Papers

Fresh from arXiv, typeset properly. Highlighted words trade on the market.

Computer Science

cs Watch Computer Science

Sunday 30 August

Networking

RadioSight: Predictive mmWave XR Network Optimization from Dynamic Neural Radio Fields

Lihao Zhang, Paul Kudyba, Zhenlin An et al. (+1 more)

Next-generation extended reality (XR) networks rely on mmWave communication for multi-gigabit throughput, yet highly directional links are vulnerable to user mobility and blockages, causing frequent outages under reactive beam management. Emerging neural radio fields can predict radio propagation, but prior work remains limited to offline channel reconstruction. We introduce RadioSight, a real-time multi-modal radio field system for predictive mmWave optimization and proactive Multi-User MIMO beamforming. RadioSight combines backward beam-tracing with real-time semantic object synchronization to anticipate RF geometry changes without full model retraining. Implemented as an edge-executable pipeline for commercial 28 GHz arrays, RadioSight determines each scheduling window's beams during the preceding window without exhaustive beam sweeps. Experiments show that RadioSight reduces beam-search error by up to ~50%, improves median throughput by 2x, and enhances link stability.

Read the full abstract

All authors Lihao Zhang, Paul Kudyba, Zhenlin An, Haijian Sun

Machine Learning

Adversarial Online Classification with a Preview

Roi Livni and Sahil Singla

Worst-case online classification is governed by sequential complexity, such as Littlestone dimension, and can be impossible even for statistically simple classes, such as thresholds of VC dimension one. We study a preview model in which an oblivious adversary fixes an entire labeled sequence of length $T$, a uniformly random subset of size $pT$ is revealed before prediction begins, and the remaining $(1-p)T$ examples are then presented in their original adversarial order. Against the best full-sequence hypothesis evaluated on the unrevealed examples, we characterize the dependence on the preview rate $p$: for binary classes of VC dimension $d$, the optimal excess loss is $\Theta(d/p+\sqrt{dT})$, up to the trivial cap at $T$; for multiclass classes we obtain the corresponding $\widetilde O(d_{\rm DS}/p+\sqrt{d_{\rm Nat}T})$ bound with no dependence on the number of labels. Thus a random preview can replace worst-case sequential complexity by classical statistical dimensions without randomizing the online order. To achieve the sharp binary bound, our ChainedPrediction algorithm uses an online analogue of chaining, implemented as a multiscale aggregation algorithm rather than only as an analytic argument.

Read the full abstract

Numerical Analysis

ENPINN: Energy-Norm-Guided Gradient-Enhanced PINNs for Generalized Transport Problems with Sharp Gradients

Subhendu Maity, Pratibhamoy Das, Arihant Patawari et al. (+1 more)

Physics-informed neural networks (PINNs) have emerged as a meshless alternative to conventional numerical methods for solving partial differential equations (PDEs). However, their limited ability to capture sharp gradients can lead to substantial errors when resolving boundary and interior layers. Here, we introduce an energy-norm-enhanced PINN (ENPINN) that incorporates gradient information and variational structure into the loss function to improve the resolution of layer-dominated solutions. We first examine two related formulations: weak-loss PINNs (WLPINNs), which incorporate test functions into the conventional PINN residual, and gradient-enhanced PINNs (gPINNs), which augment the loss with spatial derivatives of the PDE residual. By analyzing these formulations, we identify their limitations in resolving steep solution gradients and motivate the systematic construction of ENPINN. We establish theoretically how the energy-norm error depends on the ENPINN loss and show that a suitably modified residual-derivative term is essential for accurately capturing boundary layers. We further establish the existence of neural-network approximations with arbitrarily small energy error and derive corresponding derivative bounds, providing a theoretical foundation for the proposed framework. The performance of ENPINN is assessed through systematic comparisons with existing PINN variants for convection-diffusion-reaction problems exhibiting steep gradients. Numerical experiments include a combustion model, a coupled multi-scale system, a two-dimensional Burgers equation with an interior layer, and a three-dimensional time-dependent problem.

Read the full abstract

All authors Subhendu Maity, Pratibhamoy Das, Arihant Patawari, Ameya D. Jagtap

Game Theory

Fair Division Under Boolean Valuations: Beyond Normalization

Nisarg Shah and Paritosh Verma

We study fair division of indivisible items when agents have arbitrary two-level preferences: the value of each agent for any set of items is Boolean, which need not be monotone or additive. Notably, we do not impose the standard assumption of normalization, i.e., different agents may value the empty set at different Boolean levels. Since the preferences are nonmonotone, envy-freeness up to one item (EF1) and envy-freeness up to any item (EFX) each admit several variants, depending on which items are tested for removal and whether they are removed from the envious agent's bundle or the envied agent's bundle. This paper investigates the existence of these variants of EF1 and EFX, on their own and together with economic efficiency, incentive compatibility, feasibility constraints, and lottery-based randomization. Our results highlight that the existence landscape depends crucially on the number of normalized agents, who value the empty bundle at the lower Boolean level. The authors used significant assistance from GPT-5.6-Sol for deriving theoretical results, verified any AI-generated proofs for correctness, and expanded on the exposition and simplified arguments, with the aid of GPT-5.6-Sol and Claude Opus 5.

Read the full abstract

Machine Learning

Target-Aware State-Adaptive $p$-Dirichlet Graph Neural Regression for Non-Invasive Body-Composition Estimation

Nadejda Drenska, Matthew Lemoine, Gowri Priya Sunkara et al. (+3 more)

Accurate estimation of body-composition outcomes, including body fat percentage (BFP), bone mineral density (BMD), and appendicular lean mass (ALM), is important for evaluating metabolic, skeletal, and muscular health. Direct assessment using dual-energy X-ray absorptiometry (DXA), however, requires specialized equipment and involves ionizing radiation. We propose a target-aware, state-adaptive $p$-Dirichlet energy-flow graph neural regression ($p$SADE-GNR) framework for estimating these outcomes from non-invasive anthropometric measurements. A neural encoder maps participant representations to hidden states that are propagated over an outcome-specific participant-similarity graph by a state-adaptive forward-Euler discretization of the graph $p$-Dirichlet energy flow. Graph distances weight each original or latent coordinate by its normalized absolute training-fold correlation with the outcome. Using clinical data from the Pennington Biomedical Research Center and five-fold cross-validation, the correlation-weighted model using the original standardized measurements achieved the lowest root mean squared error in all nine primary outcome-cohort combinations and outperformed previously reported support vector regression or least-squares support vector regression reference values in eight of nine comparisons. Autoencoder, variational-autoencoder, and Gaussian-mixture variational-autoencoder representations generally did not improve primary-outcome prediction or reduce computational cost. In an exploratory age-prediction analysis including ALM, BMD, and BFP as predictors, the correlation-weighted GMVAE model achieved the lowest mean error in all three cohorts. These results support target-aware, state-adaptive $p$-Dirichlet graph neural regression for non-invasive body-composition estimation.

Read the full abstract

All authors Nadejda Drenska, Matthew Lemoine, Gowri Priya Sunkara, Yu Wang, Sri Lakshmi Sravani Devarakonda, Steven B. Heymsfield

Machine Learning

Learning Human Health and Diseases from 24-hour Wrist Movement

Yong Wang, Dylan McGagh, Katya Broomberg et al. (+21 more)

Much of human health and function unfolds beyond the clinic, through the movements of everyday life. Wrist-worn accelerometers capture these movements continuously, yet their rich signals are often reduced to a small set of predefined behavioural summary measures. Here, we present Sensori, a self-supervised foundation model that learns general-purpose health representations directly from 24 hours of raw tri-axial wrist movement. We developed and evaluated the model across four population-based cohorts from the United Kingdom, China and the United States, comprising 122,640 participants contributing 683,617 person-days of free-living recordings. Sensori condensed each day of movement into a representation that captured diverse movement behaviours, demographic characteristics, health axes and physical function. Evaluation in independent cohorts showed that these representations generalised across populations and measurement settings without retraining. When added to common clinical covariates, Sensori significantly improved prevalent disease classification for 52 of 102 eligible conditions (median delta AUROC, 0.060; range, 0.012-0.242) and incident disease risk prediction for 26 of 87 eligible conditions (median delta Uno's C-index, 0.064; range, 0.025-0.172), with the largest gains for neurological and psychiatric disorders. These findings establish 24-hour wrist movement as a rich and scalable source of health information, with the potential to support passive health monitoring and disease prediction at population scale.

Read the full abstract

All authors Yong Wang, Dylan McGagh, Katya Broomberg, Zizheng Zhang, Jonathan Carter, Junayed Naushad, Laura Brocklebank, Yang Sun, George Nicholson, Dianjianyi Sun, Canqing Yu, Jun Lv, Maxim Barnard, Hubert Lam, Andrew Steptoe, David W. Eyre, Liming Li, Zhengming Chen, Naomi Wray, Spiros Denaxas, Gary S. Collins, Huaidong Du, Aiden Doherty, Hang Yuan

Computation and Language

CoCoA: Context-Conditional Cultural Alignment for Large Language Models

Kyungdon Lee, Wei Xu, Alan Ritter et al. (+2 more)

Large Language Models (LLMs) often favor Western-associated entities across cultural contexts. Conventional debiasing methods aim for uniform neutrality, but cultural bias mitigation demands context-conditional behavior, preferring culturally appropriate entities when cultural cues are present and remaining neutral when they are absent. We propose CoCoA (Context-Conditional Cultural Alignment), a framework that learns this behavior through dual-context training on the same entity pairs under contexts with and without cultural cues. CoCoA combines a contrastive alignment objective with calibration and drift regularization, optimized through goal-aware gradient reconciliation. We evaluate CoCoA on CAMeL and Camellia, two entity-centric cultural bias benchmarks, across ten language settings and four LLMs. CoCoA reduces the Cultural Bias Score from 43 to 24 on average while maintaining near-neutral preferences at 50.2, with minimal impact on general performance across five standard benchmarks. These findings highlight that effective cultural alignment requires context-conditional modeling rather than uniform debiasing, and establish a new direction for mitigating entity-centric cultural bias in LLMs.

Read the full abstract

All authors Kyungdon Lee, Wei Xu, Alan Ritter, Dong-Ho Lee, JinYeong Bak

Multiagent Systems

Generalizable Multi-Agent Planning from Signal Temporal Logic Specifications via Diffusion

Joe Eappen, Zikang Xiong, Shreyash S. Iyengar et al. (+1 more)

Multi-agent systems in the real-world (e.g., drone swarms, autonomous cars, warehouse robots) must satisfy rich, temporal tasks while avoiding collisions. Signal Temporal Logic (STL) elegantly encodes such objectives, but current STL planning methods face critical limitations. State-of-the-art optimization-based approaches can handle arbitrary STL specifications but struggle with scalability, becoming computationally impractical as the number of agents grows. Learning-based methods efficiently handle a large number of agents with rapid planning times but fare poorly when deployment-time objectives differ from those used during training, and do not support planning tasks that require different specifications to be ascribed to different agents (i.e., heterogeneity) or team-level specifications requiring coordination of multiple agents. This fundamental trade-off between generalizability and scalability presents a challenge for realizing multi-agent STL planning algorithms in practice. To overcome this challenge, we introduce a new diffusion method for multi-agent planning with STL specifications. Using a differentiable approximation of STL, we integrate the STL gradient in the denoising process, making our approach generalizable to novel formulas whose predicates are placed anywhere within the goal region covered during training, while achieving the same scalability as existing learning-based methods. Our method supports heterogeneous specifications, and by using diffusion models, naturally enhances plan diversity, thereby significantly reducing safety-related violations (e.g., collisions) among agents. A detailed evaluation study justifies the utility of STL-guided diffusion-based multi-agent planners for constructing generalizable, scalable, and diverse plans. Videos and code are available at https://www.jeappen.com/diff-ma-stl/ and https://github.com/jeappen/diff-ma-stl .

Read the full abstract

All authors Joe Eappen, Zikang Xiong, Shreyash S. Iyengar, Suresh Jagannathan

Cryptography and Security

SpatialTrust: A Benchmark for Environmental Risk Recognition in Secure Authentication

Junbin Lu, Hsiang-Wei Huang, Saesha Wadhwa et al. (+2 more)

Visual environmental risk recognition plays an important role in secure authentication, where a user's surroundings may reveal sensitive information or introduce potential security risks. However, existing evaluations of multimodal large language models (MLLMs) rarely examine whether models can reliably recognize, localize, and explain such risks in spatially grounded authentication scenarios. We present SpatialTrust, a question-answering benchmark for evaluating environmental risk recognition in secure authentication. SpatialTrust assesses five complementary abilities: sensitive factor detection, direct factor identification, indirect factor identification, direct factor explanation, and indirect factor explanation. We evaluate both proprietary and open-source MLLMs and find that current models show limited performance, especially in understanding and explaining indirect risks, indicating that spatial risk awareness remains a challenging capability for MLLMs. In addition, we introduce SpatialTrustGuard, a structured QA-and-audit pipeline that improves Qwen3-VL-30B-A3B-Instruct from 36.78% to 41.12% overall. Our findings highlight the need for dedicated benchmarks and structured inference methods to improve the trustworthiness of MLLMs in secure authentication.

Read the full abstract

All authors Junbin Lu, Hsiang-Wei Huang, Saesha Wadhwa, Yu Ting Hsu, Jenq-Neng Hwang

Robotics

Blind Dexterity: Whole-Body Humanoid Manipulation via Pure Proprioception

Aditya Bhatt, Oleg Kaidanov, Puze Liu et al. (+1 more)

We present blind, whole-body manipulation skills on a Unitree G1 humanoid using only onboard proprioception, without cameras, markers, force-torque, or tactile sensors. Despite this minimal sensing, the trained policies exhibit surprising capability across qualitatively different tasks: push-resilient bipedal walking without IMU feedback, active soccer ball trapping with a foot, seeking and lifting a suitcase by its handle, and mounting a randomly positioned skateboard. We argue that these capabilities arise from a key underappreciated signal: the way the joint encoder readouts evolve under purposeful compliant contact, effectively forming a whole-body tactile channel. By generating contact-rich motions, the trained policies actively probe the environment; as a result, task-relevant object state (e.g., pose) becomes increasingly decodable from short proprioceptive histories. We expose this information using compact task-specific state estimators trained alongside, but fully separately from, the policies; their prediction errors decrease rapidly after informative contact. Our results indicate that joint encoder-based proprioception, combined with compliant actuation (now widely available on commercial robots and low-cost motors) is already a strong, practical substrate for whole-body dexterous manipulation and interactive perception, and therefore a natural foundation on which richer sensing can be layered.

Read the full abstract

All authors Aditya Bhatt, Oleg Kaidanov, Puze Liu, Jan Peters

Computer Vision

GeoAgent: Evaluating VLM Geolocalization Through Embodied Navigation

Arka Mukherjee, Soham Roy, Kartikeya Trivedi et al. (+1 more)

Modern Vision-Language Models (VLMs) perform well above the human baseline in image geolocalization, a task critically important in disaster response, OSINT verification, and location privacy. However, most efforts to study AI behavior on the task remain limited to static image-based retrieval, classification, and predictions. We argue that faithful recreation of the task should involve embodied navigation, where a multimodal agent autonomously explores its surroundings to gather observations before submitting a prediction. To this end, we introduce GeoAgent, an agentic environment-based benchmark that requires agents to navigate Street View environments to refine their geolocalization through sequential reasoning. Our analysis shows that modern VLMs struggle to discern regional patterns while succeeding at country- and continent-level predictions. When compared to static image-based baselines, agentic navigation significantly improves accuracy across established metrics. We also note severe bias in a developed/developing region context across frontier model architectures and poor self-improvement capabilities given incorrect priors. Overall, our work establishes the challenges of embodied navigation and geospatial reasoning. We publicly release our code and the GeoAgent environment: https://geoagent-benchmark.github.io

Read the full abstract

All authors Arka Mukherjee, Soham Roy, Kartikeya Trivedi, Shreya Ghosh

Symbolic Computation

Parallel Integration over Simple Radical Extensions

Sam Blake

The parallel Risch (Risch–Norman) method is a fast heuristic for computing elementary integrals over towers of transcendental extensions. Its justification rests on two structural facts about the integral: a bound on its denominator and a description of the logarithms that can occur. Both are known for purely logarithmic towers (Davenport–Trager) and, in the form of a structure theorem, for arbitrary derivations on multivariate rational function fields (Bronstein). We extend both facts to a simple radical extension $L=K(y)$, $y^m=q$, of such a field. The key observations are that the integral closure of $F[t_1,\dots,t_n]$ in $L$ has an explicit basis, so that all factorisation can remain in a polynomial ring, and that the derivation has a well-defined pole order $\delta_P\in\{0,1,e_P\}$ at every height-one prime $P$, so that pole orders of derivatives shift by $\delta_P$. The denominator of the integral then has the same Hermite-type shape as in the transcendental case, while the admissible logands are precisely the $S$-units of the integral closure for an explicit finite set $S$ of primes; the latter can be larger than the set generated by irreducible polynomials, as the unit $x+\sqrt{x^2+1}$ shows. For $n=1$ we relate these $S$-units to torsion in the Jacobian and, for $m=2$, to the polynomial Pell equation, obtaining a complete description of the logarithmic part in genus~0. We describe the resulting algorithm and give examples.

Read the full abstract

Computation and Language

SIC-Agents: Benchmarking and Building an Adaptive Simulator for Pediatric Serious Illness Communication Training

Zihan Wang, Anita Marie Slominska, Rennie Bimman et al. (+10 more)

Pediatric serious illness communication (SIC) is critically important, yet scalable communication training for clinicians remains limited. Compared with other dialogue simulation settings, pediatric SIC poses additional challenges, including multi-party interactions, response to parental distress and strong dependence on feedback dynamics. Existing LLM-based simulators optimize generic dialogue quality rather than curriculum-contingent behavior required for effective SIC training. In collaboration with educators and pediatric clinicians, we introduce the first benchmark suite and simulation framework tailored to pediatric SIC training. Our benchmarks, PitfallBench and DialogueBench, evaluate simulators both at the turn-level and across full dialogues. We further propose SIC-Agents, a self-improving framework that generates a clinician-editable skill document to guide simulator behavior. Our experiments show that SIC-Agents outperforms static expert prompting. To support future research, we release our benchmarks for parent simulation in pediatric SIC at https://github.com/Beikewzh/sic-benchmarks

Read the full abstract

All authors Zihan Wang, Anita Marie Slominska, Rennie Bimman, Elizabeth Di Flumeri, Amanda Mayappo-Neeposh, Conall Francoeur, Tamara Ellen Carver, Xiao-Wen Chang, Doina Precup, Esin Darici Haritaoglu, Ismail Haritaoglu, Akshatha Arodi, Naomi Goloff

Sound

Note-Level Temporal Grounding of Musical Concepts in Large Audio-Language Models

Kun Fang, Ziyu Wang and Ichiro Fujinaga

Large audio-language models (LALMs) demonstrate growing music-understanding capabilities, but whether their responses are grounded in acoustic evidence remains unclear. Musical language often involves abstract concepts whose acoustic evidence is difficult to define and evaluate precisely. We introduce MusicGroundingBench, a controlled benchmark of algorithmically generated piano audio with exact symbolic alignment, comprising three-note and two-bar settings. We evaluate two complementary capabilities: grounding, which localizes the acoustic evidence for a musical query, and understanding, which answers questions about the same excerpts. Our experiments show that cross-modal fine-tuning enables models to learn each capability, but adding grounding supervision does not consistently improve understanding across backbones. We further test whether understanding requires listening through audio-ablation controls that remove or replace the input audio, and use attention analysis to examine whether grounding supervision shifts attention toward note boundaries. Meanwhile, the two evaluated LALMs show limited zero-shot grounding even for basic musical concepts, highlighting grounded music understanding as an important open challenge.

Read the full abstract

Systems and Control · Cross-listed from math.OC

The De Cock–De Moor Conjecture: Proof, Normal Form, and Indefinite Angles

Jeffrey Humpherys

De Cock and De Moor conjectured that two matrices built from the solutions of three coupled Lyapunov equations always share their eigenvalues, an identity linking principal angles between model subspaces with the canonical correlations between the past and future of a stochastic process. The conjecture was recorded as Problem 9.1 of Unsolved Problems in Mathematical Systems and Control Theory. We prove it under the natural nonresonance condition, without stability or diagonalizability, and show that the two matrices are similar. The proof exposes a Cauchy-matrix structure created by the rank-one perturbation and reduces the problem to rational interpolation, and the main theorem has also been verified in the Lean~4 proof assistant. The same structure gives a normal form that depends only on the poles and zeros of the model. Off the stable set the eigenvalues become squared cosines of angles in an indefinite metric. We prove an inertia theorem for them and show that reflecting a pole or a zero across the unit circle acts on the angles by a principal pivot transform. For several inputs and outputs, where the conjecture fails, we identify the subspaces whose angles do give the canonical correlations. We also relate our approach to the independent proof of Gillberg and Löfberg.

Read the full abstract

Computers and Society

Applications of Risk Science to AI Fairness Evaluation: Principles, Challenges, and Best Practices

Kyra Wilson, Sabrina Kang, Saloni Dash et al. (+1 more)

Scholarly work which aims to describe potential societal impacts (e.g., risks) of proliferating technology (especially related to artificial intelligence or other algorithmic systems) is likely to have an impact beyond the scientific communities it was written for, given that general society itself is a primary object of study. However, it is an open question whether the current practices of AI evaluation scholarship follow the principles and best practices established by risk science, which aims to systematically generate knowledge related to understanding, assessing, communicating, managing, and governing risk. In this work, we examine this in depth by conducting a literature review of scholarly works purporting to evaluate the bias or fairness of technological systems used for tasks related to hiring and employment. Through analysis of 22 common fairness evaluation metrics and studies using them, we find that most characterize the severity of bias- or fairness-related consequences but do not follow best practices to characterize the uncertainty around either the occurrence of these consequences or severity estimates. Next, we conduct a case study of fairness evaluation for an AI-mediated resume screening task and demonstrate how principles of risk science can be incorporated into such an evaluation. Finally, we propose the AI Risk Report Card, which facilitates the reporting and communication of risk assessment results to stakeholders in positions to act based on the predicted risks. The outcomes of these activities suggest that further research at the convergence of risk science and AI evaluation can lead to advancements in AI assessments of societal impact by enabling shared frameworks to evaluate and discuss AI risks both within and outside of the scientific community.

Read the full abstract

All authors Kyra Wilson, Sabrina Kang, Saloni Dash, Aylin Caliskan

Computation and Language

MUDDLE: Measuring Understanding of Documents under Distractor and Length Effects

Jason Luo, Saibilila Abudukelimu, Judy Song et al. (+4 more)

Document question-answering systems increasingly answer questions over collections of retrieved documents rather than one clean source, so robustness to distracting context matters as much as reading ability. When such systems fail, it is often unclear whether the context was too long or the distractors were too close to the topic, because prior work tends to conflate these two effects. We present MUDDLE, a controlled benchmark that separates them. MUDDLE uses 270 human-annotated questions, each tied to a single source document, and instantiates every question in five conditions: the source alone, the source with two or four topically similar hard negatives, and the source with two or four random distractors. The random distractors are matched to the hard negatives in length and provenance, so an accuracy gap between the two arms reflects topical similarity rather than length. All five conditions are rendered in markdown, page images, and raw PDF, but the distractor sweep reported here is run in markdown, since a source plus its distractors exceeds current image and PDF input limits. We score answers with an LLM judge across three model families. In the complete markdown sweep, hard negatives lower accuracy more than length-matched random documents at both context sizes for gpt-5-mini, while random documents stay near the no-distractor baseline. The effect is small but directionally consistent, and for gpt-5-mini hard negatives significantly underperform length-matched random distractors when pooled across context sizes. We release the data and evaluation code for a reproducible study of context degradation.

Read the full abstract

All authors Jason Luo, Saibilila Abudukelimu, Judy Song, Andrew Feng, Shivank Garg, Vasu Sharma, Kevin Zhu

Computer Vision

Seeing Through Extreme Visual Sparsity: Surface Understanding from a Single Random Visual Patch

Sindhuja Penchala, Sudip Mittal and Noorbakhsh Amiri Golilarz

Surface material recognition from incomplete visual observations remains a challenging problem in robotic perception and environmental understanding. This paper discusses Sparse Surface Understanding Framework (SSUF), a unified dual-task learning framework that adapts four pretrained architectures-Convolutional Autoencoder (ConvAE), Vision Transformer (ViT), Swin Transformer, and Masked Autoencoder (MAE) for si-multaneous surface reconstruction and material classification. Experiments were conducted on the Touch-and-Go dataset using a sparse observation protocol in which only 10% of the original image remained visible while the remaining regions were masked. To enable a fair comparison, reconstruction-oriented models were extended with classification heads, whereas classification- oriented models were augmented with reconstruction decoders. The resulting architectures were assessed using reconstruction quality, classification performance, model complexity, and in-ference efficiency metrics. Experimental results revealed distinct strengths across the models. Swin Transformer achieved the best classification performance with an accuracy of 89.21%, an F1-score of 0.8922, and a ROC-AUC of 0.9813. In contrast, MAE produced the highest reconstruction scores among evaluated models, with a PSNR of 16.06 dB and an SSIM of 0.4501, while ViT provided the best overall balance between reconstruction and classification performance. Furthermore, all models achieved real-time inference, requiring less than 5 ms per image. Over-all, the results show that pretrained architectures can support material recognition under severe visual sparsity, while accurate image reconstruction remains challenging.

Read the full abstract

Systems and Control

Relative-Degree Wall Restricts Passivity-Based Stability Analysis in Inverter-Dominant Grids

Xiaoyu Peng, Zhongze Li, Xi Ru et al. (+2 more)

This letter reveals a fundamental limitation of passivity-based distributed stability analysis in power systems. Under the standard formulation, passivity certification inherently imposes a relative-degree compatibility constraint that excludes many high-fidelity inverter dynamic models (e.g., those that include electromagnetic transients). Potential extensions of passivity frameworks are discussed to break this limitation.

Read the full abstract

All authors Xiaoyu Peng, Zhongze Li, Xi Ru, Xinghua Chen, Feng Liu

Saturday 29 August

Human-Computer Interaction

Semantic-Enhanced Sparse Temporal Modeling for Dynamic Engagement Estimation in Multimodal Interactions

Yangchen Yu, Taoliang Wang, Jia Li et al. (+4 more)

Dynamic engagement estimation aims to infer participants' evolving engagement states from continuous multimodal interaction streams. Existing approaches typically combine acoustic, pose, and visual representations with temporal models to capture observable interaction patterns. However, conversational semantics are difficult to align with frame-level engagement annotations in the same manner as continuous audio-visual signals, leaving dialogue context less effectively exploited in existing temporal modeling pipelines. Moreover, engagement annotations often evolve smoothly over time, making dense frame-level modeling computationally redundant and potentially sensitive to local annotation noise. To address these issues, we propose SEST, a Semantically-Enhanced Sparse Temporal framework that models engagement-related information from both semantic and temporal perspectives. SEST leverages Multimodal Large Language Models to extract multimodal semantic embeddings, providing dialogue-aware contextual information beyond conventional frame-level features. In parallel, SEST replaces dense-to-dense temporal modeling with a sparse-to-dense pathway, where continuous frame sequences are first encoded into compact temporal dynamics to reduce redundant computation and broaden temporal coverage, then fused with semantic context and recovered to the original label resolution for dense engagement prediction. Experiments on the MEE2026 benchmark demonstrate the effectiveness of SEST, which achieved first place in the challenge.

Read the full abstract

All authors Yangchen Yu, Taoliang Wang, Jia Li, Nor Mazlina Ghazali, Mohd Nazim Mohtar, Bingkun Bao, Richang Hong

Programming Languages

Replay Is a Family of Assertions: Fork-Safety in Event-Sourced Agent Runtimes

Yohei Nakajima

Event-sourced agent runtimes commonly claim deterministic replay, cheap forking, and lineage. These are families of assertions, not one guarantee. A fixed-log fold may be deterministic while reactive settlement remains schedule-sensitive; a recorded oracle result can enable zero-call replay while the original interaction still changed cost or world state; and a prefix copy cannot undo an action in a discarded suffix. We present an executable F# contract that separates facts from effect requests, makes scheduling explicit, and assesses a cut relative to projection, strict execution replay, external continuation, or a declared counterfactual-world observation. Preserved counterexamples show that general settlement is neither terminating nor confluent and that disjoint declared writes do not prevent read/trigger interference. Against 5,540 public runtime logs, all 317,296 projection cuts satisfy the stated observation, while 36,251 cuts split a model request and do not license continuation; another 120,969 are Conditional because a committed oracle result must be served from the record without re-execution. In 121 observed forks, 16,625 retained-prefix events have zero structural mismatches, but none crosses an external-effect boundary. A separate public offline conformance fork crosses a committed, recorded oracle boundary. Its generator observes that the inherited result is served without a second fixture-oracle call; its verifier checks receipt/log consistency and a fork-bound HMAC assertion under a public conformance key. The result reframes cheap forking as a property-indexed semantic contract with explicit evidence and environment preconditions.

Read the full abstract

Artificial Intelligence

Ingest-Time Fact Compilation for Cost-Efficient and Reliable Question Answering over Revised Corpora

Kyle Wild, Yusuke Takahashi and Asako Uraki

Most agentic question answering (QA) systems do an important part of their semantic work at the worst possible time: every time someone asks a question. When a corpus contains revisions, drafts, revocations, deletions, and sources with different levels of authority, the model must reconstruct the governed current state on every read - then throw that work away and repeat it on the next query. This is a bit like a database that rebuilds a materialized view every time someone reads from it. We present ingest-time fact compilation, an architecture that performs this work when corpus data is ingested or changed. Raw passages are rephrased into self-contained facts; rules governing revisions, deletions, effective dates, and source trust are resolved once; and the resulting state is stored as typed records carrying source and revision provenance. At query time, an inexpensive model reads the compiled record instead of reconstructing it from noisy candidates. In a controlled synthetic experiment across five seeds, the same low-cost model produced the correct value, source, and revision in only one of 30 trials under query-time reconstruction, but in all 30 trials from the compiled substrate, at 12.89 times lower mean read cost per question. On simpler revision questions both architectures were exact, but the compiled path used 21.6 times fewer tokens. A separate test found that fact rephrasing roughly halved verbose Federal Reserve dialogue while preserving high source entailment, but left concise Wikipedia prose essentially unchanged. These results support a narrow but practical claim: resolving a corpus state once can make subsequent QA cheaper and more reliable for inexpensive models. We release the open source, MIT-licensed implementation and experimental artifacts.

Read the full abstract

Computation and Language

How To Do Things With Prompts

Kristina Šekrst and Virna Karlić

When users address large language models, they produce directive speech acts whose pragmatic features differ from those of both everyday conversation and traditional human-computer interaction, and these features change as users gain familiarity with the systems they address. This paper applies speech act and politeness theory to a corpus-pragmatic analysis of 2,000 English-language prompts drawn from publicly shared ChatGPT conversations, 1,000 from 2023 and 1,000 from 2025, using the ShareChat dataset. Each prompt is annotated for illocutionary force, directness, propositional content, and the presence of politeness markers, and the distribution of these features is compared across the two sampling years. The results show a consistent movement toward indirect, implicit, and fragmentary realizations of directive force, accompanied by a decline in politeness marking. The largest single change, a shift of 14.9 percentage points, occurs in propositional content, where explicit specification of the requested action gives way to implicit reliance on the system's inferential capacity, suggesting that users have updated their model of what the system can recover from reduced input, treating it as a competent implicature resolver. Rather than asking whether LLMs "really" understand language, we should ask: what kind of language have we created in learning to speak to them?

Read the full abstract

Software Engineering

Object-to-Source Mapping under Optimization: What a Control-Flow Trace Can Show, and Where Source-Level Coverage Breaks

Alexander Weiss and Albert Schulz

Within a validated interval, control-flow trace can reconstruct machine-level execution, but source-level structural coverage additionally requires a validated attribution from generated instructions and branch outcomes to source obligations. This paper defines that boundary and illustrates it with exact compiler output. Optimization can fuse multiple source conditions into one branch, replace a branch with a conditional select or move, and degrade provenance in debug and mapping information. In such cases, trace may be complete for the generated binary while condition-level source evidence remains unresolved. We therefore separate trace completeness, source attribution, and coverage satisfaction; distinguish source-to-object transformations from object-region attribution states; and explain when internal reconvergence makes aggregate branch observations insufficient and ordered per-instance condition-evaluation histories necessary. We compare measuring a less-optimized build, constraining code generation, and adding state-writing instrumentation. Finally, we describe a pre-compilation supplementation mechanism disclosed in WO 2026/162831 A1 that preserves selected source distinctions as trace-observable control flow without writes to application-resident coverage state. An appendix demonstrates the mechanism end-to-end on real silicon for the worked example: the supplemented -O3 build keeps each condition individually trace-observable, per-instance histories and an MC/DC verdict are reconstructed from a captured ETM trace, and code-size and cycle costs are measured. Its use remains subject to build-specific preservation, final-image mapping, trace-validity, and probe-effect verification.

Read the full abstract

Information Retrieval

SmallReason-ColBERT: An Ultra-Small Late-Interaction Retriever for Reasoning Intensive Retrieval

Abdelrahman Abdallah, Mohammed Ali and Adam Jatowt

Reasoning-intensive retrieval remains difficult for small models. Compact public ColBERTs are usually trained on general-purpose corpora and underperform reasoning-tuned 150M+ baselines on BRIGHT~\cite{bright} by several nDCG@10 points. However, no public reasoning-tuned ColBERT exists at edge scale. We introduce SmallReason-ColBERT, a 32M late-interaction retriever that closes much of this gap with three components: a varied-length contrastive warmup on ReasonIR-VL, a hard-negative contrastive polish on merged ReasonIR-HQ and BGE-Reasoner data, and a single-layer per-query-token importance head trained on top of the frozen base. The head is trained with an un-normalised weighted MaxSim score and evaluated with its length-normalised form. In a controlled re-training, replacing this training objective with the symmetric normalised score causes the loss to stall and costs $3.59$ nDCG@10. The full recipe reaches 21.41 mean nDCG@10 on BRIGHT, within $1.21$ of the 150M Reason-ModernColBERT (22.62) and above all $\le 33$M ColBERTs we evaluate. Through ablations on capacity, initialisation, and score variants, we further show that the learned head outperforms fixed IDF weighting and that simply thresholding the learned gates is harmful. https://github.com/DataScienceUIBK/SmallReason-ColBERT

Read the full abstract

Computer Vision

VG-TIE: An interpretable tabular-to-image encoding method based on visibility graphs

David Chushig-Muzo, Luis M. López-Ramos, Ángeles Rodríguez de Cara et al. (+3 more)

Tabular-to-image encoding methods enable the application of models based on both convolutional neural networks and vision transformers to tabular data, transforming feature vectors into images. Existing methods employ linear and nonlinear dimensionality reduction techniques (e.g., Principal Component Analysis (PCA), t-SNE, and UMAP) to determine pixel positions, resulting in images whose spatial layout do not inherently reflect feature relationships. This paper introduces Visibility Graphs for Tabular-to-Image Encoding (VG-TIE), a novel method that encodes the structure of feature values using Natural Visibility Graph (NVG) and Horizontal Visibility Graph (HVG) into a two-dimensional space obtained through PCA. The resulting images are model-agnostic and intrinsically interpretable. Each pixel corresponds to an input feature, its intensity reflects the magnitude and direction of deviation from the population mean, and edges represent formally defined visibility relationships between features. VG-TIE provides two interpretability methods: (i) feature ranking from node degree distributions; and (ii) local and global feature importance from pixel intensity combined with Grad-CAM. Experiments on six public tabular datasets show that VG-TIE is competitive with other tabular-to-image methods while providing interpretability on feature importance and ranking similar to intrinsic interpretable methods. The results highlight the potential of the proposed image-based transformation to provide an effective framework that expands the use of deep learning across tabular data domains.

Read the full abstract

All authors David Chushig-Muzo, Luis M. López-Ramos, Ángeles Rodríguez de Cara, Eva Milara, Luis Zhinin-Vera, Diego H. Peluffo-Ordóñez

Computer Vision

Albireo: Adaptive, Energy-Efficient Inference Framework for Video Object Detection on the Edge

Amir Taherin, José Cano, Bin Ren et al. (+2 more)

Video object detection on edge devices runs computationally expensive detectors over long frame streams, causing high energy consumption and sustained GPU utilization. Although consecutive frames are highly redundant, naive frame skipping is content-blind: it skips during critical moments such as object entry, occlusion recovery, and abrupt motion, degrading detection quality. We present Albireo, a detector-agnostic, codec-free adaptive inference framework that wraps off-the-shelf detectors and decides when detector invocation can be safely skipped based on scene content and per-object temporal state, requiring no detector modification or retraining. Albireo maintains a 10-dimensional Kalman filter (KF) per active object and invokes the detector only when prediction uncertainty exceeds a threshold; on skipped frames, boxes are predicted from the KF state at near-zero GPU cost. A KF-based rescue mechanism preserves confirmed objects through brief detector misses to prevent output fragmentation, while a lightweight empty-scene screen avoids detector calls on objectless frames. We evaluate Albireo on the BDD100K MOT validation split with three architecturally distinct detectors (YOLO11x, YOLO26x, RF-DETR-Large) on two NVIDIA Jetson platforms (AGX Thor, AGX Orin). Across all configurations, Albireo keeps AP@50 within +/-1.2 pp of per-frame inference while reducing total energy by 12.1-17.6%. On YOLO26x, it improves AP@50 by +0.8 pp while reducing energy by 17.6% (Thor) and 14.4% (Orin) and per-frame energy-delay product by 24.9% and 26.1%, respectively. Thus, the default operating point improves accuracy, energy, and latency together. In contrast, FixedSkip-2, a fixed-interval baseline with a 50% skip rate, loses 8.6 pp AP@50. Source code, evaluation pipeline, and per-clip results are available at https://github.com/amirtaherin/albireo

Read the full abstract

All authors Amir Taherin, José Cano, Bin Ren, Yanzhi Wang, David Kaeli

Cryptography and Security

AgentKernel: The Trust-Native Agentic Operating System

Zhenhua Zou, Sheng Guo, Qiuyang Zhan et al. (+3 more)

Modern AI agents routinely cross trust boundaries: they ingest untrusted content, combine it with privileged instructions, persist intermediate beliefs in long-term memory, and invoke privileged tools. This creates an attack surface in which malicious payloads can enter through model inputs and cause harmful tool actions. Yet current governance stacks remain application-level middleware that share a process trust boundary with the agents they monitor. We argue that agents need an operating-system substrate providing mandatory, non-bypassable services for identity, input mediation, memory governance, and execution control. We introduce AgentKernel, a trust-native agent operating system built around the premise that security must be a first-class design constraint. AgentKernel wraps the agent lifecycle in a mandatory enforcement boundary organized into four pillars: Identity, Perception, Cognition, and Execution. Each pillar adapts classical OS security principles to failures at the semantic plane, including delegation abuse, prompt injection, memory poisoning, and tool misuse. AgentKernel treats structural security as a capability multiplier. Kernel-managed identity supports trustworthy cross-organization collaboration; graduated perception replaces brittle single-point filters; information-flow-controlled memory improves retrieval fidelity while limiting poisoning; and semantic-to-kernel enforcement permits broader tool privileges behind a non-bypassable boundary. We position AgentKernel as the missing OS layer beneath orchestration frameworks, agent runtimes, governance platforms, and execution sandboxes, and use systematic comparison and security analysis to show how a single integrated architecture can enforce security across the full agent lifecycle.

Read the full abstract

All authors Zhenhua Zou, Sheng Guo, Qiuyang Zhan, Lepeng Zhao, Shuo Li, Zhuotao Liu

Systems and Control

Distributed Algorithms for Filtering, Estimation, and Fault Detection over Cyber-Physical-Systems: A Tutorial and Survey

Mohammadreza Doostmohammadian, Mahdi Shamsi, Hadi Zayyani et al. (+4 more)

This survey provides overview of distributed estimation, filtering, and fault detection techniques in the context of cyber-physical systems (CPS). To establish a strong foundation, we first define essential aspects of linear dynamical systems and related graph theoretic concepts, emphasizing observability conditions that are key for local state estimation. After discussing consensus algorithms, this survey highlights single-time and double-time-scale consensus-based estimation and filtering approaches. We provide a detailed comparative analysis of these methodologies, examining their computational requirements, communication overhead, and performance characteristics in resource-constrained environments. We further explore different diffusion-based estimation techniques and observationally redundant designs to enhance resilience and robustness against failures and adversarial attacks. In addition, we investigate distributed fault detection methods that enable local isolation of faults over large-scale CPS. We present both stateless and stateful detection mechanisms, along with threshold-based techniques that balance detection accuracy and false alarm rates. These approaches are essential for maintaining system integrity and preventing cascading failures in critical infrastructure. This survey concludes with an exploration of diverse real-world applications. We examine implementation challenges and algorithm adaptations in smart grid and power networks, social systems, target tracking and localization, and intelligent transportation systems. By bridging theoretical insights with practical applications, this survey and tutorial provides valuable understanding and research directions in the field of distributed algorithm design for CPS, offering both newcomers and experienced researchers a comprehensive resource for addressing current challenges and future opportunities.

Read the full abstract

All authors Mohammadreza Doostmohammadian, Mahdi Shamsi, Hadi Zayyani, Nader Meskin, Hamid R. Rabiee, Sergio Pequito, Usman A. Khan

Social and Information Networks

Measuring Healthcare Accessibility and Resilience for Smart and Connected Rural Communities: A Florida Panhandle Case Study

Dahai Yu, Zhe He, Amber DeJohn et al. (+2 more)

Rural communities in the United States face persistent healthcare disparities, with fewer providers and facilities, longer trips to care, lower health literacy, and weaker transportation and broadband infrastructure than their non-rural counterparts. Beyond any single barrier, healthcare accessibility is shaped jointly by the geographic availability of services, residents' ability to reach those services, realized utilization, and the capacity of local systems to remain operational during disruptions, which are often examined separately and at coarse spatial scales, limiting their usefulness for community planning. We present a fine-grained, multi-source longitudinal measurement study of healthcare accessibility in Florida, with a focus on the hurricane-prone Florida Panhandle. We integrate healthcare-facility points of service, monthly mobility records from January 2018 through April 2021, Census Block Group (CBG)-level demographic data, and road-network data to compare rural and non-rural communities from supply, travel, utilization, and resilience perspectives. We find that 95.57% of the Panhandle's land area is rural and that 46.7% of rural CBGs contain no healthcare facility in the pooled inventory. Rural residents have less than half the per-capita facility availability of non-rural residents. These disparities remain after measured demographic adjustment, while longitudinal patterns reveal substantial disruptions around Hurricane Michael and COVID-19. Building on this evidence, we formulate an actionable planning framework that combines E2SFCA-based access estimation, stakeholder-weighted prioritization, and constrained deployment models for mobile clinics, non-emergency medical transportation, telehealth, and disaster-resilient services. The study demonstrates how community-oriented spatial intelligence can support equitable and resilient healthcare planning in rural regions.

Read the full abstract

All authors Dahai Yu, Zhe He, Amber DeJohn, Xinyue Ye, Guang Wang