论文广场 - AcademicHub

01.

arXiv (CS.LG) 2026-06-16 DOI: arXiv:2606.16511

Tail-Shape Estimation in LLM Evaluation Is Fragile: A Protocol for Diagnosing False Positives

作者:

arXiv:2606.16511v1 Announce Type: new Abstract: Recent work motivates moving large language model (LLM) evaluation from mean-based to tail-aware metrics, including conditional value-at-risk and tail-index estimates of reward-model error. We ask whether the canonical extreme-value-theory tail-index parameter, which isolates how heavy a tail is from how large the tail mass is, adds discriminative information beyond the mean and a standard tail-magnitude statistic in LLM evaluation. We pre-register a protocol covering admissibility, goodness-of-fit, threshold-stability, and effect-size requirements for any positive tail-shape claim. The protocol is the contribution of this paper; the empirical study below is a demonstration of what its gates catch. Applied to a standard LLM toxicity-evaluation setup under two structurally different scorer families, the protocol catches three distinct modes of false positives that a naive analysis would have published, and rejects the headline tail-shape claim on both scorers. We conclude that tail-shape estimation in the LLM toxicity-evaluation setups we examined is more fragile than the recent literature suggests, and recommend the protocol as a starting point for tail-index claims in similar setups.

阅读与讨论 → 访问原文 →

02.

arXiv (CS.AI) 2026-06-16 DOI: arXiv:2605.22664

MBABench: Evaluating LLM Agents on End-to-End Spreadsheet Tasks in Finance

作者:

Thomson Yen ↗Julian Poeltl ↗Harshith Srinivas Gear ↗Yilin Meng ↗Joshua Fan ↗Adam Shen ↗Yili Liu ↗Ali Bauyrzhan ↗Siri Du ↗Haoyang Liu ↗Daniel Guetta ↗Hongseok Namkoong ↗…

arXiv:2605.22664v3 Announce Type: replace Abstract: LLM agents are increasingly expected to carry out end-to-end workflows, producing complete artifacts from high-level user instructions. To meet enterprise needs, frontier AI labs have developed agents that can construct entire spreadsheets from scratch. This is especially relevant in finance, where core workflows such as financial modeling, forecasting, and scenario analysis are commonly conducted through spreadsheets. Yet, existing spreadsheet benchmarks do not measure this advanced capability, focusing instead on question-answering or single-formula edits. To address this gap, we provide one of the first evaluations of agents on end-to-end spreadsheet tasks, focusing on economically critical financial workflows such as modeling and scenario analysis. Since deliverables therein are routinely reviewed and revised by multiple stakeholders, judging their quality necessarily involves high-level criteria such as readability or ease of modification. To reflect the multidimensional nature of solution quality, we develop an evaluation taxonomy comprising three dimensions: Accuracy, Formula, and Format, each comprising fine-grained criteria that reflect professional standards. The Claude family leads the benchmark and produces the most professional-looking outputs in our qualitative review, but even the strongest agents frequently fall short of professional finance standards and degrade sharply as the difficulty increases beyond a few chained calculations. This suggests that current agents are not yet able to reliably produce professional-quality spreadsheets at the level of complexity real-world workflows demand.

阅读与讨论 → 访问原文 →

03.

arXiv (CS.CV) 2026-06-16 DOI: arXiv:2606.14758

Disentangling Hallucinations: Orthogonal Semantic Projection for Robust Interpretability

作者:

Emirhan Bilgi\c{c}↗Baptiste Caramiaux ↗Zhi Yan ↗Gianni Franchi ↗

As Vision-Language Models are increasingly deployed in safety-critical applications, the trustworthiness of their explanations becomes crucial. Explainable AI (XAI) methods for Vision-Language Models often suffer from semantic hallucination, where attribution maps highlight prominent image regions even when prompted with incorrect text descriptions (e.g., highlighting a dog when prompted ``cat''). Although this problem is widespread, a formal mathematical analysis of XAI methods and CLIP embeddings is largely missing in the literature. We demonstrate that this phenomenon is not specific to a single architecture but is a fundamental consequence of Linear Semantic Leakage in high-dimensional embedding spaces. We propose a unified theoretical framework, Linear Semantic Attribution (LSA), which generalizes across discriminative methods. We introduce OSP, a geometric intervention that utilizes the residual property of OMP to disentangle unique semantic signals from shared concepts. We prove theoretically and demonstrate empirically that OSP minimizes hallucination by orthogonalizing the query vector against distractor concepts, rendering the attribution model blind to shared features while preserving fidelity for correct prompts. Our code is available at: https://github.com/emirhanbilgic/Orthogonal-Semantic-Projection

阅读与讨论 → 访问原文 →

04.

arXiv (quant-ph) 2026-06-17 DOI: arXiv:2606.18084

Cavity-enhanced superconducting response in an underdoped cuprate

作者:

arXiv:2606.18084v1 Announce Type: cross Abstract: Superconductors carry electrical current without resistance when paired electrons condense into a coherent macroscopic quantum state. In underdoped cuprates, evidence suggests that pairing-related correlations and superconducting fluctuations can survive above the temperature at which global coherence is lost, pointing to phase fluctuations as a key limitation on superconductivity in this regime. Motivated by recent demonstrations of cavity-modified collective states in quantum materials, we investigate whether superconducting coherence can be stabilized by engineering the electromagnetic environment of the superconductor. We study an underdoped YBa$_2$Cu$_3$O$_{7-\delta}$ thin film in a tunable terahertz cavity formed with a semi-transparent gold mirror. From temperature-dependent terahertz transmission measurements, we find that the cavity enhances the superconducting response below the critical temperature, with an increase of the inferred superfluid weight. The effect becomes more pronounced at smaller cavity lengths and is accompanied by an upward shift of the superconducting onset temperature. Calculations based on a cavity-coupled model for phase-fluctuating superconductors capture these trends and support an interpretation in terms of cavity-enhanced phase stiffness. These results showcase the potential of cavity engineering for designing emergent functionalities in correlated systems.

阅读与讨论 → 访问原文 →

05.

arXiv (CS.LG) 2026-06-12 DOI: arXiv:2606.12709

Smarter Saboteurs, Better Fixers: Scaling & Security in Linear Multi-Agent Workflows

作者:

Timothy McAllister ↗Sina Abdidizaji ↗Ivan Garibay ↗Ozlem Ozmen Garibay ↗

arXiv:2606.12709v1 Announce Type: cross Abstract: As LLM-based multi-agent systems (MAS) are deployed in the wild, the resilience of their collaboration structures against adversarial compromise becomes a critical safety concern. Attackers may leverage prompt-injection or jailbreaking to sabotage individual agents within MAS workflows, but the interaction between model scaling and system-level resilience remains poorly understood. This paper investigates how model scale affects the security of linear multi-agent workflows. Our experiments across scales of two open-weight model families on the HumanEval benchmark reveal a compliance-correction symmetry: larger models are far more likely to faithfully execute malicious instructions, with the control-to-malicious performance drop reaching 53.7pp at 27B in uncorrected pipelines. However, appending a lightweight terminal Fixer stage collapses this to 0.6pp and restores statistical parity with control-level performance, demonstrating that strictly linear collaboration structures can be viable and resilient to adversaries at this scale, and suggesting that the brittleness previously attributed to linear topology may stem from a lack of correction.

阅读与讨论 → 访问原文 →

06.

arXiv (CS.CL) 2026-06-17 DOI: arXiv:2606.17229

Rift: A Conflict Signature for Deception in Language Models

作者:

Petr Nyoma ↗

A model that lies while knowing the truth is the central case ELK cannot handle with behavioral evaluation alone. We ask whether such deception leaves an internal signature distinguishing it from honest error. Our key move is a control for wrongness: we contrast a sleeper agent (knows the truth, lies on trigger) against a naive liar (fine-tuned to emit the same wrong answers with no honest training). Both produce identical wrong outputs; any difference is about knowledge conflict, not incorrectness. We find deceptive forward passes carry a conflict signature - 2.1-2.3x higher residual rank than naive-liar passes on the same wrong answer - strong enough to identify which of two responses is the lie with 100% accuracy and no labels, across GPT-2 small/medium (three seeds) and three instruct models. Across Qwen2.5-1.5B/7B and Phi-3-mini, instructed deception raises residual rank on every tested fact (18/18, 40/40, 34/34); on Phi-3, lies separate perfectly from both honest answers and hallucinations (AUC 1.0, Wilcoxon p~6e-11). The signature survives strategic self-constructed deception (model invents its own lie, AUC 1.0), active concealment attempts (AUC 1.0), and length-controlled replication (20/20, AUC 1.0, p~1e-6). Using basis-free relative representations, a probe trained on one model family detects deception in two other families zero-shot (mean AUC 0.933), surviving simultaneous architecture and format change (AUC 0.821), and transfers across five languages (AUC 1.000, length-controlled). The signature is read-only: detectable but not injectable (0/8 both directions). Honest limitations and six negative experiments are documented in full.

阅读与讨论 → 访问原文 →

07.

arXiv (CS.CV) 2026-06-11 DOI: arXiv:2606.12142

AerialClaw: An Open-Source Framework for LLM-Driven Autonomous Aerial Agents

作者:

Ke Li ↗Jianfei Yang ↗Luyao Zhang ↗Guo Yu ↗Chengwei Yan ↗Yuan Ding ↗Di Wang ↗Nan Luo ↗Gang Liu ↗Xiao Gao ↗Quan Wang ↗

Unmanned aerial vehicles (UAVs) are increasingly used in inspection, search and rescue, environmental monitoring, and emergency response. However, most UAV applications still rely on pre-defined command sequences or task-specific pipelines, where developers manually connect perception, planning, flight control, simulation, logging, and safety modules. This limits the flexibility, reproducibility, and extensibility of autonomous aerial systems. This paper presents AerialClaw, an open-source software framework that enables UAVs to operate as decision-making aerial agents rather than merely command-following platforms. Given a natural-language mission, AerialClaw allows an LLM-based agent to understand the task, maintain context, invoke executable aerial skills, observe perception and runtime feedback, and iteratively update its decisions in a closed loop. The framework adopts a modular brain-skill-runtime architecture, combining hard skills for atomic UAV operations, Markdown-based soft skills for reusable task strategies, document-driven agent state and capability boundaries, memory-driven reflection, safety-oriented runtime validation, and platform-agnostic execution adapters. AerialClaw supports lightweight mock execution, PX4 SITL with Gazebo, and AirSim-based simulation, together with a web console, pluggable model backends, example missions, simulation assets, and staged deployment scripts. By combining standardized aerial skills, document-driven agent state, memory, and closed-loop LLM decision-making, AerialClaw provides a reproducible and extensible open-source framework for building UAV systems that can interpret missions, make decisions, execute skills, and adapt their behavior from feedback.

阅读与讨论 → 访问原文 →

08.

arXiv (CS.CV) 2026-06-17 DOI: arXiv:2606.17403

Bridging Spatial And Frequency Views For Disaster Assessment: Benefits And Limitations

作者:

Shikha V. Chandel ↗Yadav Raj Ghimire ↗Timothy Agboada ↗Leila Hashemi-Beni ↗

Rapid assessment of building damage from satellite imagery is essential for effective disaster response and recovery. While most deep learning methods rely on spatial-domain features, frequency-domain representations can capture complementary structural cues such as debris patterns and collapse-induced textures. This study presents a controlled comparison of spatial-domain, frequency-domain, and dual-domain deep learning approaches for multi-class building damage classification using post-disaster imagery from the xView2 (xBD) dataset. To ensure fairness, all models are built on an EfficientNet-B0 backbone and trained under identical settings, differing only in their input representations and fusion strategies. Performance is evaluated using accuracy, macro F1-score, per-class metrics, and confusion matrices. Results show that dual-domain models provide measurable improvements over single-domain approaches. The dual spatial configuration achieves the highest test accuracy (0.4688) and lowest loss, while the spatial-only model attains the best macro F1-score (0.4254), indicating more balanced class performance. In contrast, frequency-only models perform worst and exhibit overfitting, suggesting limited generalization. Despite these gains, all models struggle to detect subtle damage levels, particularly the Minor class, due to class imbalance and fine-grained visual ambiguity. While dual-domain approaches improve detection of severe damage, challenges remain. These findings highlight the benefits and limitations of hybrid representations and motivate future work on data balancing, advanced fusion, and regularization.

阅读与讨论 → 访问原文 →

09.

arXiv (CS.CV) 2026-06-16 DOI: arXiv:2606.16234

Propagating Structural Guidance: Synthesizing Fluorescein Angiography from Fundus Images and Sparse OCT Scans

作者:

Tengfei Ma ↗Ruiqi Wu ↗Chenran Zhang ↗Ye Geng ↗Na Su ↗Xiangyuan Duanmu ↗Tao Zhou ↗Yi Zhou ↗Wen Fan ↗

Fundus fluorescein angiography (FFA) is critical for assessing retinal vascular abnormalities, but its acquisition is invasive and not always feasible. In contrast, color fundus photography (CFP) is non-invasive and widely accessible, which has motivated studies on CFP-to-FFA synthesis. However, prior works rely solely on CFP surface texture, fundamentally limiting the ability to reconstruct functional vascular information and subtle pathological changes. To address this, we propose a novel framework that synthesizes FFA from CFP with structural guidance provided by optical coherence tomography (OCT). We construct a multi-modal retinal imaging dataset with paired CFP, FFA, and OCT from 3,676 patient eyes–the first tri-modally aligned dataset in retinal imaging. To bridge the spatial gap between OCT and fundus modalities, we propose a Spatially Aligned Cross-Modal Fusion (SACMF) module that projects depth-resolved OCT features onto the fundus plane and injects them into the CFP encoder via adaptive layer normalization. Beyond feature fusion, we further introduce Token-wise Cross-Modality Alignment (TCMA), a token-level contrastive learning strategy that explicitly aligns CFP and FFA representations at corresponding spatial positions. Our method achieves superior synthesis performance compared to state-of-the-art methods. Moreover, extensive experiments demonstrate that the FFA images synthesized by our approach bring greater improvements in downstream disease diagnosis performance than existing methods, highlighting the clinical potential of our approach as a non-invasive decision-support tool in routine workflows. The code is available at https://github.com/while-plus/OCT-guide-FFA-Syn.

阅读与讨论 → 访问原文 →

10.

arXiv (CS.AI) 2026-06-19 DOI: arXiv:2606.20518

FlowEdit: Associative Memory for Lifelong Pronunciation Adaptation in Flow-Matching TTS

作者:

Harshit Singh ↗Ayush Pratap Singh ↗Nityanand Mathur ↗

arXiv:2606.20518v1 Announce Type: new Abstract: Flow-matching text-to-speech systems achieve remarkable zero-shot quality but remain static after deployment: pronunciation errors on out-of-vocabulary proper nouns persist unless the model is retrained. We introduce FlowEdit, a life-long adaptation framework for frozen flow-matching TTS that learns pronunciation corrections as latent conditioning edits rather than weight updates. When corrective feedback is provided, FlowEdit optimizes a token-level perturbation in the text embedding space, then stores the correction in a Modern Hopfield Network serving as content-addressable episodic memory. At inference, corrections are retrieved via soft attention with a similarity gate, enabling fuzzy morphological matching. On our curated benchmark of 312 multilingual proper nouns across 18 language families, FlowEdit reduces target-word Phoneme Error Rate by 92.7% relative to the zero-shot baseline while maintaining identical general-speech quality. Corrections complete in approximately 15 seconds on a single GPU.

阅读与讨论 → 访问原文 →

11.

PLOS Medicine 2026-05-13 DOI: HASH:09e1b7c2032e6adf96b8b52c7c1c674e

Contribution of nosocomial transmission to <i>Klebsiella pneumoniae</i> neonatal sepsis in Africa and South Asia: An observational study of infection clusters inferred from pathogen genomics and temporal data

作者:

Erkison Ewomazino Odih ↗

by Erkison Ewomazino Odih, Jabir A. Abdulahi, Anne V. Amulele, Matthew Bates, Eva Heinz, Weiming Hu, Kajal Jain, Rindidzani Magobo, Courtney P. Olwagen, John M. Tembo, Tolbert Sonda, Jonathan Strysko, Caroline C. Tigoi, Kyle Bittinger, Jennifer Cornick, Ebenezer Foster-Nyarko, Wilson Gumbi, Steven M. Jones, Chileshe L. Musyani, Carolyn M. McGann, Ahmed M. Moustafa, Patrick Musicha, James C. L. Mwansa, Moreka L. Ndumba, Thomas D. Stanton, Donwilliams O. Omuoyo, Oliver Pearse, Laura T. Phillips, Paul J. Planet, Charlene M. C. Rodrigues, Fatou Secka, Kirsty Sands, Erin Theiller, Allan M. Zuza, Sulagna Basu, Grace J. Chan, Kenneth C. Iregbu, Jean-Baptiste Mazarati, Semaria Solomon Alemayehu, Timothy R. Walsh, Rabaab Zahra, Angela Dramowski, Sombo Fwoloshi, Appiah-Korang Labi, Lola Madrid, Noah Obeng-Nkrumah, David Ojok, Boaz D. Wadugu, Andrew C. Whitelaw, Anudita Bhargava, Atul Jindal, Ramesh K. Agarwal, Alexander M. Aiken, James A. Berkley, Susan E. Coffin, Nicholas A. Feasey, Nelesh P. Govender, Davidson H. Hamer, Shabir A. Madhi, Mari Jeeva Sankar, Kelly L. Wyres, Kathryn E. Holt Background Klebsiella pneumoniae is the leading cause of sepsis among neonates in low- and middle-income countries (LMICs) in Africa and Asia, contributing substantially to the overall burden of antimicrobial-resistant infections and mortality among neonates globally. Pathogen sequencing has been used to investigate case clusters and confirm nosocomial transmission in a small number of neonatal units. Here we utilise pathogen sequence data to estimate the fraction of K. pneumoniae neonatal sepsis attributable to nosocomial transmission in African and South Asian countries. Methods and findings We estimated the proportion of invasive K. pneumoniae disease involved in nosocomial transmission clusters in a given neonatal unit, using single-linkage clustering based on pairwise temporal and genetic distances estimated from bacterial whole-genome sequences aggregated from 10 contributing studies. Analysing 1,523 K. pneumoniae isolates from 27 units in 13 countries in Africa and South Asia between 2013 and 2023, we inferred 156 nosocomial transmission clusters, ranging from 2 to 188 neonates each (83 of the clusters comprised ≥3 cases). Overall, we estimated that 1,035 neonatal infections (68.0%) were part of nosocomial transmission clusters. Excluding the first infection in each cluster as a potential index case, we estimate at least 879 (57.7%) infections were acquired via nosocomial transmission. Sensitivity analyses showed that results were robust to the choice of genetic distance estimation methods and thresholds used to define clusters, and cluster estimates were stable over temporal distance thresholds ranging from 2 to 8 weeks. Isolates were mostly extended-spectrum beta-lactamase (ESBL) producers (90.9%) and included 172 multi-locus sequence types (STs). Fourteen STs, including several globally recognised multidrug-resistant lineages, were associated with transmission clusters at multiple units, and these were collectively responsible for two-thirds of all infections. Carriage of carbapenemase genes (adjusted odds ratio, aOR = 2.08 [95% confidence interval, CI: 1.04, 4.14]; p = 0.04) and ESBL genes (aOR = 2.48 [95% CI: 1.26, 4.90]; p = 0.006) were significantly positively associated with transmission in a logistic regression model with site as a covariate. Limitations of this study include the lack of sufficient clinical data to allow high-resolution investigation of transmission dynamics and lack of facility-level data to investigate contributors to the observed differences in transmission burden across sites. Conclusions Nosocomial transmission contributes to a substantial proportion of K. pneumoniae sepsis in neonatal care units in Africa and South Asia. Reducing transmission within these settings through improved infection prevention and control and other measures could substantially reduce the neonatal sepsis burden. A high burden of transmission clusters is associated with the same drug-resistant lineages that are recognised as high-risk clones associated with hospital outbreaks in high-income countries, indicating global connectivity of the antimicrobial-resistant pathogen population.

阅读与讨论 → 访问原文 →

12.

medRxiv (Medicine) 2026-06-23 DOI: HASH:1cf973aafa93507cab684f5c9cc58815

Antibodies against influenza A/H1N1pdm2009 and B/Victoria strains but not A/H3N2 are increased in recent onset type 1 narcolepsy versus matched controls

作者:

Yan ↗Lin ↗Guillard ↗Zhang ↗Macaubas ↗Pizza ↗Biscarini ↗Plazzi ↗Mallajosyula ↗Davis ↗Maecker ↗Mignot ↗…

Study Objectives: Onsets of Narcolepsy type-1 (NT1) increased following A/H1N1 vaccination with PandemrixTM in Europe and with A/H1N1pdm2009 infections in China and other countries. To test if other strains could trigger narcolepsy, we measured strain-specific antibodies in patients with recent onset NT1 compared to controls. Methods: Antibodies against hemagglutinin (HA) and neuraminidase (NA) were tested in 62 patients with very recent onset (onset and blood collection following a single flu season, mean +/- SEM: 0.44 +/- 0.06 years since onset) and 100 controls matched by age, sex, season and year of collection (2000-2025). Results were next extended to 181 recent onset patients (mean +/- SEM: 1.00 +/- 0.05 years) versus 260 controls, matched by sex, season and year, but having a slightly higher mean age. HA inhibition (HAI) and NA inhibition (NAI) assays were conducted using flu strains known to circulate during the corresponding flu seasons. HAI results are shown as % positive (titers >= 40) and NAI results as geometric mean titers. Odds ratio (OR) and coefficient were used to compare antibody titers in NT1 versus controls. The contribution of each assay to prediction was finally quantified in the larger sample set using Shapley decomposition. Results: NT1 patients had increased anti-HA and anti-NA antibodies against A/H1N1pdm2009 (anti-HA OR = 3.86, anti-NA coefficient = 0.35) and B/Victoria (anti-HA OR =1.90, anti-NA coefficient = 0.22), but not A/H1N1pre2009, A/H3N2, or B/Yamagata, independent of HLA-DQB1*06:02 status, age, sex, and flu season. Correlations between anti-HA and anti-NA antibodies titers were weak to moderate but significant (r2=-0.10 to 0.34). Multivariable model outperformed age-only baseline (McFadden R2 = 0.19 vs. 0.03; AUC = 0.79 vs. 0.64; likelihood-ratio test X2 = 51, p

阅读与讨论 → 访问原文 →

13.

arXiv (CS.CL) 2026-06-16 DOI: arXiv:2606.16118

Know Your Limits : On the Faithfulness of LLMs as Solvers and Autoformalizers in Legal Reasoning

作者:

Olivia Peiyu Wang ↗Sanna Wong-Toropainen ↗Daneshvar Amrollahi ↗Ryan Bai ↗Tashvi Bansal ↗Arush Garg ↗Leilani H. Gilpin ↗

Large Language Models (LLMs) achieve strong performance on reasoning tasks, but whether this reflects faithful logical inference or heuristic approximation remains unclear. We study this question in legal entailment by comparing three paradigms, including pure LLM classification, LLM-based Formal Reasoning, and solver-based Formal Reasoning using the Z3 SMT solver, on a re-annotated subset of ContractNLI across five LLMs. Our re-annotation reveals a systematic and measurable gap between pragmatic legal interpretation and strict formal entailment, where a substantial proportion of legally sound inferences are not formally grounded without additional unstated assumptions. While introducing formal structure improves accuracy, with LLM-based Formal Reasoning achieving the highest benchmark performance, we show that this gain does not imply faithful reasoning. We identify three recurring failure modes: scope laundering, where LLMs report solver-inconsistent classifications without executing the underlying formal reasoning, producing conclusions that appear logically grounded but are not; implicit constraint blindness, where LLMs overlook logical constraints present in formal representations; and program synthesis failures, where LLMs generate incorrect Z3 code despite structured prompting. Critically, scope laundering persists across all models, raising serious concerns about the faithfulness of LLM-based formal reasoning as a proxy for symbolic execution. These results reveal a fundamental gap between benchmark accuracy and logical faithfulness.

阅读与讨论 → 访问原文 →

14.

arXiv (CS.CV) 2026-06-11 DOI: arXiv:2606.11702

MedCTA: A Benchmark for Clinical Tool Agents

作者:

Tajamul Ashraf ↗Hyewon Jeong ↗Fida Mohammad Thoker ↗Bernard Ghanem ↗

To make clinically grounded decisions, medical AI agents are expected to go beyond simple recognition and be capable of tool retrieval, evidence acquisition, and integration. Existing benchmarks largely evaluate isolated perception or single-turn question answering, and therefore provide limited visibility into failures of planning, tool recruitment, and rollout reliability. We introduce MedCTA, a benchmark for evaluating medical tool agents on clinician-validated, step-implicit tasks grounded in realistic multimodal clinical inputs, including radiology images, pathology slides, and reports. MedCTA comprises 107 real-world clinical tasks with clinician-verified executable trajectories over 5 deployed tools, and supports process-aware evaluation of tool selection, argument validity, execution stability, trajectory fidelity, and outcome quality. We benchmark 18 open- and closed-source multimodal models and find that even frontier systems remain brittle in multi-step clinical tool use: autonomous rollouts are dominated by protocol failures, premature stopping, and incorrect tool recruitment, while gold-standard tool routing yields large but still incomplete gains. These results show that strong backbone perception does not translate into reliable agentic behavior in clinical settings. MedCTA provides a rigorous testbed for auditing, diagnosing, and advancing trustworthy medical AI agents. The dataset and evaluation suite are available at https://ivul-kaust.github.io/MedCTA/

阅读与讨论 → 访问原文 →

15.

arXiv (CS.CV) 2026-06-16 DOI: arXiv:2606.17048

Exact Posterior Score Estimation for Solving Linear Inverse Problems

作者:

Abbas Mammadov ↗Ozgur Kara ↗Kaan Oktay ↗Iskander Azangulov ↗Adil Kaan Akan ↗Hyungjin Chung ↗James Matthew Rehg ↗Yee Whye Teh ↗

Diffusion and flow-based models learn powerful data priors by training a denoiser to reverse Gaussian corruption. To use this prior to solve a linear inverse problem, one needs to sample from the posterior, but the score that the prior provides is the unconditional score, not the posterior score. Existing methods either steer a fixed pretrained denoiser with approximate measurement-matching corrections, or train a conditional restoration model that abandons the denoising structure of the prior. We derive the exact posterior score in closed form for linear Gaussian inverse problems under general Gaussian interpolants, and show that posterior sampling reduces to a denoising problem at an operator-dependent shifted pivot under an anisotropic noise covariance. We turn this identity into Exact Posterior Score (EPS), a denoising training objective that preserves the input/output structure of standard pretraining and can therefore be trained from scratch or fine-tuned from a pretrained denoiser. At inference, EPS uses the same sampler as the underlying backbone, with no likelihood gradients or projections. We evaluate EPS on five linear inverse problems across FFHQ and ImageNet, where it outperforms training-free and training-based baselines on fidelity, perceptual, and distributional metrics, while using roughly an order of magnitude fewer denoiser evaluations than gradient-based posterior samplers.

阅读与讨论 → 访问原文 →

16.

arXiv (CS.AI) 2026-06-17 DOI: arXiv:2606.17081

The Price of Anarchy in Disaggregated Inference

作者:

Athos Georgiou ↗

arXiv:2606.17081v1 Announce Type: cross Abstract: Disaggregated inference architectures physically separate prefill and decode phases onto distinct GPU pools, creating competing "agents" that share a fixed hardware budget. We provide, to our knowledge, the first formal game-theoretic analysis of this architecture, using NVIDIA Dynamo as a concrete case study. We model disaggregated serving as three coupled games: a two-player resource game between prefill and decode pools, a selfish caching game over the hierarchical KV cache, and a congestion game with positive externalities for request routing. We empirically validate the latter two; the P/D resource game is treated analytically (Section 9.2). We characterize how GPU saturation induces regime transitions that shift the game's payoff structure: below saturation, selfish behavior has bounded Price of Anarchy (PoA); at saturation, superlinear latency and cache externalities drive our empirical estimator PoA-hat (defined in Section 6.4) upward. Based on this analysis, we design an adaptive controller that detects saturation transitions in real time and adjusts routing parameters accordingly, shifting from cache-affinity exploitation to load-balanced congestion avoidance. We instantiate our framework on a 3-node NVIDIA B200 cluster running Dynamo with two models, Nemotron-4-340B (TP=8, full-node workers with cross-InfiniBand KV transfers) and Llama-3.1-70B (TP=4), and find the same three-regime PoA-hat structure with the same first post-knee grid point (C=128) on both models. Adaptive routing shifts each model to a better operating point. Our strongest result is on the 70B 1P/5D topology, where PoA-hat drops 3.1x (66.4 to 21.5) in the saturated phase at a 13% throughput cost. On the 70B 1P/2D, PoA-hat drops 2.2x and TTFT P99 drops 7.6x (see Section 8.5).

阅读与讨论 → 访问原文 →

17.

arXiv (CS.LG) 2026-06-11 DOI: arXiv:2602.02229

Prediction-Powered Risk Monitoring of Deployed Models for Detecting Harmful Distribution Shifts

作者:

Guangyi Zhang ↗Yunlong Cai ↗Guanding Yu ↗Osvaldo Simeone ↗

arXiv:2602.02229v2 Announce Type: replace Abstract: We study the problem of monitoring model performance in dynamic environments where labeled data are limited. To this end, we propose prediction-powered risk monitoring (PPRM), a semi-supervised risk-monitoring approach based on prediction-powered inference (PPI). PPRM constructs anytime-valid lower bounds on the running risk by combining synthetic labels with a small set of true labels. Harmful shifts are detected via a threshold-based comparison with an upper bound on the nominal risk, satisfying assumption-free finite-sample guarantees on the type-I error. We demonstrate the effectiveness of PPRM through extensive experiments on image classification, large language model (LLM), and telecommunications monitoring tasks.

阅读与讨论 → 访问原文 →

18.

arXiv (quant-ph) 2026-06-12 DOI: arXiv:2605.08341

Quantum metrology via partial quantum error correction

作者:

Yinan Chen ↗Zongyuan Wang ↗Sisi Zhou ↗

arXiv:2605.08341v2 Announce Type: replace Abstract: We introduce a method for error-corrected quantum metrology where only partial quantum error correction (QEC) is needed to suppress local noise and maintain the probe states' super-standard-quantum-limit (super-SQL) sensing performance. This stands in contrast to the existing QEC-assisted sensing schemes in Phys. Rev. Lett. 112, 080801 (2014) and Phys. Rev. Lett. 112, 150802 (2014), where a probe state is encoded into the logical subspace of a quantum code and error correction involves measurements on all checks of the code. Here, we encode the probe states into superpositions of energetically different states of the underlying quantum code. For our probe states, error correction using a subset of checks is enough to suppress noise both before and after phase imprinting. We analyze the tradeoff in noise suppression. For noise parallel to our phase imprinter of weight $l$, we achieve a suppression of $p^\delta$ where $p$ is the noise strength and $\delta = \lfloor (l+1)/2 \rfloor$. We propose an adaptive imprinter weight increasing strategy to maintain super-SQL performance as we scale up the system. In all our examples, checks and phase imprinters are chosen to be local operators avoiding non-local connectivity.

阅读与讨论 → 访问原文 →

19.

PLOS Medicine 2026-05-11 DOI: HASH:149a28e3ce6ea28b01955dbe0de39e43

Connected or chained by social media? Child and adolescent mental health in a digital era

作者:

Silja Kosola ↗

by Silja Kosola Social media has evolved from connection to compulsion, disproportionately harming children and adolescents. Addictive designs together with developmental vulnerability fuel mental health risks and highlight the urgent need for stricter age limits and stronger protections. In this Perspective, Silja Kosola outlines how social media disproportionately harms child and adolescent mental health, and argues that while recent policy changes aimed at protecting youth from social media are welcome, stricter age limits and greater accountability of social media companies are needed.

阅读与讨论 → 访问原文 →

20.

arXiv (CS.AI) 2026-06-18 DOI: arXiv:2606.19168

Beyond Safe Data: Pretraining-Stage Alignment with Regular Safety Reflection

作者:

Jinhan Li ↗Kexian Tang ↗Yihan Xu ↗Zhuorui Ye ↗Kaifeng Lyu ↗

arXiv:2606.19168v1 Announce Type: new Abstract: To achieve deeper safety alignment for large language models (LLMs), recent efforts have studied how to push safety interventions earlier into the pretraining stage, primarily by filtering unsafe data or rewriting it into safer forms. We argue that pretraining-stage alignment should go beyond making the data safe: LLMs may compose seemingly benign knowledge and capabilities into unsafe behaviors. To this end, we propose Safety Reflection Pretraining, a pretraining-stage alignment method which regularly inserts short safety reflections into pretraining corpora to integrate self-monitoring directly into language modeling, establishing a foundational capability that is subsequently reinforced by compatible post-training. Our experiments with 1.7B models pretrained on FineWeb-Edu show that Safety Reflection Pretraining improves safety classification accuracy and substantially reduces the success rates of inference-stage and finetuning attacks. Complementary to our real-world experiments, we also introduce a fully controlled synthetic environment, MedSafetyWorld, with a clear definition of safety and a reasoning structure under which models can easily generalize unsafe behaviors from safe data. Ablations in MedSafetyWorld further demonstrate a clear advantage of Safety Reflection Pretraining in preventing models from acting on unsafe behaviors generalized from safe data, compared with data filtering and rewriting. Taken together, our findings suggest that pretraining alignment should not only make the training data safe, but also shape the behaviors that models are likely to acquire from safe data.

阅读与讨论 → 访问原文 →

21.

arXiv (CS.AI) 2026-06-12 DOI: arXiv:2412.08610

Competition and Diversity in Generative AI

作者:

Manish Raghavan ↗

arXiv:2412.08610v3 Announce Type: replace-cross Abstract: Recent evidence, both in the lab and in the wild, suggests that the use of generative artificial intelligence reduces the diversity of content produced. The use of the same or similar AI models appears to lead to more homogeneous behavior. Our work begins with the observation that there is a force pushing in the opposite direction: competition. When producers compete with one another (e.g., for customers or attention), they are incentivized to create novel or unique content. We explore the impact competition has on both content diversity and overall social welfare. Through a formal game-theoretic model, we show that competitive markets select for diverse AI models, mitigating monoculture. We further show that a generative AI model that performs well in isolation (i.e., according to a benchmark) may fail to provide value in a competitive market. Our results highlight the importance of evaluating generative AI models across the breadth of their output distributions, particularly when they will be deployed in competitive environments. We validate our results empirically by using language models to play Scattergories, a word game in which players are rewarded for answers that are both correct and unique. Overall, our results suggest that homogenization due to generative AI is unlikely to persist in competitive markets, and instead, competition in downstream markets may drive diversification in AI model development.

阅读与讨论 → 访问原文 →

22.

arXiv (quant-ph) 2026-06-15 DOI: arXiv:2606.14632

No classical particle limit for massless quanta

作者:

Riccardo Falcone ↗Simon Fuchs ↗

arXiv:2606.14632v1 Announce Type: new Abstract: We investigate whether relativistic massless classical particles may emerge as the classical limit of massless quanta. To address this question independently of any specific dynamics, environment, or pointer basis, we develop an axiomatic and purely kinematical framework for the coarse-graining approach. In this formulation, a candidate classical phase space is taken as the outcome space of a POVM subject only to minimal classicality and covariance under the relevant spacetime symmetry group. Applying this framework to the Poincaré group, we prove a no-go theorem for massless particles: the covariance requirement is incompatible with the operational conditions for classicality. The theorem leaves open field-like limits of massless quanta, for example the emergence of electromagnetic or gravitational fields, while ruling out classical massless particles, such as classical photons or gravitons.

阅读与讨论 → 访问原文 →

23.

arXiv (CS.LG) 2026-06-16 DOI: arXiv:2602.05779

How Controlling the Variance can Improve Training Stability of Sparsely Activated DNNs and CNNs

作者:

Emily Dent ↗Jared Tanner ↗

arXiv:2602.05779v2 Announce Type: replace Abstract: The Edge-of-Chaos (EoC) theory developed for the random initialization of deep networks allows more efficient training by both preserving information in the initial outputs of the network and minimising exploding or vanishing gradients through characterisation of the intermediate layers as Gaussian processes. This EoC theory provides formulae for the choice of the initialisation distribution variances of the weights and biases. For activations which are approximately linear around the origin, the EoC theory typically encourages the Gaussian process variance to converge towards zero with increasing depth. Here we consider the less studied setting of highly sparsity inducing activations where a large region of values near the origin are set to zero. In this setting we prove a new phenomenon whereby initialisations leading to larger fixed Gaussian processes are beneficial to training stability. This theory informs a new, yet simple, initialisation strategy that allows training DNNs and CNNs with as large as 90\% sparsity in the hidden layers.

阅读与讨论 → 访问原文 →

24.

arXiv (CS.CV) 2026-06-11 DOI: arXiv:2510.17816

Cross-Domain Multi-Person Human Activity Recognition via Near-Field Wi-Fi Sensing

作者:

Xin Li ↗Jingzhi Hu ↗Yinghui He ↗Hongbo Wang ↗Jin Gan ↗Jun Luo ↗

Wi-Fi-based human activity recognition (HAR) provides substantial convenience and has emerged as a thriving research field, yet the coarse spatial resolution inherent to Wi-Fi significantly hinders its ability to distinguish multiple subjects. By exploiting the near-field domination effect, establishing a dedicated sensing link for each subject through their personal Wi-Fi device offers a promising solution for multi-person HAR under native traffic. However, due to the subject-specific characteristics and irregular patterns of near-field signals, HAR neural network models require fine-tuning (FT) for cross-domain adaptation, which becomes particularly challenging with certain categories unavailable. In this paper, we propose WiAnchor, a novel training framework for efficient cross-domain adaptation in the presence of incomplete activity categories. This framework processes Wi-Fi signals embedded with irregular time information in three steps: during pre-training, we enlarge inter-class feature margins to enhance the separability of activities; in the FT stage, we innovate an anchor matching mechanism for cross-domain adaptation, filtering subject-specific interference informed by incomplete activity categories, rather than attempting to extract complete features from them; finally, the recognition of input samples is further improved based on their feature-level similarity with anchors. We construct a comprehensive dataset to thoroughly evaluate WiAnchor, achieving over 90% cross-domain accuracy with absent activity categories.

阅读与讨论 → 访问原文 →

25.

arXiv (CS.CV) 2026-06-17 DOI: arXiv:2606.18008

PhaseWin: An Efficient Search Algorithm for Faithful Visual Attribution

作者:

Zihan Gu ↗Ruoyu Chen ↗Junchi Zhang ↗Li Liu ↗Xiaochun Cao ↗Hua Zhang ↗

Visual attribution is a fundamental tool for interpreting modern vision and vision-language models, particularly when their decisions must be inspected, diagnosed, or audited. Its goal is to explain how a model's decision depends on local regions of the visual input, typically by assigning an importance ordering over candidate image regions. Given an image partitioned into $n$ regions, faithful attribution can be cast as an ordered subset-search problem, in which progressively inserting the selected regions should recover the target model response as early as possible. Exhaustive search over region subsets incurs exponential cost, while the widely used greedy search still requires a quadratic number of model evaluations, because every selection step rescores all remaining candidates. We propose PhaseWin, an efficient subset-search algorithm for faithful visual attribution. PhaseWin reorganizes greedy region selection into a phased window-search procedure: rather than re-evaluating the full candidate set at every step, it alternates between global candidate screening, adaptive pruning, and localized window refinement, while preserving the essential region-ranking behavior of greedy search. We analyze PhaseWin under monotone evidence-accumulation conditions and show that, under feature-level structural assumptions, it attains controllable linear evaluation complexity together with near-greedy faithfulness guarantees. Extensive experiments on image classification, object detection, visual grounding, and image captioning show that, among all compared attribution methods, PhaseWin reaches high faithfulness with the fewest forward passes, empirically realizing the predicted reduction from $O(n^2)$ to $O(n)$. The code is available at https://github.com/Qihuai27/phasewin-va.

阅读与讨论 → 访问原文 →

探索全球前沿学术脉络