels splitting into communities with different boundary strengths, multiple labels merging into tightly bounded communities, and consumption spheres that no existing label describes. Comparison with prior theoretical predictions is broadly consistent, with the notable exception that Hip-Hop exhibits rich internal differentiation across both datasets, challenging its prevailing single-centered characterization. By providing a label-independent coordinate system grounded in listener behavior, this framework opens a path toward tracking how genre boundaries and internal structures evolve over time, a question that static label systems cannot address.

We present an efficient and accurate implementation of hybrid exchange-correlation (XC) functionals in the SIESTA code, enabling large-scale simulations based on Hartree-Fock-type exact exchange combined with strictly localized numerical atomic orbitals (NAOs). Our approach exploits a fitted representation of the NAOs in terms of Gaussian-type orbitals (GTOs), which allows for the analytical evaluation of four-center electron repulsion integrals (ERIs) via the LIBINT library. This framework is seamlessly integrated with SIESTA's real-space grid and sparse-matrix infrastructure, and is combined with multiple screening techniques to control the computational complexity. We also introduce a fully analytical formulation of hybrid-functional forces and a dynamic parallel distribution scheme that ensures excellent scalability. We validate our implementation through benchmark calculations on a broad set of systems (including semiconductors, insulators, and two-dimensional materials) and demonstrate that the HSE06 functional significantly improves the prediction of band gaps compared to PBE, in close agreement with G0W0 and experimental data. We analyze in detail the trade-offs between accuracy and computational efficiency as a function of the number of Gaussians, basis set range, and integral screening thresholds. Our results confirm that hybrid functional calculations in SIESTA are now feasible for large extended systems, making accurate first-principles predictions of electronic and structural properties accessible at scale.

Accurate quantification of protein-nanoparticle interactions is essential for applications in nanobiotechnology, nanomedicine, and drug delivery. Motivated by recent computational and experimental work, we combine coarse-grained united-atom (UA) models with molecular docking to characterize protein adsorption on SiO_2 nanoparticles. We construct orientation-resolved heatmaps in which polar and azimuthal angles uniquely specify the relative protein-nanoparticle pose, and the map amplitude reports binding propensity via the minimum UA adsorption energy or the docking score. Each angular bin corresponds to a distinct docked complex, enabling systematic comparison of binding geometries across models. To relate docking score landscapes to Boltzmann-averaged UA adsorption energetics, we analyze eight birch pollen allergen proteins previously studied experimentally. Similarity between the two orientational distributions is quantified using the Jensen-Shannon divergence (JSD). We find encouraging agreement between the two approaches in several cases, while also identifying limitations and routes for improvement, including optimized angular resolution and iterative refinement of interaction parameters. Overall, this framework provides a quantitative bridge between coarse-grained energetics and docking outputs at protein-nanoparticle interfaces, supporting improved predictive modeling and mechanistic insight into protein-nanoparticle binding landscapes.

Accurate large-scale Kohn-Sham density functional theory (DFT) calculations are essential for modeling complex material systems, including interfaces, defects, nanoclusters, and twisted two-dimensional heterostructures. Achieving chemical accuracy at scales of $10^4$-$10^5$ electrons with practical time-to-solution, however, remains challenging for existing DFT implementations. We present GPU-centric computational methods and algorithmic innovations within a finite-element (FE) discretized projector augmented-wave (PAW) formulation (PAW-FE) for accurate, efficient, and scalable electronic-structure calculations on modern exascale systems. The FE discretization, developed within a collinear spin formalism, accommodates generic boundary conditions and employs multi-resolution quadrature for accurate evaluation of atom-centered PAW integrals on coarse grids. The resulting generalized Hermitian eigenproblem is solved using residual-based Chebyshev filtered subspace iteration (R-ChFSI). Exploiting R-ChFSI's tolerance to inexact matrix-multivector products, we employ an approximate inverse PAW overlap matrix, mixed-precision arithmetic (FP32/TF32), and low-precision nearest-neighbor communication (BF16) during filtered subspace construction, along with block-wise computation-communication overlap to reduce cost while preserving robustness. These strategies yield up to $8\times$ and $20\times$ CPU-GPU speedups on Intel and AMD GPU architectures, respectively. Compared to plane-wave PAW methods, PAW-FE achieves close to 8$\times$ reduction in time-to-solution for 10,000-electron systems on NVIDIA GPUs, with larger gains at scale, and around 6$\times$ over norm-conserving FE approaches. We demonstrate scalability to 130,000-electron systems, establishing PAW-FE as an exascale-ready method for chemically accurate first-principles simulations.

Metasurfaces have revolutionized compact wavefront control using planar, subwavelength structures. However, conventional meta-optical devices predominantly operate within a far-field paradigm, assuming electromagnetic decoupling between the source and metasurface, which limits control to post-emission wavefront shaping. Here, we define and experimentally demonstrate near-field meta-optics - a regime where strong source - structure coupling enables simultaneous control of emission and radiation. By integrating an inverse-designed dielectric metasurface directly within the near field of a terahertz photoconductive antenna (PCA), we show that the metasurface co-defines the emission process itself. Our meta-PCA, incorporating a 50-um-high metasurface - one-third the thickness required by reciprocal far-field designs - collapses emission from ~60deg divergence to a sharp < 10deg forward beam, while enhancing on-axis intensity 50-fold compared to bare GaAs. Unlike far-field metasurfaces that typically trade efficiency for thinness while remaining laterally large, our device achieves extreme compactness in both dimensions. Remarkably, it exceeds the outcoupling efficiency of a bulky, millimeter-scale silicon lens by 10%, despite a volume reduction of over three orders of magnitude. These results establish near-field meta-optics as a transformative paradigm for developing high-efficiency, ultra-compact on-chip photonic systems across the electromagnetic spectrum.

This paper presents the optimal-control suggestion for congestion on freeways using data assimilation (DA) of distributed fiber-optic sensing (DFOS). To simultaneously maximize throughput and avoid/mitigate congestion, it is necessary to execute optimal control for the current traffic state as active transportation and demand management (ATDM) according to multi-objective optimization with real-time monitoring data. However, optimal control cannot be estimated due to intermittent observed data obtained from conventional sensors. To solve the issue, this paper proposes the ATDM optimal control estimation with DA of DFOS, which can monitor traffic flow in real time without dead zones. Our real-time DA method enables us to estimate the effectiveness of control scenarios by simulation. This paper also provides a method to uniquely determine the optimal-control solution among the Pareto solutions for multi-objective optimization. Throughput and mean speed across the entire road are considered as the objective functions. Variable speed limit (VSL) and inflow control are taken as ATDM examples. Validation results on a Japanese freeway show that (i) the optimal control scenario varies depending on the traffic state, especially congestion level; (ii) optimal control considering VSL alone improves throughput by 5-14% while the improvement rate for mean speed is 0-8%; (iii) throughput and mean speed are improved by 10-15% and 20-30%, respectively when VSL and inflow control are considered. This paper also implies the importance of balance management for the lane occupancy and proactive optimal control before congestion occurs.

Antenna miniaturization remains a critical technological challenge across frequency scales - from microwave RF links in phones and wearables to VLF for underwater-to-air communications and ionospheric probing. At deeply subwavelength scales conventional antennas require complex and lossy matching circuits due to absent intrinsic material resonances, motivating resonant electrically small emitters (ESEs) like mechanical resonators and quantum emitters. Here, we extend the theory of electrically small antennas (ESAs) to this broader ESE class, deriving the fundamental efficiency limit for a unit volume emitter at given frequency and bandwidth. Our figure of merit (FOM) - quantifying proximity to this limit - enables direct comparison across ESE types, frequencies, bandwidths and scales. We demonstrate its utility using public data from ELF and VLF Navy facilities alongside two mechanical ESEs reported in literature. The measurements reveal that mechanical antennas operate near theoretical FOM limit, questioning claims of possible further orders-of-magnitude gains. A naturally resonant emitter is still subject to the Chu-Harrington limit (CHL) under its standard assumptions. Indeed, we derive novel CHL-dictated constraints on atomic ESE properties: lower bound on excited-state lifetime and an upper bound on transition dipole moment.

We present the first quantum-hardware implementation of a Hamiltonian simulation algorithm that produces signed vector-field solutions to the time-domain Maxwells equations using a Schrodingerisation-based approach. The electromagnetic fields are discretized using finite-difference operators, and the resulting non-unitary matrices are mapped to Bell-basis Trotter blocks, enabling efficient circuit construction. We introduce a measurement procedure that retrieves not only field amplitudes, but also physical directions of the electric and magnetic field values at select spatial points. Implementing this logic on quantum hardware relies on relative-phase-based sign reconstruction. Numerical results obtained using IonQ QPU, show good agreement with analytical solutions of benchmark problems in two dimensions and on simulators; in three dimensions. We further extend our approach to compute fields scattered from simple bodies, by enforcing appropriate boundary conditions. Our work lays the foundational steps towards realizing quantum-hardware solutions for computational electromagnetics.

The reliability of machine learning in multiscale physical systems depends on how physical structure is embedded into the learning process. We investigate this in the context of turbulent multiphase flows, focusing on the prediction of subgrid interfacial area density, a key quantity governing interphase transport that remains unresolved in large-eddy simulations. In this work, we develop and evaluate two machine learning subgrid closure models to predict the three-dimensional subgrid interfacial area density: a purely data-driven 3D encoder-decoder network, and a physics-constrained variant regularized by a fractal geometric prior. Across a range of Weber numbers, the physics-based model improves predictive accuracy, reduces error variance, and suppresses nonphysical artifacts relative to purely data-driven approaches. We also show that these gains are regime-dependent: the embedded inductive bias enhances generalization in corrugation-dominated regimes where its underlying assumptions hold, but becomes ineffective in fragmentation-dominated regimes characterized by topology change and droplet breakup. These results reveal a broader principle for scientific machine learning: the utility of physics-informed models depends not only on the presence of inductive bias, but on its alignment with the governing physical regime. This suggests a path toward regime-aware learning frameworks for modeling of complex multiscale systems.