❌

Reading view

↗️

Safe to Stop? Risk-Constrained Stopping for Sequential Clinical Diagnosis Agents

βœ‡cs.AI, q-bio.NC updates on arXiv.org
By: Yuexin Wu Β· Vasile Rus
arXiv:2609.09678v1 Announce Type: new Abstract: Clinical diagnosis agents must decide not only what test to request next, but also when to diagnose or defer. Existing agent benchmarks largely evaluate accuracy after fixed or unconstrained interaction, leaving autonomous stopping reliability implicit. We present Cros, a risk-constrained stopping layer combining state-wise error ranking, policy design on disjoint development splits, and LTT-style exact tests of selective diagnostic error and minimum autonomous coverage for complete sequential policies. Its finite-sample guarantee requires the candidate family, testing rule, and any randomization to be frozen before calibration labels are accessed. On a 1,834-episode MIMIC-derived abdominal-pain benchmark, the full ranker achieves exploratory state-error AUROC 0.853, compared with 0.715 for maximum class probability and 0.552 for the backbone's native stop score. On the previously viewed 367-episode evaluation split, analytically averaging over the frozen Cros weights yields 16.9% selective error at 78.8% coverage, cost 5.57, and 0.68 tests, versus 30.8% error at 100% coverage, cost 8.14, and 1.53 tests under native stopping. Forced continuation is non-monotone: error is 28.3% with HPI alone and 34.3% after full workup. However, the uniform-weight mixture ablation is cheaper on this viewed split despite missing the locked development margins, and Cros nominally satisfies the joint criterion in only 6 of 20 development resplits. Because evaluation labels were inspected during earlier development, these findings provide exploratory feasibility and audit evidence, not a confirmatory safety certificate.
  •  
  • ↗️
↗️

Which Medical Questions Deserve Rationales? Perturbation-Sensitive Selection for Robust QA

βœ‡cs.AI, q-bio.NC updates on arXiv.org
By: Yuexin Wu Β· Dayou Yu Β· Vasile Rus
arXiv:2609.09684v1 Announce Type: cross Abstract: Medical question-answering datasets often contain answer labels, whereas high-quality rationales remain scarce, noisy, or costly to validate. This changes the acquisition question: rather than asking which questions should be labeled, we ask which already-labeled questions should receive rationale supervision under a fixed token budget. We study an offline version of this problem in which candidate rationales are visible to the selector but withheld from downstream training unless selected. We propose root-mean-square Robustness-based Sample Prioritization (RMS-RSP), which perturbs hidden states only at rationale tokens and measures the resulting shift in the gold-versus-best-distractor margin. Across five medical QA datasets, MedGemma-4B-IT, three training seeds, ten budgeted non-RSP selectors, and an unbudgeted full-supervision reference, RMS-RSP provides a deliberately qualified result. Its locked-budget accuracy is 60.61% on average versus 60.08% for Random, with a statistically resolved gain only on AfriMed-QA (+1.44 points). Its full-budget accuracy area is not better than Random. However, after three answer-option reorderings, RMS-RSP improves robust accuracy and semantic consistency by 1.91 and 2.85 points on average, respectively, with the same direction on all five datasets. Training on every pool rationale raises macro accuracy to 63.74%, but consumes 29--254 times more rationale tokens and does not uniformly improve robustness. These findings do not establish universal accuracy gains; they instead suggest that rationale-local boundary sensitivity can identify supervision that improves invariance to semantically equivalent formatting changes.
  •  
  • ↗️
↗️

Measuring Competency, Not Performance: Item-Aware Evaluation Across Medical Benchmarks

βœ‡cs.AI, q-bio.NC updates on arXiv.org
By: Zhimeng Luo Β· Lixin Wu Β· Adam Frisch Β· Daqing He
arXiv:2509.24186v2 Announce Type: replace-cross Abstract: Accuracy-based evaluation of Large Language Models (LLMs) measures benchmark-specific performance rather than underlying medical competency: it treats all questions as equally informative, conflates model ability with item characteristics, and thereby produces rankings that vary with benchmark choice. To address this, we introduce MedIRT, a psychometric evaluation framework grounded in Item Response Theory (IRT) that (1) jointly models latent competency and item-level difficulty and discrimination, and (2) includes benchmark integrity validation to ensure items within each topic measure a single, coherent underlying ability. We prospectively evaluate 71 diverse LLMs on a USMLE-aligned benchmark across 11 medical topics. As internal validation, MedIRT correctly predicts held-out LLM responses on unseen questions with 83.3% accuracy. As external validation, IRT-based rankings outperform accuracy-based rankings across 6 independent external medical benchmarks -- including expert preferences, holistic clinical tasks, safety judgments, and open-ended queries -- achieving 4 wins, 0 losses, and 18% lower variance. As a substantive finding, topic-level competency profiles expose striking domain-specific heterogeneity that aggregate accuracy masks. As a diagnostic tool, difficulty-tier analysis reveals two distinct response profiles (difficulty-sensitive responding and difficulty-insensitive responding) that require fundamentally different interventions. These results establish item-aware psychometric evaluation as a more valid and stable foundation for assessing LLMs in medicine, with potential implications for any high-stakes domain where benchmark integrity can be validated, and items vary meaningfully in difficulty and discrimination.
  •  
  • ↗️
↗️

ST-BiBench: Benchmarking Multi-Stream Multimodal Coordination in Bimanual Embodied Tasks for MLLMs

βœ‡cs.AI, q-bio.NC updates on arXiv.org
By: Xin Wu Β· Zhixuan Liang Β· Yue Ma Β· Mengkang Hu Β· Zhiyuan Qin Β· Xiu Li
arXiv:2602.08392v2 Announce Type: replace-cross Abstract: Multimodal Large Language Models (MLLMs) have significantly advanced the landscape of embodied AI, yet transitioning to synchronized bimanual coordination introduces formidable challenges in multi-stream multimodal integration. We introduce ST-BiBench, a comprehensive multi-tier framework for evaluating spatio-temporal multimodal coordination. Our approach centers on Strategic Coordination Planning, assessing high-level cross-modal reasoning over multiple action and perception streams. To investigate the "proximity paradox"-where semantically coherent plans fail to align with spatially grounded visual inputs-we incorporate Foundational Spatial Grounding to verify workspace awareness and arm-selection logic. Furthermore, we probe model frontiers through Fine-Grained Action Control, investigating whether MLLMs can directly synthesize high-dimensional continuous action modalities (16-Dim) from complex multimodal metadata. Evaluating 30+ state-of-the-art MLLMs, we uncover a persistent and pervasive "coordination paradox"-a significant gap between high-level strategic reasoning and fine-grained physical execution. Results reveal that while frontier MLLMs excel at logic-driven strategy, they frequently suffer from perception-logic disconnection and multi-stream interference during multimodal fusion. ST-BiBench provides a platform for identifying critical bottlenecks in multi-stream multimodal fusion and cross-modal alignment for complex embodied tasks.
  •  
  • ↗️
↗️

GPX3 suppresses gallbladder cancer progression by modulating redox balance, glycolysis, and anti-tumor immunity

βœ‡(Multiomics OR Omics) AND (Lung OR gastric OR Hepatocellular)
By: Zuyi Ma Β· Jia Sun Β· Xin Wu Β· Changzhen Shang Β· Binglu Li

Oncogenesis. 2026 Apr 2. doi: 10.1038/s41389-026-00603-7. Online ahead of print.

ABSTRACT

Gallbladder cancer (GBC) is an aggressive malignancy characterized by metabolic plasticity and profound immune evasion. However, the functional role of glutathione peroxidase 3 (GPX3), a secreted antioxidant enzyme, in these processes remains unclear. Multi-omics analyses of paired GBC and adjacent non-tumor tissues revealed consistent downregulation of GPX3, which correlated with reactive oxygen species (ROS) accumulation and enhanced glycolytic activity. Functional restoration of GPX3 in GBC cells reduced intracellular ROS levels, suppressed the expression of glycolysis-related enzymes, and consequently impaired tumor proliferation, migration, and invasion. In xenograft models, GPX3 overexpression markedly attenuated tumor growth and lung metastasis. Notably, GPX3 restoration also enhanced CD8+ T cell infiltration and elevated pro-inflammatory cytokine production, suggesting reversal of tumor-associated immunosuppression. These findings identify GPX3 as a critical tumor suppressor that integrates redox regulation, metabolic reprogramming, and immune activation to restrict malignant progression. Targeting GPX3 or its downstream pathways may represent a promising therapeutic strategy to simultaneously suppress gallbladder cancer aggressiveness and reinforce anti-tumor immunity.

PMID:41927557 | DOI:10.1038/s41389-026-00603-7

  •  
  • ↗️
↗️

GPX3 suppresses gallbladder cancer progression by modulating redox balance, glycolysis, and anti-tumor immunity

βœ‡Omics In Lung
By: Zuyi Ma Β· Jia Sun Β· Xin Wu Β· Changzhen Shang Β· Binglu Li

Oncogenesis. 2026 Apr 2. doi: 10.1038/s41389-026-00603-7. Online ahead of print.

ABSTRACT

Gallbladder cancer (GBC) is an aggressive malignancy characterized by metabolic plasticity and profound immune evasion. However, the functional role of glutathione peroxidase 3 (GPX3), a secreted antioxidant enzyme, in these processes remains unclear. Multi-omics analyses of paired GBC and adjacent non-tumor tissues revealed consistent downregulation of GPX3, which correlated with reactive oxygen species (ROS) accumulation and enhanced glycolytic activity. Functional restoration of GPX3 in GBC cells reduced intracellular ROS levels, suppressed the expression of glycolysis-related enzymes, and consequently impaired tumor proliferation, migration, and invasion. In xenograft models, GPX3 overexpression markedly attenuated tumor growth and lung metastasis. Notably, GPX3 restoration also enhanced CD8+ T cell infiltration and elevated pro-inflammatory cytokine production, suggesting reversal of tumor-associated immunosuppression. These findings identify GPX3 as a critical tumor suppressor that integrates redox regulation, metabolic reprogramming, and immune activation to restrict malignant progression. Targeting GPX3 or its downstream pathways may represent a promising therapeutic strategy to simultaneously suppress gallbladder cancer aggressiveness and reinforce anti-tumor immunity.

PMID:41927557 | DOI:10.1038/s41389-026-00603-7

  •  
  • ↗️
↗️

KEditVis: A Visual Analytics System for Knowledge Editing of Large Language Models

βœ‡cs.AI, q-bio.NC updates on arXiv.org
By: Zhenning Chen Β· Hanbei Zhan Β· Yanwei Huang Β· Xin Wu Β· Dazhen Deng Β· Di Weng Β· Yingcai Wu
arXiv:2603.29689v1 Announce Type: cross Abstract: Large Language Models (LLMs) demonstrate exceptional capabilities in factual question answering, yet they sometimes provide incorrect responses. To address this issue, knowledge editing techniques have emerged as effective methods for correcting factual information in LLMs. However, typical knowledge editing workflows struggle with identifying the optimal set of model layers for editing and rely on summary indicators that provide insufficient guidance. This lack of transparency hinders effective comparison and identification of optimal editing strategies. In this paper, we present KEditVis, a novel visual analytics system designed to assist users in gaining a deeper understanding of knowledge editing through interactive visualizations, improving editing outcomes, and discovering valuable insights for the future development of knowledge editing algorithms. With KEditVis, users can select appropriate layers as the editing target, explore the reasons behind ineffective edits, and perform more targeted and effective edits. Our evaluation, including usage scenarios, expert interviews, and a user study, validates the effectiveness and usability of the system.
  •  
  • ↗️
↗️

Generalizable Heuristic Generation Through LLMs with Meta-Optimization

βœ‡cs.AI, q-bio.NC updates on arXiv.org
By: Yiding Shi Β· Jianan Zhou Β· Wen Song Β· Jieyi Bi Β· Yaoxin Wu Β· Zhiguang Cao Β· Jie Zhang
arXiv:2505.20881v2 Announce Type: replace-cross Abstract: Heuristic design with large language models (LLMs) has emerged as a promising approach for tackling combinatorial optimization problems (COPs). However, existing approaches often rely on manually predefined evolutionary computation (EC) heuristic-optimizers and single-task training schemes, which may constrain the exploration of diverse heuristic algorithms and hinder the generalization of the resulting heuristics. To address these issues, we propose Meta-Optimization of Heuristics (MoH), a novel framework that operates at the optimizer level, discovering effective heuristic-optimizers through the principle of meta-learning. Specifically, MoH leverages LLMs to iteratively refine a meta-optimizer that autonomously constructs diverse heuristic-optimizers through (self-)invocation, thereby eliminating the reliance on a predefined EC heuristic-optimizer. These constructed heuristic-optimizers subsequently evolve heuristics for downstream tasks, enabling broader heuristic exploration. Moreover, MoH employs a multi-task training scheme to promote its generalization capability. Experiments on classic COPs demonstrate that MoH constructs an effective and interpretable meta-optimizer, achieving state-of-the-art performance across various downstream tasks, particularly in cross-size settings. Our code is available at: https://github.com/yiding-s/MoH.
  •  
  • ↗️
↗️

Automated Reinforcement Learning: An Overview

βœ‡cs.AI, q-bio.NC updates on arXiv.org
By: Reza Refaei Afshar Β· Joaquin Vanschoren Β· Uzay Kaymak Β· Rui Zhang Β· Yaoxin Wu Β· Wen Song Β· Yingqian Zhang
arXiv:2201.05000v2 Announce Type: replace-cross Abstract: Reinforcement Learning and, recently, Deep Reinforcement Learning are popular methods for solving sequential decision-making problems modeled as Markov Decision Processes. RL modeling of a problem and selecting algorithms and hyper-parameters require careful consideration, as different configurations may entail completely different performances. These considerations are mainly the task of RL experts; however, RL is progressively becoming popular in other fields, such as combinatorial optimization, where researchers and system designers are not necessarily RL experts. Besides, many modeling decisions are typically made manually, such as defining state and action space, size of batches, batch update frequency, and time steps. For these reasons, automating different components of RL is of great importance, and it has attracted much attention in recent years. Automated RL provides a framework in which different components of RL, including MDP modeling, algorithm selection, and hyper-parameter optimization, are modeled and defined automatically. In this article, we present the literature on automated RL (AutoRL), including the recent large language model (LLM) based techniques. We also discuss the recent work on techniques that are not presently tailored for automated RL but hold promise for future integration into AutoRL. Furthermore, we discuss the challenges, open questions, and research directions in AutoRL.
  •  
  • ↗️
↗️

Adaptive Multi-Agent Reasoning for Text-to-Video Retrieval

βœ‡cs.AI, q-bio.NC updates on arXiv.org
By: Jiaxin Wu Β· Xiao-Yong Wei Β· Qing Li
arXiv:2602.19040v1 Announce Type: cross Abstract: The rise of short-form video platforms and the emergence of multimodal large language models (MLLMs) have amplified the need for scalable, effective, zero-shot text-to-video retrieval systems. While recent advances in large-scale pretraining have improved zero-shot cross-modal alignment, existing methods still struggle with query-dependent temporal reasoning, limiting their effectiveness on complex queries involving temporal, logical, or causal relationships. To address these limitations, we propose an adaptive multi-agent retrieval framework that dynamically orchestrates specialized agents over multiple reasoning iterations based on the demands of each query. The framework includes: (1) a retrieval agent for scalable retrieval over large video corpora, (2) a reasoning agent for zero-shot contextual temporal reasoning, and (3) a query reformulation agent for refining ambiguous queries and recovering performance for those that degrade over iterations. These agents are dynamically coordinated by an orchestration agent, which leverages intermediate feedback and reasoning outcomes to guide execution. We also introduce a novel communication mechanism that incorporates retrieval-performance memory and historical reasoning traces to improve coordination and decision-making. Experiments on three TRECVid benchmarks spanning eight years show that our framework achieves a twofold improvement over CLIP4Clip and significantly outperforms state-of-the-art methods by a large margin.
  •  
  • ↗️
↗️

Exploring Partial Multi-Label Learning via Integrating Semantic Co-occurrence Knowledge

βœ‡cs.AI, q-bio.NC updates on arXiv.org
By: Xin Wu Β· Fei Teng Β· Yue Feng Β· Kaibo Shi Β· Zhuosheng Lin Β· Ji Zhang Β· James Wang
arXiv:2507.05992v2 Announce Type: replace-cross Abstract: Partial multi-label learning aims to extract knowledge from incompletely annotated data, which includes known correct labels, known incorrect labels, and unknown labels. The core challenge lies in accurately identifying the ambiguous relationships between labels and instances. In this paper, we emphasize that matching co-occurrence patterns between labels and instances is key to addressing this challenge. To this end, we propose Semantic Co-occurrence Insight Network (SCINet), a novel and effective framework for partial multi-label learning. Specifically, SCINet introduces a bi-dominant prompter module, which leverages an off-the-shelf multimodal model to capture text-image correlations and enhance semantic alignment. To reinforce instance-label interdependencies, we develop a cross-modality fusion module that jointly models inter-label correlations, inter-instance relationships, and co-occurrence patterns across instance-label assignments. Moreover, we propose an intrinsic semantic augmentation strategy that enhances the model's understanding of intrinsic data semantics by applying diverse image transformations, thereby fostering a synergistic relationship between label confidence and sample difficulty. Extensive experiments on four widely-used benchmark datasets demonstrate that SCINet surpasses state-of-the-art methods.
  •  
  • ↗️
There are no more articles
❌