Peer Revenue Play credits. Real papers.

Papers

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

Computer Science Artificial Intelligence

cs.AI Watch Artificial Intelligence

Thursday 27 August

Artificial Intelligence

Style as a Confound: False Positives in AI Detection of Non-Native Academic Writing

Hyeonchu Park, Gahye Jeong and Bugeun Kim

AI text detectors are increasingly employed in academic settings, but it remains unclear whether their outputs reflect AI authorship itself or broader linguistic features associated with polished academic English. Previous studies have reported high false-positive rates (FPRs) for non-native English writing, but population-level comparisons confound authorship with differences in topic, domain, and writing style. Professional editing provides a useful setting for examining this issue because it changes the linguistic form of manuscripts while preserving authorship and content. We examined 135,389 document pairs from a professional English editing service (2018-2025), comprising non-native manuscripts and their native-edited versions, to assess how editing affects detector responses controlling for content and authorship. For the 13 AI text detectors, FPRs for human-written texts varied widely, from 0.0% to 100.0%. Responses varied across detectors: the same edits increased AI scores in some detectors but decreased them in others. Notably, score changes correlated with the extent of editing. The findings identify professional editing style as a key confounding variable in AI detector outputs, rather than establishing a full separation of text origin from linguistic style, raising concerns about fairness and reliability in academic settings.
Read the full abstract

Artificial Intelligence

Accelerating Scientific Research with Gemini in the Real-World

Samuel Schmidgall, Xiaokai Zhu, Marian Shaw et al. (+32 more)

We present an extension and comprehensive real-world validation of Co-Scientist, a Gemini-based multi-agent system designed to accelerate end-to-end scientific research across hypothesis generation, experimentation, and manuscript generation. Moving beyond in silico hypothesis generation, this specialized configuration transitions Co-Scientist into an execution-grounded research partner advancing closed-loop scientific workflows across materials science, biology, and computer science. In materials science, Co-Scientist interfaced with a semi-automated chemical vapor deposition reactor to design a safe precursor route for MXenes; experimental execution produced a lamellar 2D material sharing key structural similarities with the Ti3C2Tx MXene lattice, although further experiments are needed to confirm the atomic structure. Leveraging Gemini 3 Deep Think for rapid, lab-in-the-loop execution, it also tailored growth recipes to laboratory constraints in minutes, enabling single-attempt growth of monolayer MoS2, MoSe2, and WS2 semiconductors. In biology, Co-Scientist predicted emergent swarming phenotypes of engineered E. coli across inducer (IPTG) gradients from sparse imaging data, quantitatively matching unpublished wet-lab morphological measurements. In computer science, Co-Scientist autonomously discovered an inference-time scaling architecture that outperformed six frontier models on HealthBench (Hard and Professional) while reducing potential clinical harm under blinded physician evaluation. Finally, a double-blind study of end-to-end generated papers with 30 domain experts across 450 reviews demonstrates that Co-Scientist's reliability modules reduce hallucination and plagiarism while improving research safety. Together, these results demonstrate progress toward closed-loop multi-agent scientific AI systems capable of accelerating real-world scientific discovery.
Read the full abstract

All authors Samuel Schmidgall, Xiaokai Zhu, Marian Shaw, Lin Yang, Valentin Liévin, Jingyun Yang, Yuchen Zhuang, Tim Strother, Alex Bijamov, Min Woo Sun, Anil Palepu, Justin Chen, David Steiner, Jacqueline Shreibati, Wei-Hung Weng, Yilin Zhao, Xingjian Hu, Nicholas Zahn, Sadhya Garg, Julia Kirby, Yuxiang Gan, Jiaoli Li, Divy Thakkar, Shekoofeh Azizi, David Racz, Juraj Gottweis, Vivek Natarajan, Chenglin Wu, Tal Danino, Keran Rong, Haozhe Wang, Benoit Schillings, Yong Cheng, Quoc V. Le, Tao Tu

Artificial Intelligence

Five Primitives for Governing Autonomous AI Agents at Runtime

Jiten Oswal and John Cadeddu

Enterprise deployments of autonomous AI agents inherit a control model built for human users and long-lived services, and the fit fails in three specific ways: agent principals are ephemeral, appearing and vanishing faster than provisioning; their actions are selected by a model rather than programmed, so the set of things they may attempt is not known in advance; and the population is discovered rather than provisioned, because anyone who can call an API can create one. We argue that governing such agents is a runtime problem – not a model-alignment problem and not a build-time problem – and we derive five primitives from the questions that must be answered before an action takes effect and after it has: discovery, identity, governance, attestation, and supply chain. For each we state what fails if it is absent and why the others cannot structurally supply it. We describe an implementation in which an agent's action is mediated against policy before it takes effect, authorised against a per-tenant action vocabulary, and recorded in a hash-linked signed ledger a third party can verify with the vendor out of the loop. We report what the architecture costs: the enforcement point sits on the request's critical path, identity requires a sidecar per workload, and fail-closed mediation converts availability incidents into denial. We are explicit about implementation status: four primitives are built and running in private pilots, and the fifth is built as separate tooling and not yet integrated into the request path. We keep it in the set deliberately: a five-part decomposition that exactly matches what its authors happened to build is not a taxonomy but a description of a codebase.
Read the full abstract

Artificial Intelligence

Relational Over-Regularization: Graph-Based AI-Generated Text Detection via Sentence Transition Deviation

Hyeonchu Park and Bugeun Kim

Detecting AI-generated text (AIGT) remains challenging because existing approaches rely on token-level statistical signals or independent stylometric features, causing them to overfit to specific generators and fail under distribution shift. We identify a structural signal at the sentence-pair level: LLMs produce inter-sentence transition variance that deviates from human writing through inflated variance driven by recurring similarity bursts at paragraph boundaries and templated transitions. We formalize this as Relational Over-Regularization (ROR) and validate it across four benchmarks (p < 0.001). The central contribution is this relational problem formulation, not a novel GNN architecture; CSFG is one concrete instantiation for operationalizing ROR. To exploit this signal, we propose the Cross-Source Stylometric Fingerprint Graph (CSFG), a graph-based framework that encodes positional, sequential, semantic, and transition deviation signals as learnable GNN edge features. The per-edge signed deviation {\delta}_ij operationalizes ROR without hand-crafted thresholds and acts as a false-positive calibrator. CSFG achieves 97.14% accuracy under binary detection, outperforming the strongest graph-based baseline by 11.14 pp, with a false-positive rate of 1.57% and robust generalization to unseen LLMs in the inflated-variance regime; detection degrades for generators whose transition variance falls at or below the human baseline.
Read the full abstract

Artificial Intelligence

SIGMA: Structured Noise-Effect-Aware Grouped Multi-Agent Aggregation

Li Mingqian

Cooperative multi-agent reinforcement learning (MARL) faces significant challenges in maintaining robust coordination under noisy observations. Although observation disturbances are often introduced independently across agents, their downstream effects on cooperative decision-making can become structured through underlying cooperation structures. We characterize this phenomenon as structured noise effects, where noise-induced decision effects exhibit local correlation among agents with stronger task-related dependencies while remaining globally heterogeneous across different agents and local structures. Existing robust MARL methods, however, rarely explicitly characterize or exploit such structure-dependent noise effects. To address this limitation, we propose SIGMA, a hierarchical collaboration framework that exploits cooperation structures to learn robust representations under noisy observations. SIGMA first organizes agents into adaptive local structures through density-based grouping and performs intra-group consensus aggregation to preserve shared task-relevant information while smoothing agent-specific representation deviations. Inter-group attention then adaptively integrates information across different groups to preserve global coordination while accommodating their heterogeneous contributions. Experiments on noisy-observation tasks in StarCraft II empirically validate the structured noise effects and demonstrate that SIGMA consistently improves robustness under observation noise while maintaining competitive performance in noise-free environments.
Read the full abstract

Artificial Intelligence · Cross-listed from cs.CL

FOCUS & RePAIR: Mitigating Text Degeneration via Token-Level Guidance for Pruned Large Language Models

Junyoung Lee, Sehyeon Park, Shinhyoung Jang et al. (+5 more)

Pruning is a practical approach to compress large language models (LLMs), but it can amplify text degeneration, especially repetition loops, even when perplexity and task accuracy remain largely unchanged. In this work, we present a token-level analysis of this failure mode by viewing decoding as a dynamical process that enters and persists in a small set of recurrent contexts. Our analysis decomposes degeneration into loop entry risk and loop persistence, and shows that persistence is controlled by the escape mass assigned to plausible alternatives within the token sampling set. Motivated by these findings, we propose two token-level guidance objectives for post-pruning fine-tuning. FOCUS reweights distillation toward high-confidence teacher regions to suppress leakage, while RePAIR uses onset-centered positive/negative continuation pairs with a margin loss to promote plausible alternatives and prevent early commitment to repetition loops. Experiments on open-ended continuation and instruction-based generation show that both methods consistently reduce repetition and improve generation quality.
Read the full abstract

All authors Junyoung Lee, Sehyeon Park, Shinhyoung Jang, Seonha Ryu, Hojeong Kim, Hyunsei Lee, Il Hong Suh, Yeseong Kim

Artificial Intelligence · Cross-listed from cs.CL

Do LLMs Understand Personality? Rethinking Persona Fidelity Evaluation through Structured Behavioral Inference

Mengfan Li, Zesheng Wei, Xuanhua Shi et al. (+1 more)

As large language models are increasingly deployed to simulate diverse human characters, ensuring persona fidelity, defined as the extent to which an agent's behavior consistently reflects the psychological and stylistic characteristics of a target persona, has become a critical requirement. However, existing evaluation paradigms primarily rely on either holistic LLM-based judges, which are prone to "holistic appraisal hallucination”, or static psychometric inventories, which fail to capture the context-dependent fidelity required in dynamic dialogue. To address these limitations, we propose PRISM (Persona Reasoning with Inverse SFL-based Modeling), a psycholinguistically grounded framework that reformulates persona fidelity evaluation as a structured inverse inference task. Inspired by Systemic Functional Linguistics (SFL), PRISM decomposes persona fidelity into three functional dimensions: Task Framing, Interpersonal Stance, and Linguistic Style. It estimates dimension-specific evidence over a persona-conditioned label space and aggregates these signals into an interpretable and auditable evaluation process. Experiments show that PRISM yields more accurate and stable judgements than traditional holistic judging, providing a more reliable framework for persona fidelity evaluation.
Read the full abstract

All authors Mengfan Li, Zesheng Wei, Xuanhua Shi, Yang Deng

Artificial Intelligence · Cross-listed from cs.CV

PailitaoGR: Latent Think-with-Images for Generative Image Retrieval

Xiaomeng Fan, Yueran Liu, Shengyu Zhou et al. (+6 more)

Generative retrieval has demonstrated strong performance by directly generating product semantic identifiers (SIDs). Extending this paradigm to image search, however, is nontrivial because real-world query images contain diverse information, including the search target, useful auxiliary evidence, and irrelevant visual content. This requires the model to identify and focus on the search target while selectively utilizing auxiliary evidence. In this paper, we propose PailitaoGR, a Latent Think-with-Images method for generative image retrieval, which internalizes target-focused perception and selective auxiliary-evidence utilization into a the generative retrieval model, enabling Zooming without Cropping and Reading without OCR. Specifically, we design a target-focused perception mechanism that identifies and enhances visual tokens of the search target, consisting of a target Enhancer and a learning strategy based on on-policy distillation and attention guidance loss, enabling the model to focus on search-target regions. We also design a selective auxiliary-evidence utilization mechanism that identifies and enhances visual tokens of auxiliary evidence, including an auxiliary enhancer and an in-capacity incremental contrastive distillation strategy, enabling the model to exploit auxiliary evidence. We construct training and validation sets sampled from real-world online image-search logs. Experiments show that our method outperforms existing baselines by an average of 13.8%, validating its effectiveness.
Read the full abstract

All authors Xiaomeng Fan, Yueran Liu, Shengyu Zhou, Chenghan Fu, Wanxian Guan, Feng Li, Chuan Yu, Jian Xu, Bo Zheng

Artificial Intelligence · Cross-listed from cs.CV

CoGeo-GS: Concept-Driven and Geometry-Aware Multi-Object Removal in 3D Scenes

Yuanxiang Ni, Xianliang Huang, Chenhang Ma et al. (+4 more)

Multi-object removal in 3D scenes is challenging due to severe occlusions, semantic entanglement, and the difficulty of maintaining geometric and multi-view consistency. Existing 3D Gaussian Splatting (3DGS) methods perform well for single-object editing but scale poorly to multi-object scenarios, often requiring repetitive optimization and yielding unstable geometry in removed regions. We propose CoGeo-GS, a concept-driven framework for controllable multi-object removal in 3D scenes. CoGeo-GS assigns concept-aware semantic tags to Gaussians, enabling flexible object selection and reducing interference between foreground objects and background structures within a single optimization stage. To recover plausible geometry, we introduce a geometry-aware completion pipeline that combines monocular depth priors with diffusion-based refinement and boundary-aligned blending. A geometry-regularized refinement strategy further stabilizes reconstruction and preserves multi-view consistency. Experiments demonstrate that CoGeo-GS outperforms existing methods in visual quality and reconstruction fidelity.
Read the full abstract

All authors Yuanxiang Ni, Xianliang Huang, Chenhang Ma, Chen Xiao, Yuewen Ma, Ruxin Wang, Hao Zhang

Artificial Intelligence · Cross-listed from cs.MA

Risks and Controls for Multi-Agent Systems: an analytical framework for deployment of AI agents across organisational boundaries

Alistair Reid, Simon O'Callaghan, Dustin Venini et al. (+2 more)

This report presents a framework to help organisations, policymakers and researchers reason about the risks that emerge when AI agents interact with each other, how those risks change as interactions cross organisational boundaries, and the controls that may help address them. As organisations deploy AI agents, those agents will increasingly interact with each other: inside the organisation, with the agents of partners, customers and suppliers, and with unknown counterparties on the open internet. Failures can emerge from the interactions themselves, and once those interactions cross an organisation's perimeter, no single organisation can fully see, control or govern them. The report introduces three deployment tiers, defined by the minimum common governance binding any two interacting agents: singular governance, where one organisation governs every agent; federated governance, where multiple organisations deploy into a shared environment under agreed rules; and open environments, where agents operate with no central authority and shared standards are adopted voluntarily if at all. Within each tier, the report examines risk factors, failure modes and available controls. It identifies who is positioned to apply the controls, and where no actor is positioned to act, it characterises the gap and the collective action required to close it.
Read the full abstract

All authors Alistair Reid, Simon O'Callaghan, Dustin Venini, Liam Carroll, Tiberio Caetano

Artificial Intelligence

AgentJudgeBench: A Multi-Difficulty Benchmark for Evaluating LLM Judges on Agentic Tool-Calling

Abhigya Verma, Amit Kumar Saha, Seganrasan Subramanian et al. (+1 more)

LLM judges are widely used to evaluate agentic tool-calling systems, yet their reliability on structured, dependency-driven workflows remains largely unexamined. We present AgentJudgeBench, the first benchmark to systematically study LLM-as-a-judge reliability for agentic tool-calling over workflow DAGs, as distinct from the broader LLM-as-a-judge task of open-ended text or preference evaluation. The benchmark comprises 3,808 instances spanning six DAG topologies and three difficulty tiers, evaluated with five generators (3B-70B open-weight models and GPT-5.4) and six judges (20B to frontier scale) under paired with- and without-ground-truth conditions. Judge alignment degrades monotonically with task difficulty, 1.5x faster without ground truth, and on hard queries without ground truth all six judges converge to a narrow 77-82% band regardless of scale, revealing a structural ceiling driven primarily by task difficulty, though its height is partly prompt-dependent for weaker generators, that model capacity alone cannot overcome. Ground-truth exposure is not uniformly beneficial: it reduces alignment for GPT-5.4 (1.5 pp) and Gemini-2.5-Pro (3.9 pp), consistent with over-anchoring. Among mitigation strategies, chain-of-thought reasoning and judge temperature both have negligible effect, while structured evaluation rubrics improve alignment by up to 6.5 pp but do not generalize uniformly across judge-generator pairs. With ground truth, QwQ-32B best matches the programmatic reference, while a human validation study identifies GPT-OSS-120B as the most human-aligned judge; without it, frontier judges lead only marginally within the shared ceiling. These results expose fundamental limitations of current LLM judges and yield practical guidelines for reliable evaluation in agentic systems.
Read the full abstract

All authors Abhigya Verma, Amit Kumar Saha, Seganrasan Subramanian, Sai Harshitha Aluru

Artificial Intelligence · Cross-listed from cs.LG

J-Zero: Unified Challenger–Solver–Judge Self-Evolution from Zero Data

Gyouk Chu, Myeongho Jeon, Teresa Yeo et al. (+1 more)

Self-evolving language models have recently emerged as a promising path toward superintelligence, with the advantage of reducing the cost of human supervision. While considerable progress has been made in verifiable domains, self-evolution in unverifiable domains remains less explored. We propose Judge co-adaptation from Zero data (J-Zero), a unified Challenger–Solver–Judge self-evolution framework that supports self-improvement across both domains. The Challenger and Solver co-evolve through an adversarial interaction: the Challenger generates increasingly difficult tasks, while the Solver learns to produce higher-quality responses to them. In parallel, the Judge co-adapts using preference pairs whose ordering is known in advance from how each response was produced, i.e., the Solver's answer over the Challenger's, and the Solver's decomposed-and-recombined answer over its one-shot answer, rather than from the Judge's own scores. J-Zero outperforms the baselines by an average of 4.2 points on verifiable and 8.0 points on unverifiable domains, and continues to improve through at least ten iterations, whereas the baselines degrade after two. Further analysis identifies Judge co-adaptation as the key driver of this sustained improvement.
Read the full abstract

All authors Gyouk Chu, Myeongho Jeon, Teresa Yeo, Eunho Yang

Artificial Intelligence · Cross-listed from math.NA

Physics-Informed Stochastic Configuration Machine: A Backpropagation-Free Neural Network with Fast Training for Nonlinear Differential Equations

Yuehao Song, Zhong Chen, Lihui Cen et al. (+2 more)

While Physics-Informed Neural Networks (PINNs) have emerged as a transformative paradigm for solving complex differential equations, their reliance on backpropagation-based gradient descent and automatic differentiation (AD) imposes significant computational bottlenecks and severe non-convex optimization challenges. To overcome these fundamental limitations, we propose the Physics-Informed Stochastic Configuration Machine (PI-SCM), a novel backpropagation-free framework for both forward and inverse problems in differential equations. The core mathematical contribution lies in the analytical evaluation of local Jacobians for nonlinear differential operators, which facilitates a linearized representation of the physical loss and projects it into a unified, linearized algebraic subspace. This reformulation allows for the explicit determination of optimal network weights via a sequence of generalized linear least squares solvers, effectively bypassing the iterative traps of traditional nonlinear optimizers. We develop a progressive algorithmic suite comprising localized construction (PI-SC-I), sliding-window updating (PI-SC-II), and global updating (PI-SC-III), and rigorously establish their universal approximation properties. Extensive experiments demonstrate that PI-SCM achieves high-fidelity predictive accuracy and robust parameter identification while accelerating the training process by orders of magnitude compared to standard PINNs. Our work provides a highly efficient and scalable foundation for next-generation, real-time Scientific Machine Learning applications.
Read the full abstract

All authors Yuehao Song, Zhong Chen, Lihui Cen, Liang Wu, Kai Zhang

Artificial Intelligence

DuMateBench: Evaluating Autonomous Agents in Complex Real-World Workflows

Zechun Niu, Yukun Zhao, Jiaxin Zhang et al. (+12 more)

Autonomous agents are increasingly adopted to complete complex, multi-tool workflows in real-world settings. However, existing benchmarks typically separate tasks by application or capability and evaluate agents in environments that are cleaner and more stable than those encountered in practice. We introduce DuMateBench, a real-session benchmark reconstructed from anonymized and privacy-screened user sessions collected from a large-scale production agent platform. Each task preserves the relevant pre-solution interaction history, persistent configurations, and workspace state, and is then validated through human verification. The resulting benchmark comprises 200 tasks spanning 8 broad scenarios and 17 fine-grained capability categories, with most tasks requiring multiple capability coordination. We execute these tasks in isolated Docker containers injected with three forms of real-world environmental complexity: Insufficient, Unstable, and Noisy, and assess performance using a hybrid deterministic and LLM-as-Judge evaluation protocol. Experiments across five representative autonomous-agent frameworks paired with four state-of-the-art LLMs reveal substantial gaps in strict task completion. Complementary robustness, efficiency, and diagnostic analyses further show that performance under environmental perturbations is jointly shaped by the capabilities of the LLM and the surrounding agent framework. The code and data are publicly available at https://dumatebench.com/.
Read the full abstract

All authors Zechun Niu, Yukun Zhao, Jiaxin Zhang, Xu Shen, Jinhua Si, Han Tian, Can Xu, Yunfan Song, Jiaxin Mao, Yansong Gao, Yuchen Li, Jianmin Wu, Lingyong Yan, Shuaiqiang Wang, Dawei Yin

Artificial Intelligence

Multi2AV-Safety: Benchmarking Safety in Multimodal-to-Audio-Video Generation

Kaichao Jiang, Changtao Miao, Baiqi Wu et al. (+8 more)

Recent audio-video generators increasingly support joint conditioning on text, images, audio, and video. These capabilities also enable attacks that exploit cross-modal interactions or obscure harmful intent to bypass safeguards and induce harmful audio-video outputs. However, existing generation-safety benchmarks have not kept pace with these advances, providing limited coverage of multimodal input combinations and obscured attack intents. To address these gaps, we introduce Multi2AV-Safety, the first full-coverage red-team benchmark for multimodal-to-audio-video generation, comprising 11,024 attack instances across all 11 non-singleton T/I/A/V conditioning configurations, 4 attack-intent categories, and 5 harm categories. Our evaluation of recent state-of-the-art models, including four multimodal-conditioned audio-video generators and eight safety guards, reveals substantial vulnerabilities in both generation and safeguarding, with multimodal compositional risk and obscured attack-intent risk emerging as two complementary challenges. Guided by these findings, we introduce PerceptGuard, an omni-modal guard integrating compositional-risk and attack-intent supervision through structured risk perception learning. By jointly training rationale generation and safety classification, it learns shared risk representations that enable a safety head to make efficient predictions at inference without rationale decoding, while retaining the ability to generate explanations on demand. Across 34 safety benchmarks, PerceptGuard combines SOTA multimodal safety detection with highly competitive unimodal performance, strengthening input-side safeguards against multimodal attacks on omni models. In particular, it improves safeguarding against the above risks, achieving an overall recall of 86.06% on Multi2AV-Safety and outperforming GuardReasoner-Omni by 14.56%.
Read the full abstract

All authors Kaichao Jiang, Changtao Miao, Baiqi Wu, Zhiyuan Lu, Kang Yang, Peiwei Zhao, Junchi Chen, Yunfeng Diao, He Liu, Qi Chu, Tao Gong

Artificial Intelligence

PILOT in the Loop: Live Self-Improvement for Long-Horizon Agents

Yang Xiao, Yusong Sun, Haoyi Wu et al. (+7 more)

Long-horizon agent runs generate experience that can improve both the current run and future work. Most self-improvement methods process this experience only after execution ends, so they cannot redirect the active run or immediately apply and validate lessons learned from it. We argue that self-improvement should instead be live, using emerging experience both to redirect the active run and to update the persistent harness. Existing agent architectures do not fully support this goal. Single-agent self-correction combines task execution and trajectory assessment within one context, while subagent delegation separates execution but typically cannot redirect an active subagent. We present PILOT, a supervisor-worker harness for live self-improvement through two coupled mechanisms: (1) live steering lets a separate supervisor redirect or abort the active worker during execution; and (2) live self-evolution distils procedures and failure modes revealed during execution into reusable skills and memory. Across two frozen backbones and three benchmarks, PILOT ranks first in five of six configurations. On Terminal-Bench 2.0, PILOT outperforms counterpart harnesses by up to 9.8 percentage points. In the self-improvement setting, PILOT gains 14.6 points with GLM-5.1 and 12.4 points with Kimi-K2.6. Mean output tokens fall by 42.9% and 47.4%, while successful evaluations per million output tokens rise by 110.3% and 134.0%, respectively.
Read the full abstract

All authors Yang Xiao, Yusong Sun, Haoyi Wu, Wenyang Hui, Wen Da, Zhaokai Luo, Mu Chuan, Yao Hu, Wenjie Li, Chengyue Jiang

Artificial Intelligence · Cross-listed from cs.RO

RTNav: Towards Real-Time Zero-Shot Object Navigation

Easop Lee, Lingyu Zhang and Boyuan Chen

Navigation in unknown environments to find unforeseen objects has become increasingly feasible with capable vision and language foundation models. However, these models also introduce non-negligible inference latency, which becomes an important concern when agents must operate continuously in the real world. Most state-of-the-art methods are still developed in synchronous simulators, where the environment waits for the agent to act and inference time is effectively free. As a result, agents are often designed around the sequential execution of perception, reasoning, and action, with little regard for time constraints. Under real-time execution, where wall-clock time counts towards the task budget, the inefficiencies of these architectures become clear. We show that recent zero-shot object navigation methods suffer consistent performance degradation under such realistic timing conditions. Motivated by this observation, we propose RTNav, a simple but effective architecture that treats inference latency, asynchronous environment stepping, and bounded compute as explicit design considerations. Evaluated on real-time variants of HM3D-v1, HM3D-v2, and HM3D-OVON, RTNav improves the success rate by up to 11% and the Success weighted by Completion Time by up to 5.1 points over prior work.
Read the full abstract

Artificial Intelligence · Cross-listed from cs.MA

GVS5H: Zero-Shot Self-Orchestration with Ledger-Based Control for Improved LLM Coding Performance

Victor Gao, Vida Khosrowshahi, Ali Khosrowshahi et al. (+3 more)

Frontier coding performance is typically attained with large, costly proprietary models. We introduce ledger-based zero-shot self-orchestration (GVS5H), a training-free method in which fresh instances of one model decompose problems and coordinate through a shared file system. Across eleven open and closed-weight models on the 100 latest hard LiveCodeBench problems, the method yields as much as 25.6 points improvement, boosting several cheaper models to frontier-level performance. Orchestrated Qwen3.8 Flash Next scores 93.0% against Fable 5's 90.4% at 9% of the cost, while the smaller Qwen3.8-27B reaches 92.4%. Gains are not universal: some models are unchanged or worse. Transcript analysis attributes the gain to decomposition and persistent context. Inference-time organization can reach or exceed frontier coding accuracy at a fraction of the cost on self-hostable weights.
Read the full abstract

All authors Victor Gao, Vida Khosrowshahi, Ali Khosrowshahi, Xihao Sun, Juhyun Lee, Simon Lee

Wednesday 26 August

Artificial Intelligence

Is Reasoning Always Useful? Rethinking Reasoning Utility in Universal Multimodal Embeddings

Wenxiao Fan, Jingling Fu, Luohang Liu et al. (+6 more)

Reasoning-enhanced universal multimodal embeddings (UME) improve heterogeneous retrieval, but plausible rationales do not necessarily produce discriminative rankings. We study this gap by comparing the discriminative (DISC) and reasoning-driven generative (GEN) branches of UME-R1, a state-of-the-art reasoning UME method. We decompose reasoning utility into positive-target gain, hard-negative gain, and their margin difference. Positive similarity increases for 56.6%, but 15.7% are false-helpful cases where reasoning moves hard negatives closer even more. Local-neighborhood and token-attribution diagnostics suggest why: reasoning often de-condenses retrieved neighborhoods, but utility requires separator-aligned movement, while influential CoT tokens frequently encode evidence shared by positives and hard negatives. Motivated by these diagnostics, we propose SURE (Score-structure Utility Router for Embeddings), which improves UME-R1-7B by 1.5 points and yields consistent gains on two additional embedding models on MMEB-V2, without retraining, label-based policy selection, or extra VLM forward passes.
Read the full abstract

All authors Wenxiao Fan, Jingling Fu, Luohang Liu, Xinyuan Shan, Lichen Ma, Yu He, Junshi Huang, Yan Li, Kan Li

Artificial Intelligence · Cross-listed from cs.CV

UNWIND: Any-Length Facial Video for Stress Detection without Temporal Windowing

Stefanos Gkikas, Christian Arzate Cruz, Eric Nichols et al. (+2 more)

Automatic stress recognition from facial video provides a non-contact approach for affective monitoring. However, most existing video-based methods divide complete recordings into shorter temporal segments before performing classification. Such segmentation requires additional decisions concerning segment duration, overlap, and prediction aggregation, and may restrict the model from exploiting information distributed across the entire recording. We introduce UNWIND, a facial-video framework for stress detection that analyzes a complete recording as a single model input, eliminating the need for temporal windowing or external segmentation. UNWIND reorganizes the video by folding its temporal dimension into the channel dimension of a two-dimensional spatial representation, which is subsequently processed through a unified asymmetric-attention architecture. With a temporal stride of $\tau=1$, the framework processes the entire $120$-second sequence, corresponding to $3{,}600$ frames sampled at $30$~fps, in a single input. We evaluate seven temporal-stride settings on a stress dataset comprising $58$ subjects, using a stratified subject-level protocol that covers configurations from dense frame retention to sparse temporal sampling. The highest test accuracy, $70.02\%$, is obtained at $\tau=15$, while processing all frames at $\tau=1$ achieves a comparable accuracy of $69.73\%$. Computational requirements range from $12.48$ to $348.78$ GFLOPs across the evaluated stride settings, illustrating the balance between temporal sampling density and computational efficiency. The findings show that effective facial-video stress recognition can be achieved without dividing recordings into temporal windows and that complete-recording inference can be performed within a single unified model.
Read the full abstract

All authors Stefanos Gkikas, Christian Arzate Cruz, Eric Nichols, Giorgos Giannakakis, Randy Gomez

Artificial Intelligence

HiPACE: Hierarchical Phase-Boundary Analysis and Controlled Evaluation of Feature Absorption in Sparse Autoencoders

Jinyuan Zhang, Peng He, Yin Yuan et al. (+2 more)

Sparse autoencoders (SAEs) decompose LLM activations into sparse dictionary atoms, so that each distinct concept gets its own feature. One recurring behavior complicates this premise: feature absorption, in which a parent concept and its children–fruit and {apple, banana, pear}, say–collapse into a shared family direction. Prior work documents absorption empirically; missing is a closed-form prediction of when the shared direction is the cost-optimal representation of an active semantic family. This paper closes that gap. For a hierarchical Bernoulli generator with $k$ active children and residual scale $\alpha$, the $L_0$-penalized reconstruction objective admits a closed-form phase boundary $\lambda_c(k,\alpha)=\alpha^2 k/(k-1)$: above it, pure parent absorption is strictly cheaper than pure child coding. Building on this boundary, we introduce HiPACE, an evaluation protocol that tests the boundary's structural consequence in real SAE dictionaries–measuring parent–child decoder structure over WordNet families, freezing the discovery-selected statistic before testing on unseen families, and contrasting genuine families against randomized sibling nulls. The boundary proves sharp in its native regime, predicting the synthetic transition within $\pm15%$ on all 30 tested cells. In Pythia-160m SAEs, the parent–child decoder gap recovers the predicted ordering with partial correlations up to $-0.93$ that sustain on the locked holdout and exclude sibling nulls ($p=0.002$). Controlled activation composition connects the theory's active-child count to the recovered family directions, and residual-stream interventions show that signed family directions increase parent-category logits, reversing under sign flip and vanishing under random controls–establishing causal sufficiency at the family-subspace level.
Read the full abstract

All authors Jinyuan Zhang, Peng He, Yin Yuan, He Hu, ShengShuo Jiao

Artificial Intelligence

Delay-of-Gratification as a Multi-Agent Survival Micro-benchmark for Long-Horizon LLMs: Social Exposure, Personas, and Tool Use Budgets

Olga Manakina, Igor Bogdanov and Chung-Horng Lung

Large language models (LLMs) are increasingly deployed as multi-turn agents that must sustain goals, use tools, and adapt to other agents over extended interactions. However, existing research lacks auditable, multi-turn, multi-factorial experiments that quantify LLM behavior under explicit constraints, with time-resolved statistics that reveal how behavior unfolds over long horizons. To address this gap, we develop a multi-agent micro-benchmark inspired by the Stanford marshmallow experiment: ReAct agents operate minute-by-minute with a "raise a question" tool under a per-step budget, while we factorially manipulate social context (broadcast vs. isolated), personas (age, hedonic drive), and metacognitive policy (mandatory vs. optional tool use). We analyze outcomes with Kaplan-Meier (KM) survival curves and discrete-time hazard models over a long risk horizon across 19,200 agent trajectories in 64 cells. Behavior shows a sharp early "eat" impulse, and only 75.9% of agents persist to the end. In a discrete-time hazard model, isolation reduces per-minute risk relative to broadcast, whereas a must-use self-questioning policy increases risk. On average, agents ask $\approx 7.12$ questions and hit the per-step budget in $\approx 6\%$ of minutes. Questioning declines faster under broadcast than isolation. Ablation experiments demonstrated that removing hedonic drive and/or persona age increases survival and completion, narrows the broadcast/isolated gap, but leaves the must vs. may ordering intact. The combined ablation (no hedonic + no persona age) yields the highest completion (approaching $1.0$). These results establish delay-of-gratification as a compact, multi-turn interaction benchmark that captures social contagion and tool-use dynamics in LLM agents, providing a reproducible testbed and statistics for analyzing long-horizon, multi-agent behavior.
Read the full abstract

Artificial Intelligence

Evaluation of Multi-Turn Consistency in LLM Agents: Survival Analysis and Failure-Rationale Taxonomy

Igor Bogdanov, Olga Manakina and Chung-Horng Lung

Large language model (LLM) agents may perform well on isolated tasks yet drift into inconsistency over extended interaction. We evaluate temporal consistency in a controlled 20-step multi-agent setting inspired by delayed-gratification studies. At each step, an agent chooses between continuing to delay a reward or claiming it immediately (terminating the episode). Across a full-factorial manipulation of social visibility (private vs public), persona stressors, and deliberation policy, we run 84,540 trajectories spanning 8 model families. Treating the first reward-claim as a time-to-event outcome, we estimate Kaplan-Meier survival curves and fit discrete-time hazard regression to quantify how experimental factors shift failure risk over time. Then, to analyze rationales and language patterns associated with failure, we build a seven-category taxonomy from 13,780 deliberation traces from agents who choose to terminate the episode, using an LLM-assisted labeling paired with human audit ($\kappa=0.83$). Rationale profiles change systematically with time and context: early failures are more impulse-driven, later failures more fatigue- and cost-benefit-framed, while public settings increase norm-oriented justifications. We also find a deliberation-inconsistency association: among failures, longer deliberation correlates with higher rates of intra-rationale contradiction (simultaneous pro-delay and pro-claim statements), challenging the assumption that more reasoning text implies greater consistency. Together, the survival and rationale analyses reveal distinct temporal reliability regimes and model-specific "failure fingerprints", offering an evaluation lens for diagnosing inconsistency in multi-turn agent behavior.
Read the full abstract

Artificial Intelligence

The Architect, the Adversary, and the Judge: Closed-Loop Generation of Standards-Aligned Assessment Items at Scale

Wenhui Chen, Ziyao Lin, Jianlin Chen et al. (+2 more)

We present CLAIM, a production pipeline for K-12 assessment-item generation coupling a two-stage generate-then-attack protocol (the model drafts as a "curriculum architect", then re-enters the same conversation as a hostile adversarial reviewer), bi-directional few-shot conditioning on accepted and rejected items, the latter carrying the evaluator's diagnosis, and a knowledge dictionary of 44,844 error-correction rules mined from that feedback and retrieved per standard and item type. Across 43,227 scored items over 755 Common Core ELA standards, three item types, and ten LLMs, the pipeline reaches a 97.8% expert-evaluator pass rate on a 9,074-item production run. We then ask what that rate certifies. Re-scoring a stratified sample with three judges from other vendors, blind to the deployed verdict, reproduces the format ordering under every judge and recovers a larger open-set deficit than the deployed evaluator does; but agreement on the accept/reject binary is weak at production prevalence (kappa about 0.13), and the judges agree with each other no better. The level is therefore judge-relative, and with no student-response data our quality evidence is evaluator-judged throughout. The corpus also exposes a robust asymmetry. Multiple-choice and multiple-select generation saturate at 98% or above for both frontier models under a dozen static rules, whereas fill-in-the-blank generation is capability-tiered (82.8-96.7% across five models under a matched rule set, standards, and judge) and plateaus under prompt-only optimization, with error mass shifting between answer-key over-inclusion and omission as rules accumulate. We analyze this as open-set boundary determination, a task autoregressive decoders are structurally ill-equipped to solve, and show the asymmetry recurring when the evaluator itself is distilled: fail-recall rises from 8% to 63% while F1 saturates at 0.25.
Read the full abstract

All authors Wenhui Chen, Ziyao Lin, Jianlin Chen, Peiji Long, Chi Man Vong

Artificial Intelligence · Cross-listed from cs.LG

Multiple latent orderings better predict language model preferences

Aviral Chawla, William H.W. Thompson and Jean-Gabriel Young

Language models are frequently employed in settings where they are asked to make value judgments and choices. These observed choices often exhibit intransitivity: A model may prefer item $A$ to $B$ and $B$ to $C$, while also preferring $C$ to $A$. Existing work that models LLM preferences treats such inconsistencies as sampling noise around a single latent ordering. We instead propose that intransitivity reflects the aggregation of multiple latent, internally consistent orderings. We first show that observed inconsistencies cannot be explained by a single ordering under any monotone link function. We then introduce a noise-augmented mixture Bradley-Terry (MBT) model that infers latent preference components from repeated pairwise comparisons. Across seven models and four tasks, a mixture of orderings often explains structural inconsistencies better than single-utility models. We find that aggregate preferences often hide underlying preference heterogeneity. A case study on Moral Machine dilemmas shows that models which disagree on aggregate orderings can still share latent components. Together, these results suggest that LLMs reflect plural preferences. Alignment and evaluation pipelines that treat LLM preferences as a single function, therefore, risk averaging over coherent orderings that different users may endorse differently.
Read the full abstract

Artificial Intelligence

Didactic knowledge or Clinical Cases? How Data Types Shape Medical Large Language Models

Yuzheng Fan, Haochun Wang, Sendong Zhao et al. (+3 more)

Medical large language models are commonly trained on mixtures of didactic data (e.g., textbooks) and clinical data (e.g., patient records), yet how these data types differentially shape model capabilities remains unclear. We address this issue with token-matched experiments that vary the didactic-to-clinical ratio and analyze how data composition affects performance, capability profiles, and error patterns across knowledge-intensive and clinic-oriented tasks. We uncover an asymmetric transfer across task types: clinical data improves clinic-oriented tasks while remaining competitive on knowledge-intensive ones, whereas didactic data mainly improves knowledge-intensive tasks. Error analysis suggests a knowing-doing gap, where improvements in knowledge recall do not reliably generalize to clinical reasoning. We further observe that modest amounts of clinical data yield most of the gains on EHR-grounded tasks, while the optimal mixture ratio varies with the knowledge and clinical reasoning demands of downstream tasks. These findings suggest that medical LLM data curation should be application-driven, with higher proportions of clinical data preferred for reasoning-intensive use cases.
Read the full abstract

All authors Yuzheng Fan, Haochun Wang, Sendong Zhao, Xiao Han, Ming Ma, Bing Qin

Artificial Intelligence · Cross-listed from cs.CL

Multilingual Safety Signals Are Multi-Layered: Filtering Safety-Degrading Data for Safer LLMs

Jiakun Li, Guowei Song, Sijia Li et al. (+3 more)

Preserving safety alignment during large language models fine-tuning is critical, however, recent studies have demonstrated that even benign fine-tuning data may contain safety-degrading samples that silently undermine safety alignment. Existing approaches typically identify such samples using representations from a single safety-sensitive layer. While this assumption has shown effectiveness in monolingual settings, its validity for multilingual models remains unclear due to potential cross-lingual differences in representation patterns. Through a cross-lingual analysis, we show that sensitive layers are only partially shared across languages, with safety-relevant signals often distributed across multiple layers. Motivated by these observations, we propose MMSAFE, a multi-layer framework for multilingual safety-degrading data identification that captures both shared and language-specific safety signals. Extensive experiments across multiple models, languages, and safety benchmarks demonstrate that MMSAFE reduces the average harmful-response ratio by 60% compared with random filtering and achieves stronger average performance than the strongest single-layer baseline, demonstrating the effectiveness of multi-layer modeling for robust multilingual safety alignment.
Read the full abstract

All authors Jiakun Li, Guowei Song, Sijia Li, Xingwei He, Hongzheng Chai, Yuan Yuan

Artificial Intelligence · Cross-listed from cs.CV

Defending Wearable VLMs Against Private Attribute Inference

Zhimin Li, Pan Wang, Jingxian Chen et al. (+4 more)

Wearable VLM pipelines promise continuous multimodal assistance from egocentric visual capture: a user asks a task-driven question about the surrounding scene, and the system uses compact visual tokens to support language reasoning. The challenge motivating this work is that the same egocentric evidence needed for useful assistance can also reveal private attributes about the wearer or nearby bystanders. We investigate this as a joint privacy-utility problem for split VLM inference, where visual encoding occurs within a trusted device boundary but intermediate visual tokens may be transmitted to downstream reasoning components. This exposes an understudied leakage surface: even when final textual responses are benign, external attackers or untrusted downstream components can recover private attributes from transmitted visual tokens. To evaluate this tension, we construct a paired privacy-utility benchmark with 3,221 image-question records, each paired with a utility question and privacy labels covering location, income, sex, and interests. We further propose Token-Guided Attribute Privacy (TGAP), a pre-LLM token disentangler that learns a residual transformation of visual tokens before they leave the trusted boundary. TGAP combines utility preservation, identity regularization, semantic privacy suppression, and image-driven representation suppression, avoiding the utility loss caused by coarse hard or attention masking. On the benchmark used for source-model evaluation, TGAP reduces privacy accuracy from 56.7% to 7.4%, a 49.3% absolute drop, while maintaining relaxed utility at 74.4%. These results suggest that securing the compact token interface is a practical path toward privacy-preserving wearable multimodal AI.
Read the full abstract

All authors Zhimin Li, Pan Wang, Jingxian Chen, Yuantao Tang, Anthony Chen, Qian Lou, Jingtong Hu

Artificial Intelligence · Cross-listed from cs.LG

Diff Mining: Logit Differences Reveal Finetuning Objectives

Greg Kocher, Robert West, Clément Dumas et al. (+1 more)

Finetuning has become the gold standard for refining existing behaviors and inducing new ones in language models, yet it often remains unclear exactly which behaviors emerge during this process. As models grow ever more capable, understanding finetuning better becomes increasingly important, particularly since unwanted behaviors may arise during finetuning. In this paper, we introduce Diff Mining, a simple yet effective framework for identifying what a finetuned model has learned by comparing its logits to those of its base model. Diff Mining effectively surfaces salient tokens that are amplified in the finetuned model, serving as a fingerprint of its training – even on text unrelated to the finetuning domain. Unlike many existing model diffing methods which require model internals, Diff Mining only needs access to output logits and scales to large models. The framework consists of two modular stages: (i) extracting per-context logit differences between the finetuned and base models on a reference corpus, and (ii) aggregating the resulting signals to construct an interpretable token set representing the finetune. For aggregation, we explore both a simple Top-K frequency method and a Non-negative Matrix Factorization (NMF)-based approach for disentangling multiple finetuning objectives into distinct token clusters. Empirically, Diff Mining succeeds across diverse settings: on finetune domain detection, it significantly outperforms state-of-the-art model diffing methods both in identifying relevant tokens and in downstream performance when an interpretability agent is given access to the extracted token set; on models with injected biases, it identifies more than one third of the biases without targeted probing. Overall, our framework shows promise in developing auditing tools to detect finetuning objectives.
Read the full abstract

All authors Greg Kocher, Robert West, Clément Dumas, Julian Minder

Artificial Intelligence

Don't Overthink, Don't Underthink: Toward Adaptive Reasoning in Agentic AI

Md Jueal Mia and M. Hadi Amini

Recent advances in Large Language Models (LLMs) have shown that increased inference-time reasoning can improve performance on complex tasks. However, many existing approaches rely on fixed or preallocated reasoning controls, such as fixed token budgets, pre-execution difficulty estimates, or activation-space interventions, and are often evaluated on standalone reasoning benchmarks rather than full agentic workflows. These assumptions may not hold in agentic AI systems, where reasoning requirements evolve dynamically through planning, tool use, memory retrieval, and agent-to-agent interactions. Consequently, reasoning can become either excessive or insufficient, resulting in unnecessary computation, increased latency, planning drift, excessive tool use, or incomplete solutions. We argue that a major challenge for next-generation agentic AI is not merely how much reasoning a language model should perform, but how it should allocate reasoning according to evolving task demands. We characterize over-reasoning and under-reasoning as recurring failure modes of misallocated reasoning and evaluate them on MATH-500 and the GAIA public validation benchmark. Using tool-decision latency, token consumption, token-limit exhaustion, and answer correctness, our results suggest that cases classified as over-reasoning are associated with higher computational cost without proportional accuracy gains, whereas cases classified as under-reasoning are consistently associated with incorrect or incomplete solutions. These findings motivate future research on adaptive reasoning mechanisms for agentic AI.
Read the full abstract