Artificial Intelligence · Cross-listed from cs.RO
Safe Multi-Robot Coordination via VLM-LLM Reasoning and Reachability Analysis
Mohamed Dwedar, Ahmad Hafez, Alexander Jesser et al. (+1 more)
Safe coordination in heterogeneous machine-to-machine (M2M) robotic systems is challenging when
robots differ in sensing capabilities, environmental awareness, and motion execution roles. This paper presents a centralized safety-aware M2M framework for cooperative goal-directed
navigation in a heterogeneous mobile
robot team comprising a
vision-capable quadruped and a camera-less robotic vehicle. The objective is to guide both platforms toward a goal region while avoiding static and dynamic obstacles and preventing unsafe inter-
robot interactions. Under the principle of shared perception, the
vision-capable
robot provides semantic environmental awareness through a centralized server over an MQTT broker, enabling the camera-less platform to navigate using this shared scene representation alongside its own odometry, IMU, and state feedback. A
vision-
language model (VLM) interprets the visual stream, and the extracted semantic data is mapped into conservative metric
geometric constraints, including inflated obstacle sets, safe corridors, and goal regions. A large
language model (LLM) proposes high-level task allocations, while physical command authority is restricted to a
robot-specific zonotope reachability gate. This verification engine propagates independent reachable tubes to evaluate obstacle avoidance, safe-corridor containment, and inter-
robot separation predicates before approving commands. Online experiments across clear-path and dynamic-obstacle scenarios show that the pipeline reliably approves safe motion, triggers conservative replanning or holding maneuvers upon constraint violation, and enforces a strict architectural separation between advisory semantic
reasoning and formally verified motor execution.
Read the full abstract ▾
All authors Mohamed Dwedar, Ahmad Hafez, Alexander Jesser, Amr Alanwar
Artificial Intelligence · Cross-listed from cs.CV
Evaluating ADC-only deep learning pipelines for breast cancer detection and segmentation using standalone diffusion-weighted MRI
Pablo García Marcos, Paula Puerta Gonzĺez, Guillermo Lorenzo et al. (+4 more)
Dynamic contrast-enhanced (DCE) imaging is the gold standard technique for the detection and characterization of breast cancer using
magnetic resonance imaging (MRI). However, DCE-MRI requires long acquisition times and the administration of contrast into the bloodstream, which can cause allergic reactions. Alternatively,
diffusion-weighted MRI (DW-MRI) is a standard complementary technique for breast MRI that does not require contrast, has shorter acquisition times, and enables calculation of apparent
diffusion coefficient (ADC) maps that correlate with tumor cellularity. Yet, despite these technical advantages, deep learning research has focused on DCE-based models and has barely explored the tumor detection performance of DW-MRI and ADC maps either in combination with DCE-MRI or as standalone alternatives. Here, we evaluate the application of different state-of-the-art deep learning techniques for detection and
segmentation of breast cancer using ADC-only images. This is, to our knowledge, the first comprehensive evaluation of ADC-only breast cancer pipelines for classification, detection, and
segmentation tasks.
Read the full abstract ▾
All authors Pablo García Marcos, Paula Puerta Gonzĺez, Guillermo Lorenzo, Héctor Gómez, Covadonga del Camino, Angel Rio-Alvarez, Víctor M. González
Artificial Intelligence · Cross-listed from cs.CL
LabourCrew: A Multi-Agent RAG Framework for Trustworthy Adversarial Deliberation and Statutory Reasoning over Labour Law
Fatema Tuj Johora Faria, Mukaffi Bin Moin, Jubayer Al Mahmud et al. (+2 more)
In statutory question answering, every claim must be traceable to evidence, not merely relevant, since unverifiable labour-rights answers carry serious legal consequences. Current systems fall short: single-pass RAG cannot detect insufficient evidence, while multi-
agent legal-debate systems treat grounding as a prompting convention, letting
agents cite unretrieved evidence. To address this gap, we introduce LabourCrew, a multi-
agent RAG framework built around three grounding mechanisms: StatuteGraph, a graph index that explicitly links chapter, section, proviso, and cross-reference structure rather than fixed-length spans; an Evidence Exchange Protocol that confines advocates and an interpreter to an evidence ledger, making citation to unretrieved text impossible, while a fault-tolerant supervisor board runs advocates in parallel so individual failures degrade rather than crash the system; and a Calibrated Trust Gate that replaces categorical accept/reject decisions with a trust score, thresholded via conformal risk control for a distribution-free bound on the false-accept rate. We evaluate on LabourActQA, a 500-item Bangla question set from the Bangladesh Labour Act, 2006, spanning seven
reasoning categories and three difficulty tiers. The framework drives the empirical false-accept rate to 0.081, within the target level ($\alpha = 0.10$), achieves the highest Answer Relevancy among HyDE RAG, Graph-RAG, and Hierarchical RAG (0.862 $>$ 0.839, 0.815, 0.828), and degrades gradually rather than catastrophically as question difficulty increases. These results show that calibrated abstention, not
retrieval quality alone, is what makes legal question answering auditable in low-resource statutory domains.
Read the full abstract ▾
All authors Fatema Tuj Johora Faria, Mukaffi Bin Moin, Jubayer Al Mahmud, M. F. Mridha, Md. Alam Hossain
Artificial Intelligence
Early Prediction of Pathological Complete Response to Neoadjuvant Chemotherapy Using Temporal Deep Learning on DWI
Pablo García Marcos, Md. Tarequl Islam, Paula Puerta González et al. (+5 more)
Early identification of non-responders to neoadjuvant chemotherapy (NACT) is crucial for timely treatment adaptation in breast cancer. However, many existing predictive models rely on multiparametric
magnetic resonance imaging (MRI), late treatment time points, or extensive clinical data, which limits their applicability. This study proposes a deep learning framework for early prediction of pathological complete response (pCR) using only
diffusion-weighted MRI (DW-MRI) acquired at baseline and after the first NACT cycle. This framework feeds cropped tumor-centered patches to an EfficientNet-based temporal model that directly learns tumor shape and local tissue characteristics without explicit radiomic feature engineering. The model, trained with 10-fold cross-validation, achieved an area under the receiver operating characteristic curve (AUC) of 0.90 for pCR prediction after one cycle, providing actionable information after a single treatment cycle while avoiding gadolinium administration and reducing dependence on heterogeneous clinical data. By focusing on the baseline-to-first-cycle window instead of later stages, the approach supports earlier escalation or de-escalation of NACT, and its exclusive reliance on DW-MRI facilitates protocol standardization, multi-centre deployment and
privacy-preserving data sharing. These results demonstrate that DW-MRI-based deep learning on tumor-centered patches constitutes a minimally invasive, clinically deployable strategy for early pCR prediction, with direct implications for personalized treatment adaptation in neoadjuvant breast cancer therapy.
Read the full abstract ▾
All authors Pablo García Marcos, Md. Tarequl Islam, Paula Puerta González, Guillermo Lorenzo, Héctor Gómez, Covadonga del Camino, Angel Rio-Alvarez, Víctor M. González
Artificial Intelligence
Neoadjuvant chemotherapy response prediction using pretreatment diffusion and contrast-enhanced magnetic resonance imaging with clinical variables
Pablo García Marcos, Paula Puerta González, Guillermo Lorenzo et al. (+6 more)
Prediction of pathological complete response before neoadjuvant chemotherapy may facilitate more tailored therapeutic planning for breast cancer patients. This work proposes a deep-learning model for pretreatment data only, combining apparent
diffusion coefficient maps, dynamic contrast-enhanced
magnetic resonance imaging, and clinical variables. The study uses the public ACRIN 6698/I-SPY2 multicenter
dataset. The architecture employs EfficientNet-B0 pretrained encoders for image feature extraction and late fusion with clinical information. Multiple clinical variables were evaluated, including age, race, histological type, HR/HER2 subtype, SBR grade, and maximum diameter. Only HR/HER2 subtype improved the average area under the receiver operating characteristic curve (AUC) and was retained in the final model. Using stratified five-fold cross-validation, standalone apparent
diffusion coefficient maps achieved a mean AUC of 0.79, whereas dynamic contrast-enhanced
magnetic resonance imaging achieved 0.74. Adding HR/HER2 subtype improved performance to 0.83 and 0.81, respectively. The final configuration, using both imaging modalities and HR/HER2 subtype, achieved an AUC of 0.86. These results support pretreatment
multimodal learning for response prediction, although external validation is required before clinical use.
Read the full abstract ▾
All authors Pablo García Marcos, Paula Puerta González, Guillermo Lorenzo, Héctor Gómez, Covadonga del Camino, Adán Rodríguez, Ignacio Peláez, Angel Rio-Alvarez, Víctor M. González
Artificial Intelligence · Cross-listed from cs.CL
Can Large Language Models Identify Meaningful Touchpoints in Conversion Attribution?
Jinqi Wu, Sishuo Chen, Zhangming Chan et al. (+9 more)
Touchpoint selection in conversion attribution, namely identifying meaningful touchpoints contributing to conversions, is essential for e-commerce recommendation and online advertising. Current selection methods rely heavily on collaborative-filtering-based heuristics, which fail to align with user-perceived semantic intent. Through human annotation, we reveal a significant semantic gap: many implicitly-related, semantically relevant touchpoints remain undetected by existing rules. Therefore, we systematically evaluate the capability of Large
Language Models (LLMs) in identifying these hidden associations. Our evaluation shows that while LLMs effectively uncover a substantial portion of implicitly-related touchpoints, significant room for improvement remains in their selection performance. Furthermore, we analyze the impact of different prompting strategies and foundation model choices on identification performance, providing valuable insights into their
reasoning patterns and effectiveness. These insights offer a new roadmap for transitioning conversion attribution from mechanical rule-matching to human-aligned semantic
reasoning. Moreover, we leverage the LLM-attributed conversion labels for enhancing industrial CVR model training and achieve significant offline performance gains, showing the potential of LLMs in conversion attribution.
Read the full abstract ▾
All authors Jinqi Wu, Sishuo Chen, Zhangming Chan, Yong Bai, Chao Yi, Han Zhu, Shuodian Yu, Lei Zhang, Sheng Chen, Chenghuan Hou, Jian Xu, Chaoyou Fu
Artificial Intelligence
Retrieving Relations, Detecting Fallacies: A RAG Approach to Political Debate Analysis
Deborah Dore, Greta Damo, Elena Cabrio et al. (+1 more)
Fallacies are arguments that employ invalid
reasoning, making their automatic detection critical in sensitive contexts such as high-stakes political debates, where public opinion is shaped. Spotting a fallacious argument requires contextual knowledge beyond its pure surface text. This entails world knowledge pertaining to the subject matter under discussion, as well as knowledge of the relationships that exist between arguments within the argumentative discourse. Prior work on fallacy analysis has shown that argumentative discourse structure can beneficially improve classification performance. However, such structure is typically encoded only as static classifier features, limiting its flexibility. Building on this intuition while addressing this limitation, we introduce a guided
retrieval-augmented methodology for fallacy detection and classification that leverages argumentative relations of support and attack to dynamically steer the extraction of relevant documents. We evaluate our approach on the ElecDeb60to20
benchmark across 42
retrieval configurations and 14 models, performing
retrieval over a 15GB knowledge base of collected political-related documents. Our approach improves macro-F1 up to 0.864 for fallacy detection and up to 0.725 for classification over non-
retrieval baselines. These results show that incorporating external knowledge significantly enhances fallacy detection and classification when
retrieval is argumentatively guided.
Read the full abstract ▾
All authors Deborah Dore, Greta Damo, Elena Cabrio, Serena Villata
Artificial Intelligence · Cross-listed from cs.RO
Constructing Predictive Surgical Path for AI-based Capsulorhexis Skill Transfer
Mohammad Javad Ahmadi and Hamid D. Taghirad
Automated training of surgeons is one of the most crucial factors that significantly minimize
surgical training risks and expenses. With recent advances in artificial intelligence (AI) knowledge and available data from various surgeries, AI's involvement in
surgical training is becoming very promising. It is recommended that at the early stages of AI development, it interferes in the surgery as a third
agent alongside the trainer. As trust in AI increases, this process will lead to an AI
agent acting as a trainer in the future. The first phase in which AI can intervene in the training process is to suggest an improved
surgical path to the trainer. A platform must be constructed in the first step, to accomplish this task and to enhance the movement path of trainee surgeons. This paper introduces this platform along with an annotated capsulorhexis surgery
dataset called the ARAS-Farabi
dataset. In this research, a deep convolutional neural network is pre-trained with JIGSAWS and ARAS-Farabi
surgical datasets that can extract
surgical skill characteristics from surgery tool tip motion data. The proposed platform develops a reference model from the feature space of an expert surgeon's movement trajectory and proposes an improved path to enhance the skill of a novice surgeon. An optimization with two loss functions is utilized to create a path that raises the skill level of the novice surgeon's path while simultaneously predicting and preserving his/her intent. The results of this study reveal that, with the assistance of an AI
agent, the trainee surgeon's movement path can be enhanced by at least 20 percent while maintaining his intentional objective. In addition to the recommended deep network, various tangible indicators have also been developed in this research to verify the level of trainee improvement.
Read the full abstract ▾
Artificial Intelligence · Cross-listed from cs.RO
Geo-VLA: Geometry-Aware Vision-Language-Action Planning via Internalization of Map Semantics
Ran Chen, Jiaxing Ren, Zhikun Zhang et al. (+3 more)
Vision-
language-action (VLA) models have advanced end-to-end autonomous driving by leveraging foundation models for semantic
reasoning and long-tail generalization. However, their planning performance remains limited in complex driving environments because image-only representations inadequately capture planning-relevant road geometry and
topology. In this paper, we propose Geo-VLA, a plug-and-play framework that enhances VLA models by learning geometry-aware visual representations. During training, Geo-VLA internalizes
geometric map semantics to strengthen road-structure representations, while requiring no HD maps or additional lane information during inference. To support this approach, we introduce Geo-QA, a geometry-focused question-answering
dataset that injects road geometry into
vision-
language representations through contrastive learning and instruction tuning. Experiments on NAVSIM v1 demonstrate that Geo-VLA consistently improves VLA planners with distinct action-generation architectures, achieving 92.1 PDMS and establishing a new state-of-the-art among single-camera VLA planners.
Read the full abstract ▾
All authors Ran Chen, Jiaxing Ren, Zhikun Zhang, Yunhao Hou, Junbao Zhuo, Bochao Zou
Artificial Intelligence · Cross-listed from q-bio.QM
Rigorous Evaluation of Large Language Models for Malaria Drug Discovery: Trade-offs in Performance, Scale, and Resource Utility
Marvellous O. Ajala, Zainab Ashimiyu-Abdusalam and Comfort Adesina
We introduce Malaria-Instruct, a curated instruction-following
dataset derived from the ChEMBL Legacy Malaria corpus for Malaria virtual screening, and conduct a systematic evaluation of five open-source LLMs; Gemma-2 2B/9B, TxGemma-2B/9B, and LlaSMol-Mistral-7B, on a rigorous out-of-distribution data split. Performance was benchmarked against classical ML models (Random Forest, XGBoost) and frontier proprietary models (Gemini 2.5, OpenAI o3) under few-shot conditions. Fine-tuned LLMs substantially outperformed all baselines: TxGemma-9B achieved the highest ROC-AUC ($0.731 \pm 0.005$) and LlaSMol-Mistral-7B the best enrichment factor (EF@1% $\approx$ 4.99). Domain-specific fine-tuning proved categorically indispensable with TxGemma-9B collapsing from ROC-AUC 0.731 to 0.499, under its best few-shot condition, and neither Gemini 2.5 (ROC-AUC $\approx$ 0.53) nor o3 (ROC-AUC $\approx$ 0.59) achieved reliable discrimination without fine-tuning. Biomedical pretraining conferred a measurable advantage at equivalent scale, while chemistry-aware pretraining yielded superior prospective enrichment. Fine-tuned open-source LLMs represent a compelling, resource-efficient paradigm for antimalarial VS, outperforming both classical pipelines and proprietary
reasoning models under structurally challenging conditions.
Read the full abstract ▾
Artificial Intelligence · Cross-listed from cs.CY
Mapping General-Purpose AI Governance in Twenty AI Middle-Power Jurisdictions
Josephine Schwab, Nathan Naidoo, Ferruccio Barazzutti et al. (+2 more)
The most capable general-purpose AI (GPAI) models are mostly built in two jurisdictions, the United States and China, but the risks they carry land globally. Regionally advanced economies hosting no frontier developer, which we call AI middle-powers, are writing their own rules to govern GPAI. This paper investigates which GPAI-relevant provisions these AI middle-powers have enacted, mapping twenty jurisdictions including the European Union at the level of the individual provision, across four governance areas that trace the accountability chain for the model layer: systemic risk assessment, evaluation and verification, prohibitions with monitoring and detection, and serious incident reporting. Confirmed absence is recorded as data alongside positive provision. We find that jurisdictions converge on form, but diverge on force. Sixteen engage in at least three of the four governance areas, yet only about one in five provisions sit in binding law, and three-quarters of the instruments that do bind do so without defining GPAI. The institutional infrastructure shows the same shape: four in five of the mapped governance actors hold mandates that predate GPAI, and obligations attach wherever the inherited regime already reached, which is the application layer rather than the model. Where these states engage the model layer, they build capacity to observe it rather than impose duties on those who build it, and almost every evaluation body was constituted without the power to act on what it finds. Nominal coverage of the full accountability chain reaches eleven jurisdictions, but only five hold more than one provision in every area and, outside the EU, no jurisdiction imposes a binding evaluation duty on a model developer. The
dataset gives researchers and policymakers a provision-level basis for identifying where regimes could align, and where coordination would have to start from scratch.
Read the full abstract ▾
All authors Josephine Schwab, Nathan Naidoo, Ferruccio Barazzutti, Sheryn Lee, Caio Vieira Machado
Artificial Intelligence · Cross-listed from cs.SE
What Does an Evaluation License? A Commit-Bound Census of Claim Replay in Inspect Evals
Xi Qin and Jizhou Tong
Benchmarks can run without determining what their results license. We freeze a large evaluation collection and attempt to replay its historical claims. Most units stop because the evidence required for replay is not bound. Where replay is possible, different claims remain stable at different resolutions. We make this otherwise implicit inference step explicit and executable.
Read the full abstract ▾
Artificial Intelligence · Cross-listed from cs.CR
Incident-Data Robustness Analysis of the OWASP Top 10 for LLM Applications (2026): How a Community-Expert Ranking Holds Up Against a Large-Scale LLM Incident Corpus
Kyriakos "Rock" Lambros and Steve Wilson
The OWASP Top 10 for LLM Applications ranks the risks that a community of security practitioners judges most important. We ask a narrower question: checked against the record of real incidents, does that expert ranking agree with the data? We assembled a large-scale corpus of LLM-security incidents (7,714 snapshotted and 6,639 labeled against the 20-entry taxonomy) drawn from CVE, GHSA, OSV, and AIAAIC, and derived an incident-based ranking with a
Bayesian measurement-error model that corrects each category's count for classifier precision and recall. The 2026 candidate list blends the two signals at fixed weights, 0.75 on the expert vote and 0.25 on the data, so the corpus corrects the consensus without overturning it. The agreement between the two rankings is weak: Cohen's $\kappa \approx 0.20$, with a 90% interval that crosses zero. The expert ranking is nonetheless robust. A pre-registered bake-off of four frontier classifiers returns no winner. None beats the incidence floor's balanced accuracy of 0.863. A ground-truth check leaves the floor's ordering (Spearman $\rho = 0.918$ against held-out truth) in place. This is an exploratory analysis by two working-group members, not the official OWASP release, and it does not supersede the official list or process.
Read the full abstract ▾
Artificial Intelligence
A Jagged Frontier: Evaluating Robustness of Code Agents to Semantics-Preserving Transformations
Hasan Najib Mahmud, Shreya Gupta, Isha Chaudhary et al. (+4 more)
AI code
agents are increasingly deployed to resolve real software issues, yet their reliability under superficial code variations remains poorly understood. We evaluate whether coding
agents that repair repository-level issues remain reliable when the surrounding codebase is rewritten into a semantically equivalent form. We introduce a random variant sampler that applies common semantics-preserving transformations (SPTs) - spanning control-flow rewrites, dead-code injection, and identifier renaming - to produce perturbed variants. We evaluate two agentic scaffolds (mini-SWE
agent and OpenCode) each backed by one of four frontier models (Claude Opus 4.5, Kimi K2.5, MiniMax M2.5, and Qwen 3.6-27B) across instances drawn from SWE-bench Verified and SWE-bench Pro. For each instance, the
agent is run multiple times on the unperturbed and perturbed variants, yielding paired resolve-rate estimates that isolate the perturbation effect from intrinsic stochasticity. We find small degradation in most configurations: up to 6.7 percentage points mean resolve-rate drop in the most affected configurations with statistically significant degradations in 6 of 16 configurations of model, scaffold, and
dataset. Crucially, no single model ranking by robustness holds across scaffolds - Qwen is among the most robust under mini-SWE
agent on SWE-bench Verified yet the most brittle under OpenCode - revealing a jagged robustness frontier. The simpler scaffold (mini-SWE
agent) is more robust to perturbation. Our results demonstrate that even top frontier models are susceptible to semantics-preserving perturbations although the effect is not uniform, raising concerns about the deployment reliability of AI code
agents in diverse real-world codebases.
Read the full abstract ▾
All authors Hasan Najib Mahmud, Shreya Gupta, Isha Chaudhary, Nathaniel Enis, Ravi Mangal, Gagandeep Singh, Corina Pasareanu
Artificial Intelligence · Cross-listed from cs.CV
TTSD-FAR: Test-Time Self-Distillation with Fisher-Anchored Restoration for Missing-Modality Emotion Recognition in LVLMs
Muhammad Haseeb Aslam, Alessandro Koerich, Marco Pedersoli et al. (+2 more)
Large
video-
language models (LVLMs) have shown remarkable performance on
multimodal tasks like
multimodal emotion recognition (ER) in the wild. ER is inherently
multimodal, requiring a joint understanding of facial expressions, vocalizations,
language, biosignals, and gestures. However, real-world deployment remains challenging: modalities may be missing or noisy at test time. Partial observations can be viewed as a distribution shift relative to the complete-modality distribution. SOTA TTA methods based on
entropy minimization or perplexity reduction do not transfer to autoregressive LVLMs, while
retrieval augmented generation (RAG) degrades when the observed modality is weak. Because no ground-truth supervision exists to verify individual updates, adaptation across this stream risks accumulating drift and degrading once the model departs from a reliable solution. An effective solution must therefore adapt to arbitrary missing-modality patterns and remain effective during continual adaptation. We address both jointly with Test-Time Self-
Distillation (TTSD), a parameter-efficient framework in which a frozen teacher, trained on complete modalities, guides an adaptive low-rank student via self-
distillation, updating only a negligible number of parameters. Stability is built into this same loop through Fisher-Anchored Restoration (FAR), which monitors Fisher information stability to detect convergence versus drift and restores the student toward the teacher's anchor when distributional shifts are identified. Our experiments on MELD, DFEW, and BAH under 0%-50% missing modalities show that this unified adaptation-restoration design consistently outperforms
entropy-based adaptation, RAG, and perplexity-based generation over long adaptation horizons, where baselines without restoration progressively degrade while TTSD-FAR remains consistent.
Read the full abstract ▾
All authors Muhammad Haseeb Aslam, Alessandro Koerich, Marco Pedersoli, Ali Etemad, Eric Granger
Artificial Intelligence · Cross-listed from cs.LG
Selection, Recombination, or a Fresh Solve? A Candidate-Free Control for Single-Pass Test-Time Aggregation
Guiv Farmanfarmaian
An aggregation pass gives a
language model two things: candidate solutions and another chance to answer. We ask whether the candidates improve that attempt, and whether aggregation improves on selecting an existing answer. Our candidate-free control asks the same model to solve the problem again from the problem alone, using the same decoding settings and maximum output-
token allowance. On AIME-2025 and HMMT-2025, Qwen3-4B-Instruct-2507 reaches 15.5% accuracy when all four candidates are wrong, compared with 27.7% for a fresh solve. When at least two candidates are correct, the ordering reverses: 88.7% for aggregation versus 59.7% for a fresh solve. With exactly one correct candidate, the benefit remains unresolved under the specified test. Yet plurality voting reaches 96.0% in the multiple-correct regime without another model generation; its accuracy difference from aggregation is inconclusive (raw $p = 0.180$). The all-wrong accuracy loss and multiple-correct gain survive a statistical correction for adding the second
benchmark, developed after data collection, and persist when fresh responses are scored by their final answers. These findings concern one 4B model, two mathematics
benchmarks, and a single aggregation pass over candidates truncated at their first answer. All-wrong recovery can arise from solving the problem afresh. The fresh-solve control tests whether candidates improve accuracy. Comparing with selection asks whether the additional generation improves on an existing answer.
Read the full abstract ▾
Artificial Intelligence · Cross-listed from cs.SE
One Gate Is Not Enough: Composing Stateful Pre-Action Controls for Agentic AI
Gaston Besanson
Agentic AI systems take consequential actions governed by more than one pre-action control at once: authority, resource, and evidence gates that can admit, degrade, or remediate an action before it executes. This paper's central object is remediation-induced control coupling: a remediation applied by one control can change the action, evidence, or context another control evaluates, invalidating that control's earlier judgment. We formalize this coupling and give a remediate-and-regate protocol that restores per-action soundness in the current bounded, idempotent setting under its stated assumptions. We further show that the two implemented remediation operators (evidence substitution and resource-budget downroute) do not commute – a finite-model checker finds concrete counterexample instances – making remediation order part of the control-plane semantics rather than an implementation detail. A governed evidence buffer that trusts its own most recent admitted write is a further instance of the same problem at the level of state – current admissibility does not imply future reference trustworthiness – and is vulnerable to poisoning from declared-uncovered defect classes; two mitigations reduce, not eliminate, that exposure. Supporting results establish the exact condition under which positive-weight linear aggregation of gate outcomes can compensate a member veto, a unified cross-control Evidence Set, and that composition manufactures no new detection coverage, reported honestly. Empirically, on a deterministic open-data artifact composing three published engines unmodified, CH1-CH5 meet their registered decision rules across all 30 pre-registered seeds; CH6 does so under W1 but not under the smaller W2 workflow, reported as such. This is a mechanism demonstration on open payload data with a synthetic metadata layer, not a claim about production prevalence.
Read the full abstract ▾
Artificial Intelligence · Cross-listed from cs.CR
Task-Conditioned Least-Privilege Learning for Executable Terminal and MCP Agents
Alexander Tu and Michael Tu
Tool-using large
language-model
agents can complete a task while exercising authority that the user did not grant or the task does not need, causing excess-authority errors. Traditional permission gating systems alone for validating
agent environments are insufficient. We study whether post-training can teach a 4B-parameter model to choose task-conditioned authority in executable terminal and Model Context Protocol (MCP) environments to complement those measures. We propose a framework where each action is audited before execution and again from observed effects along six dimensions of risk. This
auditing is conducted using deterministic verifiers that score completion, evidence, exact state, prohibited attempts, and safe success. In conjunction with predefined task-specific sufficient-authority envelopes, we determine task-specific excess privilege values for trajectories, which are then optimized for in post-training. We find that after training using this framework on Qwen3.5-4B over 1,500 tasks, the selected seed reaches 98.48% safe success across 2,896 evaluation episodes spanning all 500 held-out tasks, compared with 64.36% for the base policy, and reduces excess-authority error events from 4.56% to 0.79%. Furthermore, external tests show capability retention and
prompt-directed improvement. A 400 task continuation study also found evidence of generalization, reducing excess-authority events by 6.99 percentage points while maintaining previous capabilities. We conclude learned restraint through least-privilege aware post-training is therefore useful as an additional control layer for tool-using
agents in executable terminal and MCP environments, but it does not replace permission gates and sandboxing.
Read the full abstract ▾
Artificial Intelligence · Cross-listed from physics.chem-ph
Coupled-cluster molecular properties across the main group that extrapolate beyond training size
Wenhao He, Xu Chen, Noah Song et al. (+10 more)
Coupled-cluster theory defines the accuracy standard for molecular electronic-structure properties but scales too steeply for routine application, whereas density-functional theory is affordable yet systematically biased. We resolve this trade-off with a single equivariant network, HARP (
Hamiltonian Read-out for Properties), that predicts an effective one-electron
Hamiltonian from one inexpensive B3LYP/def2-SVP calculation and derives a broad suite of properties from it (energy, optical gap, dipole, quadrupole, polarizability, Mulliken atomic charges, and Mayer bond orders) at coupled-cluster accuracy across nine main-group elements, including the under-served phosphorus, sulfur, and chlorine chemistries. The model is trained on a new in-house
dataset of multi-property labels computed at the CCSD(T) level for all nine elements. On a held-out test set, it reduces the error of every property by a factor of 3.8 to 270 relative to semi-local, hybrid, and double-hybrid DFT (referenced to composite CCSD(T)/cc-pVTZ), while adding only ~0.1 s wall time per molecule, delivering coupled-cluster-quality predictions at the cost of a single DFT calculation. Critically, deriving every property from a predicted
Hamiltonian rather than pooling per-atom features builds the correct size-scaling into the model architecture: on pi-conjugated oligothiophenes it matches finite-field CCSD polarizability to ~1% and the EOM-CCSD optical gap to ~3% at the largest sizes where those references remain affordable (44 and 37 atoms, where a single CCSD field point already costs ~500x the model's entire inference) and extrapolates the corrected trends to 58-atom chains, a regime where pooling-based architectures fail by construction. Accurate extrapolation is therefore set by the model's inductive bias rather than by the training data.
Read the full abstract ▾
All authors Wenhao He, Xu Chen, Noah Song, Haowei Xu, Tim S. Hindges, Bohan Li, Zihan Lin, Yu Yao, Avetik R. Harutyunyan, Fang Liu, Yao Wang, Hao Tang, Ju Li
Artificial Intelligence · Cross-listed from cs.NE
Low-Power, Neuromorphic, Acoustic Anomaly Detection for Persistent Machine Monitoring
Steven C. Nesbit, Victor M. Vergara, Michael A. Felix et al. (+4 more)
Persistent acoustic monitoring can detect machine faults without physical contact, but always-on inference is constrained by power, latency, and deployment complexity. We demonstrate autoencoder-based acoustic anomaly detection on an Intel Loihi 2 neuromorphic processor under clean and noisy conditions. Log-mel features are computed off chip; normalization, autoencoder inference, L1 reconstruction scoring, and thresholding run on chip. In a clean, microphone-position-invariant ToyADMOS ToyCar
benchmark, the on-chip model achieves 0.9959 AUC and 0.9785 standardized pAUC at maximum false-positive rate 0.1. In the DCASE 2026 Task 2 ToyCar noisy
benchmark, the model achieves source AUC 0.7990, target AUC 0.6466, and pAUC 0.6426, exceeding reported baseline metrics. Power profiling on a 16-chip Loihi 2 VPX system shows real-time throughput with 0.0406$\unicode{x2013}$0.0426 mJ dynamic energy per sample, two orders of magnitude lower than both a CPU and GPU. These results support neuromorphic acoustic anomaly detection as a practical candidate for low-power, persistent machine monitoring.
Read the full abstract ▾
All authors Steven C. Nesbit, Victor M. Vergara, Michael A. Felix, Evan T. Kain, Luis R. García Carrillo, Gerd J. Kunde, Andrew T. Sornborger
Artificial Intelligence · Cross-listed from cs.CV
From Inference to Adaptation: A Unified Optimal Transport View of Vision Language Model
Qi Yu, Zhichen Zeng, Katherine Tieu et al. (+8 more)
Vision-
language models (VLMs) have demonstrated remarkable zero-shot capabilities yet remain sensitive to real-world distribution shifts during inference. Although significant efforts are devoted to adapting VLMs at test time, they rely heavily on noisy pseudo-labels predicted directly from raw
embedding similarities during inference, which are unreliable under distribution shift and mislead the adaptation. To avoid noise amplification, existing works craft coarse-grained surrogate objectives during adaptation, which fail to explicitly model sample-level relationships across different modalities, creating objective mismatch with inference, thus leading to marginal performance improvement. In this work, we aim to bridge the detached objectives of inference and adaptation for VLMs, and propose a principled VLM TTA method called \algname. For VLM inference, we formulate the zero-shot image classification task as a cross-modal
alignment problem encoded via a Wasserstein OT formulation, providing robust pseudo-labels at the sample-level to effectively adapt VLMs. For VLM adaptation, we adopt a soft-label InfoNCE loss to adapt VLMs based on the OT-induced pseudo-labels, leveraging fine-grained supervisions to explicitly model relationships of individual image-text pairs via contrastive learning, which empowers accurate inference at the same granularity. Moreover, we theoretically reveal that the InfoNCE loss can be neatly reformulated as a Wasserstein OT formulation, thereby unifying the objectives of the inference and adaptation of VLMs to achieve their mutual benefits. Extensive experiments demonstrate the effectiveness and efficiency of our methods, outperforming the best-performing methods by up to 7% with state-of-the-art efficiency.
Read the full abstract ▾
All authors Qi Yu, Zhichen Zeng, Katherine Tieu, Xiyuan Yang, Ruizhong Qiu, Yuchen Yan, Lihui Liu, Yanjun Zhao, Lingjie Chen, Jingrui He, Hanghang Tong
Artificial Intelligence
Measuring the Partial-Credit Gap: A Strict Benchmark on Vietnam's 2025 Convex Marking Scheme
Nguyen Quoc Hung, Nguyen Dang Minh, Le Nhu Quynh et al. (+2 more)
When evaluating
language models on human exams,
benchmarks typically score each response as right or wrong and report the overall accuracy. This approach assumes that partial knowledge is worth proportional credit, an assumption that fails when an examination uses a non-additive grading scheme. The 2025 reform of Vietnam's National High School Graduation Examination demonstrates the cost of this substitution. In Part II of the exam, candidates evaluate four true/false statements per question. The grading is convex: the number of correct statements earns 0, 0.10, 0.25, 0.50, or 1.00 points. Identifying three statements correctly pays 0.50 points, not the 0.75 points that standard accuracy metrics would award. Because Part II accounts for 4.00 of the exam's 10.00 points, reporting accuracy inflates the score by rewarding partial knowledge that the state explicitly penalizes. We introduce THPT-Ladder, a
benchmark of 632 items from 21 official exams across 11 subjects, graded exactly as the ministry grades its students. The ministry publishes the marks of over a million candidates, allowing us to place models directly into the human cohort. Across eight models, the official rubric pays 0.020 to 0.159 points less per Part II question than proportional credit. This shortfall changes a model's apparent competence. For Qwen3.5-27B on the 2025 History exam, a 0.042-point shortfall drops its standing from the 90th to the 77th percentile among 481,293 candidates. A model's accuracy does not predict this penalty. At Claude Sonnet 5's accuracy level, different distributions of errors yield scores varying from 0.869 to 0.932 points per question. Official marks depend on how correct statements are grouped, meaning standard
benchmarks report a competence the institution would not certify.
Read the full abstract ▾
All authors Nguyen Quoc Hung, Nguyen Dang Minh, Le Nhu Quynh, Tran Khanh Linh, Nguyen Kieu Linh
Artificial Intelligence
Governance Records as Supervision: Verifier-Selected Self-Training for Structured Workflow Repair
Jesus Salas
Machine-verifiable workflows produce governance records linking a task contract, model attempt, verifier decision, accepted output, and target origin. We test whether verifier-admitted outputs can supervise a bounded model by consolidating occasional or expensive capability into reliable one-shot execution. On fresh, structure-disjoint PlanBench replanning cases, Qwen3-14B thinking produced 24 plans admitted by independently authored VAL. They trained the same checkpoint for non-thinking execution, without oracle targets or a stronger teacher. VAL acceptance rose from 1/80 to 57/80. A prospective replication held targets, model revision, recipe, and evaluation corpus fixed across eight LoRA seeds and three inference realizations per seed. Every seed produced a clear lift: adapters reached 45/80 to 70/80 against 1/80 for every matched base report; the exact seed-level sign-flip test gave p=0.0078125. Target-selection performance was less stable. An initial matched seed gave 102/160 accepted plans after VAL selection versus 69/160 after blinded model self-selection. Across eight prospective seeds, the contrast was seed-dependent, included one clear reverse seed, and did not replicate (p=0.3672). VAL also had a positive descriptive aggregate over schema-only selection but failed its preregistered seed-level reliability gate (p=0.0703). The verifier remains the admission authority; no reliable downstream capability advantage of semantic selection is established. A complementary Phi arm supports stronger-teacher
distillation. Earlier synthetic studies bound teachability, cumulative learning, transfer, and stopping. The evidence supports robust consolidation of one fixed, machine-checkable capability, not arbitrary planning, enterprise validity, or unrestricted self-improvement.
Read the full abstract ▾
Artificial Intelligence · Cross-listed from cs.CV
FedCoRe: Target-Adaptive Completion for Missing Modalities in Healthcare Federated Learning
Holger R. Roth, Ziyue Xu and Peter Cnudde
Federated multimodal models often assume every site has every modality, although hospitals differ in access to EHRs, chest radiographs, and ECGs. We study this setting on a MIMIC-derived respiratory deterioration task with simulated FL clients and introduce FedCoRe (
Federated Cross-Modal Representation Completion). FedCoRe learns representation- or logit-space corrections rather than generating synthetic ECGs or CXR images. When a client observes a modality that may be missing at deployment, it evaluates the same example with and without that modality to obtain paired supervision. Only clients with such pairs update the completion module, and validation may retain the unchanged prediction. We freeze the trained
multimodal predictor during evaluation so that measured differences come only from completion. Hiding ECG reduced AUROC by about 0.085; paired-example FedAvg restored 0.0415 AUROC, or 49.0% of the lost performance. We therefore report two distinct effects: paired-example FedAvg partially recovers the missing-ECG gap, while validation-selected completion is a task-specific classifier-logit correction rather than literal ECG recovery. For CXR, effect-aware completion recovers 52.8% of the loss in a controlled test where CXR is hidden. Paired-example FedAvg transfers part of this effect, but validation keeps the no-completion baseline for deployment cases whose inputs lack CXR. Thus, FedCoRe should be read as a validation-gated completion/correction framework: it can recover missing-modality signal in supported settings, but it should be deployed only when paired examples and validation evidence support that modality.
Read the full abstract ▾
Artificial Intelligence
ComponentBench: Diagnosing Component-Level Failures in Computer-Use Agents
Tianchen Guan, Xinlei Lin, Royce Cheng-Yue et al. (+2 more)
Current evaluation of computer-use
agents is split between long-horizon workflow
benchmarks and atomic GUI-grounding tests. This leaves an under-instrumented middle layer: realistic component-centered interactions (e.g., toggle a button set) that are short enough to diagnose and rich enough to capture the burdens of modern interfaces. We present ComponentBench, a
benchmark and diagnostic pipeline for component-level evaluation of computer-use
agents on modern web UIs. ComponentBench is organized around a library-agnostic ontology of 97 canonical UI components instantiated as 2,910 programmatically verified tasks across widely used component libraries, paired with cleaned human reference trajectories that enable evaluation of both task success and interaction efficiency. Beyond task collection, we introduce a scalable pipeline for
auditing realized structural difficulty after implementation and synthesizing structured failure analyses across tasks and component families. Evaluating seven models – GPT-5.4, Gemini 3 Flash, GPT-5.4 mini, GPT-5 mini, Gemini 3.1 Flash-Lite, Qwen3-VL-235B, and UI-TARS-1.5-7B – across four observation and action spaces, we show that these design choices critically impact performance. Within a single shared harness, changing only the observation and action space shifts task success by more than 30% for the same model: GPT-5 mini falls from 83.1% with accessibility-tree observations to 48.9% with coordinate-only Pixel control. Moreover, even the fastest configuration takes 3.7x as long as the matched human reference, and spatial manipulations that are trivial for humans continue to challenge current
agents.
Read the full abstract ▾
All authors Tianchen Guan, Xinlei Lin, Royce Cheng-Yue, Xiangjun Wang, Shuyan Zhou
Artificial Intelligence
SESSE: Sketch, Expand, Sort, Summarize, Evaluate – LLM-as-Judge Evaluation via Structured Decomposition
Dae Lee, Mihai Delgeanu and Adel Youssef
LLM-as-judge evaluation reduces response quality assessment to a single holistic A/B preference choice, providing no mechanism to isolate which quality dimensions drove the preference or distinguish model errors from genuine label ambiguity. We propose SESSE (Sketch, Expand, Sort, Summarize, Evaluate), a training-free framework that decomposes holistic judgment into structured sub-questions mined directly from the judge's own error cases; requiring no oracle responses, task-specific rubrics, or fine-tuning. On RewardBench (n=1,000), SESSE achieves near-parity with the chain-of-thought baseline and is competitive with RISE-Judge-32B (92.7%), a fine-tuned specialist, while remaining fully training-free. Per-criterion vote evidence provides an interpretable audit trail for diagnosing label ambiguity and judge failure modes unavailable from a single holistic output
token.
Read the full abstract ▾
Artificial Intelligence
The Lifecycle of LLM-as-a-Judge for Large-Scale Recommendation Explanations
Emma Yanyang Kong, JJ Tan, Ishan Gupta et al. (+8 more)
LLM-as-a-Judge, which leverages a large
language model to evaluate natural
language generated by another AI application or model, has become a standard, scalable approach for accelerating and extending costly human evaluation. Yet most work treats a judge as a static artifact, evaluating it once at construction or against a fixed
benchmark. We argue instead that an LLM judge operating in a deployed system is better understood as having a lifecycle. It must be built, trained, deployed, and continuously maintained as the surrounding data evolves, and each phase poses distinct technical and operational challenges. We present such a lifecycle for the LLM judges that evaluate recommendation explanations at Netflix. Everything we report comes out of a series of controlled online member-facing experiments, in which our pipeline generated and the judges assessed hundreds of thousands of distinct show-level explanations per week across a changing catalog. Our framework has four phases. (I) Birth defines the evaluation criteria and builds curated
benchmark datasets with human labels and rationales. (II) Training refines the judges' rubrics via
Reasoning-Aligned Rubric Tuning (RART), which uses a meta-judge over
reasoning output as the learning signal. (III) Deployment puts one judge in two online roles, quality gating and reflective generation. (IV) Monitoring runs a continuous Human-in-the-Loop (HITL)
alignment process that detects drift and triggers re-tuning behind a human review gate. We report results from a five-week online A/B test over tens of millions of members on the Netflix mobile app, in which judge-aligned explanations shifted member viewing toward novel content (previously unwatched) and increased successful browse-to-play sessions relative to a no-explanation control, with no quality-related escalations.
Read the full abstract ▾
All authors Emma Yanyang Kong, JJ Tan, Ishan Gupta, Lars Olds, Claire Campbell, David Fagnan, Ratna Kavuri, Veli Balin, Rohan Gosain, Louis Garcia, Minsu Jang
Artificial Intelligence · Cross-listed from cs.CY
FairGlucose: A CGM Fairness Benchmark Reveals Subgroup Disparities Hidden in Population-Level Validation
Junjie Luo, Xuzhe Zhi, Rui Han et al. (+5 more)
As CGM-based AI tools approach clinical deployment, whether their accuracy is equitable across patient demographics remains insufficiently tested. To enable this evaluation, we constructed FairGlucose, a 300-patient CGM cohort balanced across 12 demographic strata (age x gender x type 1/type 2 diabetes), with 132,480
forecasting samples and 3,945 unique behavioral events (meals, exercise, medication) logged by 81 patients. Benchmarking 33 models across four families on 2-hour glucose
forecasting, we find that population-level external validation can conceal substantial subgroup disparities. Aggregate out-of-distribution metrics appear stable (approximately 1.0), yet subgroup-level ratios range from 0.8 to 1.4, with T1D patients showing 6 mg/dL higher prediction error than T2D (p < 0.001). This disparity persists across all 33 models, suggesting a property of the prediction task rather than any single architecture. Further analysis shows that subgroup performance gaps align with the proportion of clinically hard cases, and that input-length sensitivity varies across demographics, motivating personalized configurations. Frontier LLMs underperform specialized neural models by 1-6 mg/dL; behavioral events contribute negligibly (approximately 0.1 mg/dL) even under oracle event access. These findings establish that population-level validation alone is insufficient for equity assessment of digital health AI, motivating subgroup-disaggregated reporting as a default standard.
Read the full abstract ▾
All authors Junjie Luo, Xuzhe Zhi, Rui Han, Abhimanyu Kumbara, Anand K. Iyer, Mansur E. Shomali, Ritu Agarwal, Guodong Gordon Gao
Artificial Intelligence · Cross-listed from stat.ME
Debiased Inference for AI-Generated Data without Gold-Standard Labels: Identification via Multiple Imperfect Measurements
Naoki Egami and Sooahn Shin
An increasing number of scholars use AI to measure variables they subsequently include in downstream analyses. Although AI-measured variables are often analyzed as if observed without error, ignoring prediction errors in automated measurement leads to substantial bias and invalid confidence intervals in downstream analyses, even if AI measurement accuracy is high, e.g., above 90%. Existing solutions, such as design-based supervised learning and prediction-powered inference, combine error-prone AI-based measurements with gold-standard labels, which may be costly and difficult to obtain in some application areas. In this paper, we propose debiased inference with multiple imperfect measurements (DMM), a framework that combines multiple error-prone AI measurements to enable valid downstream inference without gold-standard labels. Building on the established results on CP decomposition, DMM assumes that these measurements are independent conditional on the latent true label and observed unit-level features, such as text features represented by
embeddings. This framework allows for unknown misclassification rates to vary across annotation methods (e.g., large
language models) and across units of annotation (e.g., texts). Under this assumption, we use semiparametric inference theory to prove that the DMM
estimator is consistent and asymptotically normal, enabling valid inference for a wide range of downstream statistical analyses common in the social sciences. Our simulation results show that DMM yields valid inference and that adding accurate, though imperfect, measurements can improve efficiency. Focusing on common applications of large
language model annotations, we also develop diagnostics to assess the conditional independence assumption.
Read the full abstract ▾
Artificial Intelligence
Evaluating Structured Information Extraction with Open Models in a High Risk Public Sector Application
Elias Schubert and Felix Bießmann
The extraction of structured information from unstructured documents represents a critical component of digital transformations in all sectors. While proprietary solutions dominate commercial applications, a rapidly growing ecosystem of open-source Optical Character Recognition (OCR) engines, Large
Language Models (LLMs), and
Vision-
Language Models (VLMs) offers accessible alternatives. However, systematic evaluations on realistic, multi-step extraction pipelines remain scarce. Responsible usage of such extraction tools require comprehensive evaluations on realistic tasks, especially as these solutions will be key components of applications in the public sector that the EU AI act categorizes as high risk. To address this gap we present a comprehensive
benchmark assessing the end-to-end performance of open-source systems on a complex real-world document processing task classified as high risk: Student applications for an international study program. We conduct a comprehensive empirical evaluation with state-of-the-art OCR engines, LLMs and VLMs. Our results reveal that while VLMs generally outperform OCR+LLM pipelines, even state-of-the-art open-source models struggle to handle such tasks reliably in zero-shot settings. Only 4 of 35 configurations achieved F1 scores above 0.5, with the best OCR+LLM pipeline matching top VLM performance, though most OCR+LLM combinations performed substantially worse. Roughly 75% of all configurations scored below 0.25. Model scale influences performance, yet the relationship is non-linear: substantially larger models do not guarantee proportionally better results. Input quality, particularly the structural preservation of OCR output, emerges as a critical factor independent of downstream model capability.
Read the full abstract ▾