Artificial Intelligence
Open-Weight Masked Introspection: Measuring What Language Models Can Report About Their Own Computation
Emilio Ferrara
Are frontier models able to introspect about their internal states? Recent work suggests that under certain conditions a complex enough model can audit its own internals, call out what changed, and report back confidently about it. We tested that claim on eight open-weight models from seven families and found no such ability: asked whether their own computation had been altered, none answered better than chance. To test it we built Open-Weight Masked Introspection (OWMI), a framework that intervenes on residual-stream sites,
attention heads and sparse-autoencoder features, then interrogates the model about the change against the null conditions an answer has to beat: sham runs where nothing was altered, impact-matched random perturbations, and a text-only observer that sees only the visible output. Over 78,000 measurements, no model's report discriminates a real intervention from a sham beyond chance (AUROC ~0.5007), and an equivalence test bounds the effect below 0.15 percentage points of AUROC. Surprisingly, all the information needed is in the models. A model fine-tuned to report this class of intervention reaches near-perfect recovery on held-out directions, and a linear probe recovers intervention presence from the same activations at 75% to 95.8% accuracy, sharpening to no held-out error at the last layer before the model speaks. In one model the signal surfaces in the confidence rather than the words: its yes-or-no report never varies, while the confidence attached to it separates intervention from sham at AUROC 0.647. The failure sits in the path from internal state to verbal report, so oversight that reads a model's own testimony needs validating against an internal reference. While our results show the inability of current open-weight models to introspect, the debate is not settled for future models.
Read the full abstract ▾
Artificial Intelligence
Consilience: Conformally Calibrated Communication Control for Hidden-Profile Multi-Agent Reasoning
Abhijith Babu, Ramneet Kaur, Vishal Pramanik et al. (+7 more)
Multi-
agent LLM systems can improve
reasoning by pooling diverse perspectives, but their effectiveness depends on coordinating communication, particularly in hidden-profile settings where each
agent holds only part of the evidence required for a correct decision. Existing protocols, including fixed schedules, round-robin exchange, and unstructured debate, provide no guarantee that a conversational action is appropriate. We propose Consilience, an inference-time orchestration framework that both steers and certifies multi-
agent communication under distributed private information. At each turn, Consilience summarizes the discussion using a compact state capturing uncertainty, disagreement, evidence gain, redundancy, and premature consensus, then selects both a communication intervention (challenge, clarify, seek evidence, or route) and an appropriate speaker. Its central contribution is a round-wise conformal calibration procedure that provides a distribution-free, finite-sample guarantee: at each discussion round, conditional on reaching that round, the one-step regret of a controller's proposed action is bounded by a calibrated threshold with marginal probability at least 1 - alpha; an acceptance mechanism enforces the same guarantee for the executed action by replacing inadmissible proposals. On HiddenBench-style hidden-profile tasks spanning 12 open and closed weight
language models, Consilience improves decision accuracy and communication efficiency over fixed and unstructured discussion protocols, sometimes surpassing a full-information baseline where every
agent observes all evidence. These results demonstrate that certified adaptive communication control can be more valuable than increasing information availability, providing a practical mechanism for reliable multi-
agent LLM coordination.
Read the full abstract ▾
All authors Abhijith Babu, Ramneet Kaur, Vishal Pramanik, Olivera Kotevska, Nathaniel D. Bastian, Susmit Jha, Sunny Raj, Yanzhao Wu, Sumit Kumar Jha, Anirban Roy
Artificial Intelligence · Cross-listed from cs.CR
Beyond End-to-End Success: Diagnosing Failures in Long-Horizon Security LLM Agents
Wei Shao, Chongzhou Fang, Zuxiong Tan et al. (+4 more)
Long-horizon security LLM
agents must carry information and decisions across many dependent interactions, where later actions often depend on services, state, or access discovered much earlier. This makes final task success difficult to interpret: an
agent may fail before it ever reaches the point where the capability of interest can be exercised. We present a diagnostic methodology that instruments security tasks with checkpoints, separates failures before and after capability exposure, and uses controlled interventions to test suspected upstream bottlenecks. We evaluate the methodology across four task families involving delayed reuse of discovered information, reuse of observed state, recovery from failed strategies, and decision making after uncertain outcomes. On observed state reuse, checkpoint analysis shows that many Gemini 2.5 Flash failures occur before the model observes the state it is later expected to reuse. In a pre-specified 92-seed study, targeted protocol-disambiguation guidance increases state observation from 65.5% under a matched non-guidance control message to 95.4%. Repeating the same design with Gemini 3.7 Flash produces the opposite effect, while state observation no longer reliably predicts task completion. These results show that the dominant source of failure can shift across model generations, motivating evaluation that diagnoses where and why long-horizon security
agents fail rather than relying only on aggregate task success.
Read the full abstract ▾
All authors Wei Shao, Chongzhou Fang, Zuxiong Tan, Zequan Liang, Setareh Rafatirad, Avesta Sasan, Houman Homayoun
Artificial Intelligence · Cross-listed from eess.IV
Consistency Models for Fast MRI Reconstruction Using Regularization by Denoising
Merve Gülle, Junno Yun, Yaşar Utku Alçalar et al. (+1 more)
Diffusion models (DMs) have emerged as powerful generative priors for MRI reconstruction with promising results. Yet DM-based methods require extensive iterative refinement, limiting their practical deployment. Consistency models (CMs) provide a compelling alternative, aiming to map out the
diffusion trajectory in a single pass, enabling faster generation. In this work, we propose CM-RED, a novel MRI reconstruction method that integrates a pretrained CM into the regularization by denoising (RED) scheme. Our method builds on accelerated proximal gradient RED (RED-APG), and further incorporates controlled noise injection during the update steps to enhance generative diversity and accelerate convergence. Extensive experiments on the fastMRI knee and
brain datasets demonstrate that CM-RED achieves high-quality reconstructions across multiple anatomies, contrast weights, acceleration factors, and undersampling patterns, using only 4 network function evaluations (NFEs). The proposed method consistently outperforms existing DM- and CM-based approaches in both quantitative metrics and visual fidelity, and exhibits strong robustness to hyperparameter variations, highlighting CM-RED as an efficient and effective generative framework for accelerated MRI reconstruction. The source code and pretrained models are publicly available at https://github.com/MerveGulle/CM-RED.
Read the full abstract ▾
All authors Merve Gülle, Junno Yun, Yaşar Utku Alçalar, Mehmet Akçakaya
Artificial Intelligence
Volumetric Radiology AI in the Era of Multimodal Large Language Models
Zanting Ye, Shengyuan Liu, Xin Liu et al. (+16 more)
Advances in
multimodal large
language models (MLLMs) are extending radiological artificial intelligence (AI) beyond task-specific image analysis toward
multimodal understanding and
reasoning. Volumetric radiology, however, presents a fundamental representational mismatch: clinical interpretation often requires full-volume spatial context and acquisition-dependent quantitative information, whereas current MLLMs are commonly conditioned on selected two-dimensional (2D) images, compressed visual representations, or report-derived text. Reliable volumetric radiology AI therefore requires representations that preserve task-relevant three-dimensional (3D) information and systems that can access, verify, and integrate this information across clinical workflows. In this Review, we examine more than 200 publications through July 2026. We organize the literature around volumetric representation and
multimodal understanding at the model level, agentic orchestration at the system level, and their links to clinical applications and evaluation. We review volumetric foundation models,
language alignment and compression strategies, and agentic systems that extend MLLMs through planning, tools,
memory, and workflow interaction. We distinguish settings in which selected 2D views or report-mediated
reasoning may suffice from those that warrant native volumetric modeling. We also introduce a Claim-Design-Validation framework to assess whether technical, workflow, and clinical claims are matched by appropriate design and validation. Across the literature, native volumetric modeling and agentic capabilities depend on the spatial, quantitative, contextual, and workflow requirements of the intended task. Clinical credibility requires faithful volumetric representation, traceable system behavior, claim-aligned validation, and clearly defined human oversight in realistic workflows.
Read the full abstract ▾
All authors Zanting Ye, Shengyuan Liu, Xin Liu, Chenhui Wang, Zhisong Wang, Jiashuai Liu, Zipei Wang, Cheng Wang, Wentao Pan, Mengjie Fang, Di Dong, Mohammad Salmanpour, Arman Rahmim, Yu Gu, Yong Xia, Hongming Shan, Yixuan Yuan, Yefeng Zheng, Lijun Lu
Artificial Intelligence · Cross-listed from cs.CY
ExploraTwin, a Non-Profit Research Platform for Digital Twin Simulations
Naveen Venkat, Yuchen Qiu, Tianyi Peng et al. (+2 more)
Digital twin simulations show promise, but current empirical evidence suggests that the approach should be tested before being deployed in any particular context. To lower the friction for researchers and practitioners to test and deploy digital twin simulations, this brief commentary introduces ExploraTwin (https://exploratwin.org), an open-access, non-profit research platform for digital twin survey simulations. ExploraTwin supports two modes. In survey mode, researchers can upload a Qualtrics survey file or create a survey within the platform; select an available sample of digital twins; configure and run the simulation, and export analysis-ready data. In panel mode, researchers can assemble a small group of twins for open-ended conversations, document annotation, and moderated, focus-group-style voice discussions. We also developed CroissantTwin, a standardized data format for adding samples of digital twins to the platform. We demonstrate the survey mode workflow by using the platform to replicate 19 experiments on digital twins from the Twin-2K-500
dataset. ExploraTwin's survey execution fidelity is high: 99.6% of 197,000 answer units returned a structurally valid response on the first run.
Read the full abstract ▾
All authors Naveen Venkat, Yuchen Qiu, Tianyi Peng, George Gui, Olivier Toubia
Artificial Intelligence · Cross-listed from physics.ed-ph
Large Scale AI Grading of Handwritten Physics Assessments: Score Agreement and Olympiad Team Selection Outcomes
Praveen Pathak, Siddharth Tiwary, Charudatt Kadolkar et al. (+4 more)
Multimodal AI can read handwritten physics solutions, but high-stakes grading requires agreement with official scores and outcomes. This study evaluated GPT-5.5-based grading on 10364 scanned pages from 520 handwritten submissions by 416 unique candidates or students across three assessments: a national Physics Olympiad theory examination, the final Olympiad selection camp with theory and experiment components, and a university
quantum-mechanics examination. Each submission was graded twice by AI using the official rubrics. The second round used revised page-by-page and evidence-location instructions developed after first-round disagreement analysis. During grading, AI did not see official human marks or AI–human comparisons. Total-score correlations with official marks were high (0.91–0.97). For the final Olympiad selection, AI recovered the same five-student team as official grading. The second round improved aggregate question-part agreement, especially where first-round disagreements were larger. The main difficulty remained exact partial-credit grading, especially in experimental work. Reliable AI grading therefore depends on detailed rubrics and should be used as a second reader or audit tool under examiner control.
Read the full abstract ▾
All authors Praveen Pathak, Siddharth Tiwary, Charudatt Kadolkar, Vijay Singh, David Rakestraw, Shirish Pathare, Anwesh Mazumdar
Artificial Intelligence · Cross-listed from physics.med-ph
An integrated diffusion-weighted imaging processing and interpretation platform for MR-guided radiotherapy
Yunxiang Li, Yan Dai, Yen-Peng Liao et al. (+3 more)
Background:
Magnetic resonance imaging-guided linear accelerators (MR-Linacs) allow
diffusion-weighted imaging (DWI) to be acquired at every treatment fraction, but converting these low-signal-to-noise-ratio acquisitions into clinical decisions requires both reliable quantitative processing and an interpretation that reconciles a scattered and often contradictory literature. Purpose: To describe and evaluate an integrated, web-based platform that carries raw MR-Linac DWI to a structured, literature-grounded clinical interpretation, and to assess its
retrieval-augmented generation (RAG) interpretation module by independent expert rating. Methods: The platform couples a deep-learning processing pipeline, comprising distortion correction, denoising, and intravoxel incoherent motion (IVIM)/apparent
diffusion coefficient (ADC) fitting, with longitudinal region-of-interest analysis and a RAG interpretation
agent. The
agent reasons over a two-layer knowledge base of curated publications (a structured catalog index plus line-indexed full text), delegates arithmetic to deterministic tools, and is designed to trace each statement to a source document, section, and line range. One medical physicist and one physician independently rated the
agent's reports for nine longitudinal glioblastoma cases on a 1-5 scale across three metrics: clinical-
reasoning soundness, literature-citation quality, and overall clinical utility. Results: Across 54 ratings, the pooled mean was 4.65 +/- 0.80, with 93% of ratings >= 4; metric means were 4.6 (
reasoning), 4.5 (citation), and 4.8 (utility), and raters agreed within one point on 85% of paired ratings. Conclusions: A single platform can integrate MR-Linac DWI post-processing with traceable, expert-evaluated clinical interpretation, while highlighting the safeguards needed to verify LLM-generated
reasoning in radiation oncology.
Read the full abstract ▾
All authors Yunxiang Li, Yan Dai, Yen-Peng Liao, Jie Deng, Jill B De Vis, You Zhang
Artificial Intelligence
FL-MAESTRO: Multi-Agent LLM Orchestration for Resource-Constrained Federated Learning
Jiajun Wu, Zirui Wang, Jiayu Zhou et al. (+2 more)
In
Federated Learning (FL), the communication
topology is a runtime variable rather than a fixed design choice, since links and edge devices drop in and out during training. Each round, the server must commit three coupled decisions, namely the communication
topology, per-client resource allocation, and the aggregation rule for combining local updates. Recent agentic systems have begun bringing large
language models (LLM) into FL, but the existing line of work either operates at setup time or handles a single runtime dimension such as client selection. We propose FL-MAESTRO, a multi-
agent orchestrator that makes the joint runtime FL decision directly through three specialist LLM
agents, one per decision dimension. A coordinator combines their analyses into a single decision, and a non-LLM feasibility check confirms it before the round executes. Because the orchestrator consumes the server's predicted-failure list, it withholds clients whose updates would never be aggregated, which removes the dominant source of wasted round energy in classical FL on volatile edge networks. Because client state is read as natural-text profiles, the same orchestrator extends to heterogeneous device classes without per-class energy models. On a non-IID CIFAR-10
benchmark, FL-MAESTRO matches the accuracy of the strongest energy-aware baseline while cutting wasted round energy from over a third to near zero. Code is available at https://github.com/denoslab/FL-MAESTRO.
Read the full abstract ▾
All authors Jiajun Wu, Zirui Wang, Jiayu Zhou, Qiang Ye, Steve Drew
Artificial Intelligence · Cross-listed from cs.SE
Making Deployments Safe at Meta: Health Checks for Continuous Change-Safety
Prakash KL, Anton Korenkov, Uttam Thakore et al. (+1 more)
Continuous deployment to large scale production systems creates a tension between release velocity and reliability. Every change is a potential reliability incident, yet every delay is a missed opportunity. This paper describes the deployment time health check infrastructure that Meta uses to mediate this tension across thousands of heterogeneous services. We summarize the architecture of this prevention based distributed system's service called Service Health Checker, explain how check authors compose templated metric queries, thresholds, and workflow predicates; and discuss how the system is integrated with tiered and phased rollouts so that regressions trigger automatic rollback. We then describe the operational problems that emerged at scale, such as noise, alert fatigue, drift, and uncovered regressions, and the program of measurement, tooling, and improved defaults we deployed to address them. We close with lessons learned from years of operating deployment health checks at Meta, and the directions we are exploring next, including AI assisted health check tuning. Index Terms: deployment safety, continuous deployment, monitoring, software reliability, release engineering, software reliability engineering, AIOps, anomaly detection
Read the full abstract ▾
All authors Prakash KL, Anton Korenkov, Uttam Thakore, Christopher Hegre
Artificial Intelligence
A Temporal Planning Approach for Intelligent Flood Response
Fazlul Hasan Siddiqui, Md. Monjurul Islam and Sabah Binte Noor
Effective response to multiple, simultaneously flooded areas requires coordinating appropriate actions in the correct temporal order, under severe resource constraints. Automated planning provides a foundation for addressing this challenge by generating time-aware schedules, given a formal description of available resources, constraints, and goals. This work presents an intelligent flood-response framework that exploits temporal planning and models the complete operational life cycle of flood response. The framework incorporates priority-driven triage, route accessibility and travel costs, resource allocation, and supply management, while also supporting mid-execution re-planning in response to unexpected environmental changes. The framework is formulated both in the Action Notation Modeling
Language (ANML) and the Planning Domain Definition
Language (PDDL) 2.1, facilitating compatibility with a wider range of temporal planners. Experimental results establish the feasibility and scalability of the proposed framework, showing that flood response scenarios can be effectively modeled and solved using temporal planning, while providing guidance on planner selection.
Read the full abstract ▾
Artificial Intelligence · Cross-listed from cs.NI
Towards Traffic Modelling of Multi-Agent Systems: The Role of Coordination Topology
Davide Lamagna, Albert Cabellos, Alberto Rodriguez-Natal et al. (+2 more)
Multi-
agent LLM systems are an emerging networked workload whose rapid deployment raises questions about the traffic patterns they generate. Compared to conventional applications, these systems generate requests internally: a single user task can induce a structured sequence of model calls whose timing is governed by coordination logic rather than by user arrival rate. It is not clear whether classical traffic models, designed for human-driven workloads, apply to this setting. We present an empirical characterisation of LLM-call interarrival time distributions across sequential,
star, and full-mesh agentic coordination
topologies, using a multi-layer measurement framework over 500 repeated runs per
topology. We find that
topology fundamentally shapes the arrival process of requests to the LLM backend: fan-out coordination introduces a structural bimodality absent in sequential execution, and the reasoningphase component is best described by a log-normal distribution, with the Poisson exponential null model decisively rejected across all
topologies. These differences propagate to inference and network level metrics. The framework and analysis pipeline are released openly at https://github.com/dlamagna/agentraffic.
Read the full abstract ▾
All authors Davide Lamagna, Albert Cabellos, Alberto Rodriguez-Natal, Gábor Rétvári, Berta Serracanta
Artificial Intelligence
Lost in Translation: How Universal Ethical Values Fail to Translate Across Global Contexts
Ozioma C. Oguine, Munachimso B. Oguine, Cesar Cervera et al. (+6 more)
AI ethics frameworks treat values such as
fairness, transparency, and accountability as universal and uniformly operationalizable across contexts. We examined how 14 experts across 10 countries made sense of AI in practice, reinterpreted core values, and envisioned governance alternatives. We found that AI deployment is characterized by structurally unequal conditions, marked by infrastructural constraints, extractive practices, and a "mystification" of technology, which fundamentally shape perceptions of risks and opportunities. Our findings reveal that experts reinterpret values to fit local moral logics:
privacy as collective and relational rather than individual; transparency as trust-building accountability rather than technical disclosure; and
fairness as equity in access and representation rather than parity in outcomes. We identify these as translation gaps between encoded global frameworks and situated local practices. Finally, we propose pathways toward plural governance that redistributes epistemic authority and treats ethical negotiation as an ongoing, context-sensitive process rather than a settled technical standard.
Read the full abstract ▾
All authors Ozioma C. Oguine, Munachimso B. Oguine, Cesar Cervera, Jenny Yang, Pooja Voladoddi, Mario Rodriguez, Saif Eddin Bani Malhem, Karla Badillo-Urquiola, Daricia Wilkinson
Artificial Intelligence
Terminal Agents: A Survey of AI Agents in Command-Line Environments
Yi Bin, Xiaoyang Yuan, Haoxi Zeng et al. (+9 more)
Large
language model
agents increasingly act through terminals, yet existing surveys disperse terminal-mediated behavior across software engineering, tool use, and computer-use research. We regard terminal
agents as systems whose dominant progress-bearing action–observation loop is mediated by terminal command execution, textual feedback, and stateful environment interaction. Using terminal-mediated execution as an organizing lens, this survey establishes workload-level boundaries and connects system architecture, competence acquisition, and evaluation through a seven-dimensional terminal competence profile. Our synthesis shows that realized behavior is jointly shaped by the model, interface, harness, runtime, and environment. Executable trajectories ground learning in action consequences, verification, and recovery, whereas prevailing evaluations emphasize final outcomes and expose process quality, recovery, and governance unevenly. Bounded fixed-condition diagnostics illustrate two implications:
benchmark families expose different process signals, and matched system comparisons reveal
benchmark-dependent performance and limits of component attribution. These findings motivate explicit reporting of system and runtime conditions, supported by replayable traces and process-level evidence. The framework provides a unified basis for studying terminal-mediated agency across software engineering and emerging application domains.
Read the full abstract ▾
All authors Yi Bin, Xiaoyang Yuan, Haoxi Zeng, Wencheng Ye, Wenqi Shao, Chen Qian, Wei Ye, Yujuan Ding, Zheng Wang, Pengpeng Zeng, Jingkuan Song, Heng Tao Shen
Artificial Intelligence · Cross-listed from cs.CR
AEGIS: Preventing Cross-Domain Resource Abuse in MCP
Shriti Priya, Teryl Taylor and Frederico Araujo
The Model Context Protocol (MCP) is an open source JSON-RPC protocol that standardizes how large
language models (LLMs) interact with external systems through programmatic functions known as tools. Attackers or malicious
agents can exploit certain modalities of these MCP tools to degrade the overall quality of service of
agent-based applications. For example, an
agent may request an excessively large search radius or very long
videos, overloading backend systems and potentially causing slowdowns or denial-of-service. Each modality including text, images,
video, and location introduces distinct vectors for resource abuse, complicating the development of consistent mitigation strategies. Moreover,
multimodal and crossdomain tools expose diverse request schemas and parameters, making it difficult to define policies that are both generalizable and precise enough to enforce meaningful resource constraints. In this paper, we present AEGIS, a policy enforcement component that enables administrators to define fine-grained safeguards against resource abuse across heterogeneous MCP tools and modalities. AEGIS leverages the
reasoning capabilities of large
language models to analyze, categorize, and normalize diverse tool invocations into a unified, policy-friendly representation accessible to security practitioners. Integrated with the Open Policy
Agent and the ContextForge AI Gateway, AEGIS detects and mitigates abusive behaviors while preserving the flexibility of MCP-based
agent ecosystems.
Read the full abstract ▾
Artificial Intelligence
STCO: Conditional Neural Operators for Time-Dependent PDEs
Xingxin Yang, Zhan Zhang and Juan Li
Neural operators have emerged as efficient surrogates for time-dependent physical systems governed by partial differential equations (PDEs), but their future-state predictions are often conditioned only on observed states and static problem descriptors. For control or optimization, however, body motion, inflow, or forcing are prescribed for the query without being determined solely by the observed state. We introduce the Spatiotemporal Conditional Operator (STCO) for prescribed-condition operator learning (PCOL), a common interface that supplies prescribed target-time condition fields to heterogeneous backbone architectures while retaining their architecture-specific core computation and context pathways. Its condition interface combines Flow-Aware Graph Leaf (FAGL) with Dual-Site Feature-wise Linear Modulation (DSFiLM). Non-learned FAGL uses vorticity from the final observed frame to construct a fixed-cardinality adaptive partition, then co-locates the observed history and target-time condition fields at its regional coordinates. DSFiLM injects separate motion, inflow, and force routes before and after operator computation through current-feature-driven slot- and channel-wise gates. We evaluate twelve matched backbone architectures with different existing physical and temporal inputs. The immersed-boundary computational fluid dynamics (CFD)
benchmark spans prescribed motion, inflow disturbances, body-force actuation, and morphology. Across twelve matched backbones, three regimes, and two lead ranges, STCO yields mean paired reductions of 31.1% in relative-L2 field error and 24.7% in normalized pressure-derived load error. It also lowers longer-lead field error for 11 backbones, while interventions on individual condition groups produce measurable prediction changes for every group evaluated.
Read the full abstract ▾
Artificial Intelligence · Cross-listed from cs.LG
Decision Tree and K-Means Analysis of Raman Spectra for Edible Oils: A Physics-Informed AI Approach
Amrita Shaw, Chandrasekar S. N., Sai Muthukumar V. et al. (+2 more)
Classification of edible oils in processed foods is important for food quality, fraud prevention, and regulatory compliance. This study develops a Mutually Exclusive, Collectively Exhaustive framework integrating spectral organization, interpretable classification, Physics-Informed Artificial Intelligence (PI-AI), and Frugal AI-based feature reduction. Five edible oils were analyzed in pure form and within a fried-potato-chip
matrix using t-SNE, K-means clustering, Decision Trees, and Non-Negative Least Squares (NNLS)-based spectral decomposition. Unsupervised analyses showed stronger class organization and separability in pure oils, while food-
matrix effects caused substantial spectral overlap. Decision Trees achieved 100% classification accuracy for pure oils using only four Raman variables from 1866 spectral features. These variables represented only 0.21% of the available spectral information while retaining perfect test-set performance. Two variables associated with lipid unsaturation (about 1650 cm-1) and hydrocarbon-chain organization (about 1127 cm-1) remained important after NNLS
matrix correction. Their combined contribution increased from 50% in pure oils to about 62% and 89% in paper-subtracted and paper-plus-potato-subtracted
datasets, respectively. NNLS-based PI-AI improved food-
matrix classification by separating oil signatures from paper and potato contributions. Optimized post-pruned models achieved nearly 80% test accuracy using only five and four Raman variables, respectively. The four-feature representation reduced the data footprint by 99.44% without loss of accuracy. These findings demonstrate that Raman-based oil identification can use compact, physically meaningful, and interpretable spectral representations, supporting Frugal AI, Edge AI, portable sensing, and embedded food-quality monitoring.
Read the full abstract ▾
All authors Amrita Shaw, Chandrasekar S. N., Sai Muthukumar V., Jhinuk Gupta, Deepak L. N. Kallepalli
Artificial Intelligence · Cross-listed from physics.soc-ph
Peer-Voted LLM-Agent Stress Tests Find Feed-Induced Lexical Convergence but No Reliable Matched-Exposure Advantage for Distributed Sources
Rana Muhammad Usman and Dominic Williamson
Population-level behavior in large-
language-model (LLM)
agents cannot be characterized by single-
agent benchmarks. We introduce PV-SST, a peer-voted social-platform testbed, and report a separately frozen, preregistered matched-exposure experiment spanning four topics, four unused seeds, four open-weight model families, and three prespecified larger variants. The experiment comprises 448 trials and 112 complete model-by-topic-by-seed blocks. Relative to a topic-only control, a feed of previous-round peer posts ranked by peer-generated likes increases final-round lexical similarity in both the four-family core panel (paired mean difference +0.0082 TF-IDF cosine units, 95% block-bootstrap CI [0.0043, 0.0121], randomization p=0.000105, n=64 blocks) and the three-variant size extension (+0.0109 [0.0069, 0.0151], p=0.000001, n=48). This contrast bundles peer-post exposure with ranking and therefore does not identify a ranking-only effect. Opposite-side survival falls in the core panel (-3.9 percentage points [-6.8, -1.6], p=0.0068) but not conclusively in the larger variants (-1.0 pp [-3.1, 0.4], p=0.50). Holding
adversarial impressions fixed, four distributed sources do not reliably move honest-
agent stance more than one source. The preregistered distributed-minus-single contrast is positive but inconclusive in the core panel (+0.057 [-0.009, 0.125], p=0.112) and negative in the larger variants (-0.040 [-0.113, 0.035], p=0.332), failing the prespecified cross-model and cross-topic consistency criterion. Thus the robust result is lexical convergence under the tested peer-ranked feed, not general opinion capture or a general coordination advantage. The study evaluates synthetic LLM-
agent populations; it does not estimate effects on people or production platforms.
Read the full abstract ▾
Artificial Intelligence · Cross-listed from cond-mat.mtrl-sci
An LLM agent for end-to-end computational materials discovery
Chen Yuntong, Huang Lijiang, Li Yingqi et al. (+7 more)
Coordinating multi-scale tasks is an effective strategy for computational materials discovery, yet the diverse
algorithms and tools involved make this process highly challenging. Here we present DREAMS, a large
language model (LLM)-based
agent system capable of executing the entire screening pipeline for metal-organic frameworks (MOFs). DREAMS processes a large body of MOF literature, links publications to their
crystal structures, and curates them into a computation-ready database, which is then screened with progressively more expensive simulations. The promising candidates identified for separation under wet flue gas conditions all originate from independent studies. By bridging the heterogeneous stages from literature mining to high-throughput simulation, DREAMS operates across application domains and enables even non-experts to discover high-performance materials in an end-to-end, automated manner.
Read the full abstract ▾
All authors Chen Yuntong, Huang Lijiang, Li Yingqi, Liu Yu, Zhao Dan, Sun Mingqi, Ju Chentian, Liu Yanbing, Li Xiao-Yan, Zhao Guobin
Artificial Intelligence · Cross-listed from cs.LO
ProofJudge: Tool-Grounded LLM Evaluation of Formal Proof Quality in Mathlib
Shane Caldwell
Formal proofs in Lean 4 that pass the kernel's type checker can nonetheless vary widely in quality. We introduce ProofJudge, an agentic LLM-as-judge system that scores formal proof quality along five dimensions beyond correctness: library leverage, automation fit, structural clarity, statement quality, and Mathlib conventions. We evaluate ProofJudge on a novel
dataset of 218 declarations drawn from distinct Mathlib PRs. The judge
agent is grounded by tool access to the commit the PR is applied to, enabling it to query the library state when scoring. A judge is considered aligned with human preferences when it rates the version of the PR Mathlib accepted above the initial version that was sent back for revision. All six judge models evaluated recover the reviewers' preference well above chance, from 80.8% to 63.5%, and two open-weight judges reach roughly 70% at a tenth of the best judge's cost. We release the judge harness, evaluation
dataset, and evaluation traces as open-source artifacts to support further research.
Read the full abstract ▾
Artificial Intelligence · Cross-listed from cs.CL
G-CARL: Grounded Checklist-Aligned Reward Learning for Patient-Oriented Medical Report Interpretation
Shiao Xie, Siyu Chen, Jianwei Lv et al. (+3 more)
Personalized interpretation of medical reports has emerged as an increasingly important need among patients. Addressing this need requires both evidence-grounded medical factuality and context-dependent patient communication, yet existing medical
vision-
language tasks do not adequately capture these dual requirements. To bridge this gap, we introduce Patient-oriented Medical Report Interpretation (PMRI), a novel open-ended
multimodal generation task that requires models to explain medical reports in accurate and accessible
language based on a user's query and dialogue history. These two objectives differ fundamentally in their verifiability, yet remain tightly coupled, making them difficult to optimize jointly under conventional supervised fine-tuning and holistic reinforcement learning paradigms. To address this challenge, we propose G-CARL, a grounded, checklist-aligned reinforcement learning framework that combines multi-source
retrieval for atomic claim verification with context-aware, instance-specific weighted checklists for response coverage, providing structured supervision for factuality, user-demand satisfaction, and expression quality without constraining response diversity. We further construct MMedReport, a real-world PMRI
benchmark, along with a clinician-designed three-dimensional evaluation protocol. Extensive experiments demonstrate that G-CARL consistently outperforms existing post-training baselines in overall quality, claim-level precision, and checklist recall. Pairwise preference evaluation by clinicians further confirms that G-CARL produces interpretations that are more accurate and better aligned with patient needs.
Read the full abstract ▾
All authors Shiao Xie, Siyu Chen, Jianwei Lv, Bo Yuan, Yujin Wang, Xiandong Li
Artificial Intelligence
An Agentic Approach for Active Data Collection, Travel Behavior Modeling, and Weather-Sensitive Demand Prediction
Narges Ahmadi, Yubo Jiao, Jônatas Augusto Manzolli et al. (+2 more)
Travel behavior research increasingly combines digital data collection with predictive modeling, yet these stages are often developed and evaluated separately. This study proposes a three-
agent workflow integrating conversational data collection, structured data processing, and behavioral prediction. A chatbot-administered, image-augmented stated-preference survey collected mode choices from student commuters across five predefined weather scenarios, yielding 454 respondent-scenario observations. Weather-related associations were analyzed using a multinomial logit model, while logistic regression and random forest provided machine-learning
benchmarks. Nine locally deployed large
language models (LLMs), ranging from 2 to 35 billion parameters, were evaluated across four zero-shot
prompt-and-context conditions and extended through persona, few-shot, and
vision-based configurations. Random forest achieved 69.6% five-class accuracy, while the best text-only zero-shot LLM reached 69.9% without task-specific fitting. Habitual travel information produced the most consistent gains, Expert framing generally outperformed Role-Play, and persona information was most useful when habitual travel information was unavailable. Few-shot prompting improved prediction for several models, with gains stabilizing after a small number of examples. Using the same weather images shown to respondents, the best
vision-based configuration reached 71.5% five-class accuracy, indicating that visual context may provide additional predictive information for selected models. Overall, the study shows how conversational surveys, structured data processing, conventional behavioral modeling, machine learning, and
multimodal LLM prediction can be coordinated within an auditable multi-
agent workflow.
Read the full abstract ▾
All authors Narges Ahmadi, Yubo Jiao, Jônatas Augusto Manzolli, Jiangbo Yu, Luis Miranda-Moreno
Artificial Intelligence · Cross-listed from cs.CL
Inducing Task Models from Computer-Use Traces
Yucheng Jiang, Zora Zhiruo Wang, Ruishi Chen et al. (+1 more)
Naturalistic computer-use traces, passively recorded screenshots and mouse or keyboard actions, are a valuable resource for deriving symbolic, auditable, and reusable models of how everyday work is done. Such models matter as computer-use
agents enter real work, where
agents need to learn how tasks are actually performed, and organizations need to audit and reuse that knowledge. However, inducing such task models is challenging, as activity is observed only as low-level events and real-world work is multi-threaded with interleaved goals. Existing methods assume a given task or a single workflow, and produce step-level summaries rather than structured task models. We introduce Task Model Induction (TMI), which (i) discovers the latent tasks in an unconstrained trace, disentangling concurrent activity, and (ii) for each latent task, induces a task model pairing a hierarchical objective model of recursive goal decomposition with a procedure model of the control flow that organized the execution. Intrinsically, on controlled human and
agent trajectories, TMI recovers interleaved tasks with 0.974 agreement against ground-truth groupings and reconstructs 74.9% of the observed execution steps, far more than the strongest workflow induction baseline. Extrinsically, skills derived from TMI's task models improve held-out task accuracy by 30.0% over the strongest baseline.
Read the full abstract ▾
All authors Yucheng Jiang, Zora Zhiruo Wang, Ruishi Chen, Diyi Yang
Artificial Intelligence
AI4AI-Bench: Benchmarking LLM Agents in Algorithmic Design for Recursive Self-Improvement
Yizhe Chi, Wenyi Li, Deyao Hong et al. (+7 more)
Recursive self-improvement (RSI) asks whether an AI system can improve the process that produces AI systems, so that the next system inherits the improvement. That process is the training
algorithm: a better objective or update rule improves the compute\mbox{-}capability exchange rate for every subsequent run, including the one that produces the next
agent. Whether RSI is feasible therefore turns on whether an
agent can design training
algorithms. No
benchmark isolates that ability: existing suites are won by collecting data or by tuning hyperparameters, and none tells a change to how a run is executed apart from a change to how the model learns. We present AI4AI\mbox{-}Bench, 10 frozen research repositories spanning 10 training
algorithm families. In each task, an
agent has 4 hours on one B300 to rewrite the training
algorithm; its code is then rerun from scratch for up to 12 hours and scored by a fixed evaluator hidden from the
agent, against the repository's original
algorithm under the same procedure. Because the 10 metrics are incommensurable, every task is mapped onto one scale on which $0$ is an uninformative model, $0.1$ is the
algorithm the repository ships, and $1.0$ is the task optimum. Across 29 configurations of 6 systems on all 10 tasks the mean score is $0.166$, and the best system reaches $0.250$: even the strongest closes under a fifth of the distance between the
algorithm that was already there and the optimum. The submissions show where that distance went: most never change how the model learns at all, and the minority that do average $0.226$ against $0.126$ for the rest. More
reasoning effort mostly buys the willingness to go there, taking that minority from $8\%$ of submissions to $64\%$ and the mean score from $0.094$ to $0.196$. We release the task suite, the evaluators and every scored submission, so that the measurement can be repeated as these systems change.
Read the full abstract ▾
All authors Yizhe Chi, Wenyi Li, Deyao Hong, Xiaoqiu Wang, Mingju Gao, Kaisen Yang, Bingxiang He, Youjie Zheng, Calvin Xiao, Qinhuai Na
Artificial Intelligence
Pandora's AI Model Routing Box: Efficient Allocation with Costly Value Estimation
Adam Fisch, Shubhendu Trivedi, Fantine Huot et al. (+5 more)
Heterogeneous AI systems composed of multiple models, architectures, harnesses, or inference-time settings can improve quality and efficiency by routing queries to the specialist who can answer most effectively at the lowest cost. Routing requires estimating each specialist's expected return, but this value estimation has a cost. Cheap
estimators (e.g.,
embedding-based predictors) are fast but noisy, while accurate
estimators (e.g., fine-tuned models with access to
retrieval results or partial
reasoning traces) are expensive. We formalize this tradeoff as an instance of Pandora's Box, the classical problem of optimal search with costly inspection. Under a Gaussian signal model, the resulting policies have closed-form value-of-information expressions that determine, for each specialist and input, whether refining the value estimate is worth its cost. We call the centralized policy Pandora's Router. We extend this to a decentralized setting, Pandora's Bidder, where specialists independently decide whether to invest in self-assessment before accepting an offered price to claim a query. Experiments across three domains—a standard multi-LLM
benchmark,
retrieval-augmented specialists, and LLMs with variable inference-time
reasoning—show that Pandora's Router matches the routing quality of exhaustive estimation, while querying the expensive
estimator far less often. In the decentralized setting, value-of-information
reasoning improves allocative efficiency when competing estimates are accurate; when competing estimates are noisy, however, it can increase the strategic specialist's utility at the expense of others.
Read the full abstract ▾
All authors Adam Fisch, Shubhendu Trivedi, Fantine Huot, William W. Cohen, Michael Kaisers, Mirella Lapata, Kate Larson, Jacob Eisenstein
Artificial Intelligence
MidTool: Mid-training Data Synthesis for Agentic Tool Use
Fengqing Jiang, Yite Wang, Boyi Liu et al. (+5 more)
Mid-training is increasingly recognized as a critical stage for shaping the capabilities of large
language models. Recent work has shown that targeted mid-training can strengthen
reasoning-intensive abilities such as math and science, and can also improve agentic capabilities in software-engineering settings. In this work, we study the parallel but less explored agentic capability: general tool use. We present MidTool, an open corpus construction pipeline for agentic tool-use mid-training that combines large-scale web, PDF, and code data with synthesized supervision from real-world tool APIs, MCP skills, and document-grounded workflows. MidTool is designed to teach models how to recognize tool affordances, ground arguments from context, compose tool call workflow, and recover from incomplete information. We mid-train Qwen3-4B-Base and Qwen3-8B-Base on MidTool-Mix, and then apply follow-up post-training with both supervised fine-tuning and reinforcement learning. Compared with baselines, MidTool-Mix consistently improves downstream performance under both SFT and RL on BFCL, tau2-Bench, and MCP Universe. These results suggest that general tool use, like other important LLM capabilities, benefits from dedicated mid-training rather than being left entirely to post-training.
Read the full abstract ▾
All authors Fengqing Jiang, Yite Wang, Boyi Liu, Zhaoyang Wang, Canwen Xu, Zhewei Yao, Radha Poovendran, Yuxiong He
Artificial Intelligence
Phantom Gains: Auditing Self-Improvement Against a Measured Null
Cheng Xu, Nan Yan, Liming Chen et al. (+1 more)
Whether a
language model has improved itself is increasingly judged not by mean accuracy but by which individual problems it gains and loses. Tracking these transitions means differencing two noisy estimates, leaving them vulnerable to measurement artifacts.
Auditing three rounds of rank-$32$ LoRA self-training on Qwen3-8B against a frozen control pushed through the identical pipeline, we identify seven measurement failures, each of which inverts a reported finding when its control is absent. Several are standard practice. A ledger built on a single greedy decode manufactures capability changes on an untrained model, largely an artifact of inference batching; the expansion statistic separating acquisition from sharpening assigns that same model a rate of $0.280$. The natural threshold repair does not survive replication: estimated across the frozen comparisons such a design already contains, its null stays non-zero. We replace it with a per-problem exact test against a pooled baseline under false-discovery-rate control, which detects nothing on any held-out replicate and is unchanged under the multiple-testing rule, error rate and pool size. Applied to a ladder of arms matched in stream, volume and evaluation, the audit finds that external
distillation improves problems the base model rarely reaches while three forms of self-training do not; a regression rejects this asymmetry as a by-product of
distillation's larger overall gain ($p < 10^{-8}$). On the far smaller set of problems the base model never reaches, the evidence is inconclusive, while self-training corrupts problems solved at baseline at rates well above the measured floor. Transition-level
auditing therefore requires a separately measured null for every statistic it reports: nulls that cost no new experiments, built from baseline replicates a multi-arm study already owns, though not from as few as most possess.
Read the full abstract ▾
All authors Cheng Xu, Nan Yan, Liming Chen, M-Tahar Kechadi
Artificial Intelligence · Cross-listed from cs.CL
Inject, Align, Recover: Staged Post-Training for Retrieval-Free Document Knowledge Internalization
Qian Kou, Xiaofeng Shi, Xiaosong Qiu et al. (+1 more)
Large
language models often fail to answer questions about a bounded document collection when the source documents are not retrieved at inference time. We study this setting as document knowledge internalization: converting a fixed corpus into usable parametric knowledge for
retrieval-free question answering. We propose IAR (Inject, Align, and Recover), a three-stage post-training framework that separates structured document knowledge injection, QA behavior
alignment, and general ability recovery. Unlike conventional continued pretraining, Inject converts source documents into continuation, rewrite, and instruction-conditioned reconstruction objectives. Align then adapts the injected model with answer-only QA supervision, while Recover merges the domain-adapted model with the base instruction model to recover general capabilities. Across Common Corpus (CC) and CCI, and across Llama, Phi, Qwen, and SmolLM model families, IAR improves the domain-primary domain-general frontier for
retrieval-free document internalization. In the main comparison, IAR improves over Vanilla SFT on all four reported metrics in 7 of 8
dataset-model settings, with average gains of 3.6 percentage points in domain QA accuracy and 12.1 percentage points in mean general performance across IFEval, MMLU, and MSBench. Extended CC baselines show that LoRA and FAPM can win individual general metrics, but among methods that also reach leading or near-leading domain internalization, IAR retains one of the strongest general profiles.
Read the full abstract ▾
All authors Qian Kou, Xiaofeng Shi, Xiaosong Qiu, Hua Zhou
Artificial Intelligence
Break It Down, Pass It On: Cross-Task Skill Transfer in LLM Agents
Yiyang Feng, Biddut Sarker Bijoy, Niranjan Balasubramanian et al. (+1 more)
Large
language model (LLM)
agents can induce skills from completed tasks and reuse them later to grow more capable with experience. In practice, induced skills may transfer unreliably and can even harm the
agent that retrieves them. When
agent-induced skills transfer reliably across tasks remains an open question. We conduct a comprehensive and controlled study of how the way skills are induced shapes their transfer across tasks. Specifically, we compare task-level with subtask-level skill induction and text with code skill formats, the two axes along which existing methods differ. Task-level skills mostly reduce the
agent's performance below its no-
memory baseline while subtask-level skills raise it above on average, and text skills transfer better than code skills. To further understand our findings, we examine two complementary properties of the induced skills: specificity, which measures how closely a skill matches real tasks, and abstractness, which measures how evenly its relevance spreads across tasks. Neither property alone predicts task success, but their combined effect does, which we propose as a skill utility score. The score correlates consistently with task success when skills are transferred, and subtask-level and text skills score higher. Computing skill utility only needs the skills and task descriptions, so our score serves as a lightweight diagnostic of a skill
memory for practitioners.
Read the full abstract ▾
All authors Yiyang Feng, Biddut Sarker Bijoy, Niranjan Balasubramanian, Jiawei Zhou
Artificial Intelligence
Catching the Rug: Early Prediction of Fraudulent Memecoins on Solana via Machine Learning
Jianghai Li, Pavel Kuznetsov, Yury Yanovich et al. (+2 more)
The rapid proliferation of memecoins on blockchain platforms has increased the risk of fraudulent activities, particularly rug pulls. While previous studies have focused on Ethereum-based
tokens, this paper shifts the spotlight to Solana, the leading blockchain for memecoins by trading volume and
token count. Unlike Ethereum, where rug pulls often exploit smart contract
backdoors, Solana memecoin rug pulls are predominantly driven by liquidity manipulation and social dynamics. This research pioneers large-scale rug pull early detection in the Solana ecosystem by assembling a
dataset of 6.4 million
tokens over 7 months.
Market analysis reveals that a vast majority of these memecoins exhibit rug pull characteristics within one hour of launch, highlighting the urgency of short-horizon prediction. Despite the absence of code-level features, we demonstrate that classic machine learning models, particularly Gradient Boosting (XGBoost), achieve robust performance in detecting potential rug pulls using only the first 5 minutes of trading data. Furthermore, we evaluate cross-platform generalization between PumpFun and Raydium, revealing that multi-source data fusion significantly mitigates domain shift and improves detection reliability. This study advances the understanding of DeFi fraud on high-throughput chains and provides a practical framework for protecting investors.
Read the full abstract ▾
All authors Jianghai Li, Pavel Kuznetsov, Yury Yanovich, Konstantin Nott-Whaley, Igor Vodolazov