❌

Reading view

↗️

Stabilizing Rubric Integration Training via Decoupled Advantage Normalization

✇cs.AI, q-bio.NC updates on arXiv.org
By: Zelin Tan · Zhouliang Yu · Bohan Lin · Zijie Geng · Hejia Geng · Yudong Zhang · Mulei Zhang · Yang Chen · Shuyue Hu · Zhenfei Yin · Chen Zhang · Lei Bai
arXiv:2603.26535v2 Announce Type: replace Abstract: We propose Process-Aware Policy Optimization (PAPO), a method that integrates process-level evaluation into Group Relative Policy Optimization (GRPO) through decoupled advantage normalization, to address two limitations of existing reward designs. Outcome reward models (ORM) evaluate only final-answer correctness, treating all correct responses identically regardless of reasoning quality, and gradually lose the advantage signal as groups become uniformly correct. Process reward models (PRM) offer richer supervision, but directly using PRM scores causes reward hacking, where models exploit verbosity to inflate scores while accuracy collapses. PAPO resolves both by composing the advantage from an outcome component Aout, derived from ORM and normalized over all responses, and a process component Aproc, derived from a rubric-based PRM and normalized exclusively among correct responses. This decoupled design ensures that Aout anchors training on correctness while Aproc differentiates reasoning quality without distorting the outcome signal. Experiments across multiple model scales and six benchmarks demonstrate that PAPO consistently outperforms ORM, reaching 51.3% vs.\ 46.3% on OlympiadBench while continuing to improve as ORM plateaus and declines.
  •  
  • ↗️
↗️

Rethinking the Role of Entropy in Optimizing Tool-Use Behaviors for Large Language Model Agents

✇cs.AI, q-bio.NC updates on arXiv.org
By: Zeping Li · Hongru Wang · Yiwen Zhao · Guanhua Chen · Yixia Li · Keyang Chen · Yixin Cao · Guangnan Ye · Hongfeng Chai · Zhenfei Yin
arXiv:2602.02050v3 Announce Type: replace Abstract: Tool-using agents based on Large Language Models (LLMs) excel in tasks such as mathematical reasoning and multi-hop question answering. However, in long trajectories, agents often trigger excessive and low-quality tool calls, increasing latency and degrading inference performance, making managing tool-use behavior challenging. In this work, we conduct entropy-based pilot experiments and observe a strong positive correlation between entropy reduction and high-quality tool calls. Building on this finding, we propose using entropy reduction as a supervisory signal and design two reward strategies to address the differing needs of optimizing tool-use behavior. Sparse outcome rewards provide coarse, trajectory-level guidance to improve efficiency, while dense process rewards offer fine-grained supervision to enhance performance. Experiments across diverse domains show that both reward designs improve tool-use behavior: the former reduces tool calls by 72.07% compared to the average of baselines, while the latter improves performance by 22.27%. These results position entropy reduction as a key mechanism for enhancing tool-use behavior, enabling agents to be more adaptive in real-world applications.
  •  
  • ↗️
↗️

Behavioral Consistency Validation for LLM Agents: An Analysis of Trading-Style Switching through Stock-Market Simulation

✇cs.AI, q-bio.NC updates on arXiv.org
By: Zeping Li · Guancheng Wan · Keyang Chen · Yu Chen · Yiwen Zhao · Philip Torr · Guangnan Ye · Zhenfei Yin · Hongfeng Chai
arXiv:2602.07023v2 Announce Type: replace-cross Abstract: Recent works have increasingly applied Large Language Models (LLMs) as agents in financial stock market simulations to test if micro-level behaviors aggregate into macro-level phenomena. However, a crucial question arises: Do LLM agents' behaviors align with real market participants? This alignment is key to the validity of simulation results. To explore this, we select a financial stock market scenario to test behavioral consistency. Investors are typically classified as fundamental or technical traders, but most simulations fix strategies at initialization, failing to reflect real-world trading dynamics. In this work, we assess whether agents' strategy switching aligns with financial theory, providing a framework for this evaluation. We operationalize four behavioral-finance drivers-loss aversion, herding, wealth differentiation, and price misalignment-as personality traits set via prompting and stored long-term. In year-long simulations, agents process daily price-volume data, trade under a designated style, and reassess their strategy every 10 trading days. We introduce four alignment metrics and use Mann-Whitney U tests to compare agents' style-switching behavior with financial theory. Our results show that recent LLMs' switching behavior is only partially consistent with behavioral-finance theories, highlighting the need for further refinement in aligning agent behavior with financial theory.
  •  
  • ↗️
↗️

Fluid-Derived Organoids from Pleural Effusion and Ascites: Emerging Models for Drug Resistance and Personalized Oncology

✇(Multiomics OR Omics) AND (Lung OR gastric OR Hepatocellular)
By: Yongyang Chen · Xiaoqing Xu · Jialin Chen · Miao Yin · Jinhui Chen · Zhanghua Qi · Ming Shi · Wenmei Su

J Cancer. 2026 Mar 4;17(3):614-625. doi: 10.7150/jca.127511. eCollection 2026.

ABSTRACT

Malignant pleural effusion (MPE) and malignant ascites (MA) are common complications in advanced-stage cancers, often signifying disease progression and resistance to treatment. Compared to tissue biopsies or surgical specimens, materials derived from effusions offer advantages such as minimal invasiveness, ease of accessibility, and the feasibility of repeated collection during therapeutic interventions. Organoids generated from tumor cells in effusions, termed fluid-derived organoids (FDOs), have demonstrated the ability to maintain genetic heterogeneity and accurately replicate patient-specific tumor phenotypes. These characteristics position FDOs as promising models for investigating drug resistance mechanisms and informing personalized oncology strategies. In the context of lung cancer, organoids derived from pleural effusions have been employed to study acquired resistance to epidermal growth factor receptor (EGFR) tyrosine kinase inhibitors and immunotherapy. Similarly, in ovarian and gastrointestinal cancers, organoids derived from ascites have proven to be valuable platforms for examining chemotherapy resistance and conducting drug sensitivity testing. FDOs have shown significant potential for translational applications by effectively correlating ex vivo drug responses with clinical outcomes, thus facilitating real-time monitoring of resistance evolution. However, several challenges remain, such as achieving culture standardization, maintaining the integrity of tumor microenvironment components, and integrating with multi-omics approaches. This review provides a comprehensive overview of recent advancements in the use of pleural effusion- and ascites-derived organoids for drug resistance research, underscores their applications in personalized oncology, and explores future research directions.

PMID:41869438 | PMC:PMC13003542 | DOI:10.7150/jca.127511

  •  
  • ↗️
↗️

Fluid-Derived Organoids from Pleural Effusion and Ascites: Emerging Models for Drug Resistance and Personalized Oncology

✇Omics In Lung
By: Yongyang Chen · Xiaoqing Xu · Jialin Chen · Miao Yin · Jinhui Chen · Zhanghua Qi · Ming Shi · Wenmei Su

J Cancer. 2026 Mar 4;17(3):614-625. doi: 10.7150/jca.127511. eCollection 2026.

ABSTRACT

Malignant pleural effusion (MPE) and malignant ascites (MA) are common complications in advanced-stage cancers, often signifying disease progression and resistance to treatment. Compared to tissue biopsies or surgical specimens, materials derived from effusions offer advantages such as minimal invasiveness, ease of accessibility, and the feasibility of repeated collection during therapeutic interventions. Organoids generated from tumor cells in effusions, termed fluid-derived organoids (FDOs), have demonstrated the ability to maintain genetic heterogeneity and accurately replicate patient-specific tumor phenotypes. These characteristics position FDOs as promising models for investigating drug resistance mechanisms and informing personalized oncology strategies. In the context of lung cancer, organoids derived from pleural effusions have been employed to study acquired resistance to epidermal growth factor receptor (EGFR) tyrosine kinase inhibitors and immunotherapy. Similarly, in ovarian and gastrointestinal cancers, organoids derived from ascites have proven to be valuable platforms for examining chemotherapy resistance and conducting drug sensitivity testing. FDOs have shown significant potential for translational applications by effectively correlating ex vivo drug responses with clinical outcomes, thus facilitating real-time monitoring of resistance evolution. However, several challenges remain, such as achieving culture standardization, maintaining the integrity of tumor microenvironment components, and integrating with multi-omics approaches. This review provides a comprehensive overview of recent advancements in the use of pleural effusion- and ascites-derived organoids for drug resistance research, underscores their applications in personalized oncology, and explores future research directions.

PMID:41869438 | PMC:PMC13003542 | DOI:10.7150/jca.127511

  •  
  • ↗️
↗️

Continual Learning in Large Language Models: Methods, Challenges, and Opportunities

✇cs.AI, q-bio.NC updates on arXiv.org
By: Hongyang Chen · Zhongwu Sun · Hongfei Ye · Kunchi Li · Xuemin Lin
arXiv:2603.12658v1 Announce Type: cross Abstract: Continual learning (CL) has emerged as a pivotal paradigm to enable large language models (LLMs) to dynamically adapt to evolving knowledge and sequential tasks while mitigating catastrophic forgetting-a critical limitation of the static pre-training paradigm inherent to modern LLMs. This survey presents a comprehensive overview of CL methodologies tailored for LLMs, structured around three core training stages: continual pre-training, continual fine-tuning, and continual alignment.Beyond the canonical taxonomy of rehearsal-, regularization-, and architecture-based methods, we further subdivide each category by its distinct forgetting mitigation mechanisms and conduct a rigorous comparative analysis of the adaptability and critical improvements of traditional CL methods for LLMs. In doing so, we explicitly highlight core distinctions between LLM CL and traditional machine learning, particularly with respect to scale, parameter efficiency, and emergent capabilities. Our analysis covers essential evaluation metrics, including forgetting rates and knowledge transfer efficiency, along with emerging benchmarks for assessing CL performance. This survey reveals that while current methods demonstrate promising results in specific domains, fundamental challenges persist in achieving seamless knowledge integration across diverse tasks and temporal scales. This systematic review contributes to the growing body of knowledge on LLM adaptation, providing researchers and practitioners with a structured framework for understanding current achievements and future opportunities in lifelong learning for language models.
  •  
  • ↗️
↗️

Towards Effective and Efficient Graph Alignment without Supervision

✇cs.AI, q-bio.NC updates on arXiv.org
By: Songyang Chen · Youfang Lin · Yu Liu · Shuai Zheng · Lei Zou
arXiv:2603.08526v1 Announce Type: cross Abstract: Unsupervised graph alignment aims to find the node correspondence across different graphs without any anchor node pairs. Despite the recent efforts utilizing deep learning-based techniques, such as the embedding and optimal transport (OT)-based approaches, we observe their limitations in terms of model accuracy-efficiency tradeoff. By focusing on the exploitation of local and global graph information, we formalize them as the ``local representation, global alignment'' paradigm, and present a new ``global representation and alignment'' paradigm to resolve the mismatch between the two phases in the alignment process. We then propose \underline{Gl}obal representation and \underline{o}ptimal transport-\underline{b}ased \underline{Align}ment (\texttt{GlobAlign}), and its variant, \texttt{GlobAlign-E}, for better \underline{E}fficiency. Our methods are equipped with the global attention mechanism and a hierarchical cross-graph transport cost, able to capture long-range and implicit node dependencies beyond the local graph structure. Furthermore, \texttt{GlobAlign-E} successfully closes the time complexity gap between representative embedding and OT-based methods, reducing OT's cubic complexity to quadratic terms. Through extensive experiments, our methods demonstrate superior performance, with up to a 20\% accuracy improvement over the best competitor. Meanwhile, \texttt{GlobAlign-E} achieves the best efficiency, with an order of magnitude speedup against existing OT-based methods.
  •  
  • ↗️
↗️

Process-Centric Analysis of Agentic Software Systems

✇cs.AI, q-bio.NC updates on arXiv.org
By: Shuyang Liu · Yang Chen · Rahul Krishna · Saurabh Sinha · Jatin Ganhotra · Reyhan Jabbarvand
arXiv:2512.02393v2 Announce Type: replace-cross Abstract: Agentic systems are modern software systems: they consist of orchestrated modules, expose interfaces, and are deployed in software pipelines. Unlike conventional programs, their execution, i.e., trajectories, is inherently stochastic and adaptive to the problems they solve. Evaluation of such systems is often outcome-centric. This narrow focus overlooks detailed insights, failing to explain how agents reason, plan, act, or change their strategies. Inspired by the structured representation of conventional software systems as graphs, we introduce Graphectory to systematically encode the temporal and semantic relations in such systems. Using Graphectory, we automatically analyze 4000 trajectories of two dominant agentic programming workflows, SWE-agent and OpenHands, with four backbone Large Language Models (LLMs), attempting to resolve SWE-bench issues. Our automated analyses (completed within four minutes) reveal that: (1) agents using richer prompts or stronger LLMs exhibit more complex Graphectory, reflecting deeper exploration, broader context gathering, and more thorough validation; (2) agents' strategies vary with problem difficulty and the underlying LLM - for resolved issues, strategies often follow coherent localization-patching-validation steps, while unresolved ones exhibit chaotic or backtracking behaviors; and (3) even successful agentic systems often display inefficient processes. We also implement a novel technique for real-time construction and analysis of Graphectory and Langutory during agent execution to flag trajectory issues. Upon detecting such issues, the technique notifies the agent with a diagnostic message and, when applicable, rolls back the trajectory. Experiments show that online monitoring and interventions improve resolution rates by 6.9%-23.5% across models for problematic instances, while significantly shortening trajectories with near-zero overhead.
  •  
  • ↗️
↗️

LifeBench: A Benchmark for Long-Horizon Multi-Source Memory

✇cs.AI, q-bio.NC updates on arXiv.org
By: Zihao Cheng · Weixin Wang · Yu Zhao · Ziyang Ren · Jiaxuan Chen · Ruiyang Xu · Shuai Huang · Yang Chen · Guowei Li · Mengshi Wang · Yi Xie · Ren Zhu · Zeren Jiang · Keda Lu · Yihong Li · Xiaoliang Wang · Liwei Liu · Cam-Tu Nguyen
arXiv:2603.03781v1 Announce Type: new Abstract: Long-term memory is fundamental for personalized agents capable of accumulating knowledge, reasoning over user experiences, and adapting across time. However, existing memory benchmarks primarily target declarative memory, specifically semantic and episodic types, where all information is explicitly presented in dialogues. In contrast, real-world actions are also governed by non-declarative memory, including habitual and procedural types, and need to be inferred from diverse digital traces. To bridge this gap, we introduce Lifebench, which features densely connected, long-horizon event simulation. It pushes AI agents beyond simple recall, requiring the integration of declarative and non-declarative memory reasoning across diverse and temporally extended contexts. Building such a benchmark presents two key challenges: ensuring data quality and scalability. We maintain data quality by employing real-world priors, including anonymized social surveys, map APIs, and holiday-integrated calendars, thus enforcing fidelity, diversity and behavioral rationality within the dataset. Towards scalability, we draw inspiration from cognitive science and structure events according to their partonomic hierarchy; enabling efficient parallel generation while maintaining global coherence. Performance results show that top-tier, state-of-the-art memory systems reach just 55.2\% accuracy, highlighting the inherent difficulty of long-horizon retrieval and multi-source integration within our proposed benchmark. The dataset and data synthesis code are available at https://github.com/1754955896/LifeBench.
  •  
  • ↗️
↗️

BEAT: Visual Backdoor Attacks on VLM-based Embodied Agents via Contrastive Trigger Learning

✇cs.AI, q-bio.NC updates on arXiv.org
By: Qiusi Zhan · Hyeonjeong Ha · Rui Yang · Sirui Xu · Hanyang Chen · Liang-Yan Gui · Yu-Xiong Wang · Huan Zhang · Heng Ji · Daniel Kang
arXiv:2510.27623v3 Announce Type: replace Abstract: Recent advances in Vision-Language Models (VLMs) have propelled embodied agents by enabling direct perception, reasoning, and planning task-oriented actions from visual inputs. However, such vision-driven embodied agents open a new attack surface: visual backdoor attacks, where the agent behaves normally until a visual trigger appears in the scene, then persistently executes an attacker-specified multi-step policy. We introduce BEAT, the first framework to inject such visual backdoors into VLM-based embodied agents using objects in the environments as triggers. Unlike textual triggers, object triggers exhibit wide variation across viewpoints and lighting, making them difficult to implant reliably. BEAT addresses this challenge by (1) constructing a training set that spans diverse scenes, tasks, and trigger placements to expose agents to trigger variability, and (2) introducing a two-stage training scheme that first applies supervised fine-tuning (SFT) and then our novel Contrastive Trigger Learning (CTL). CTL formulates trigger discrimination as preference learning between trigger-present and trigger-free inputs, explicitly sharpening the decision boundaries to ensure precise backdoor activation. Across various embodied agent benchmarks and VLMs, BEAT achieves attack success rates up to 80%, while maintaining strong benign task performance, and generalizes reliably to out-of-distribution trigger placements. Notably, compared to naive SFT, CTL boosts backdoor activation accuracy up to 39% under limited backdoor data. These findings expose a critical yet unexplored security risk in VLM-based embodied agents, underscoring the need for robust defenses before real-world deployment.
  •  
  • ↗️
↗️

Lean Finder: Semantic Search for Mathlib That Understands User Intents

✇cs.AI, q-bio.NC updates on arXiv.org
By: Jialin Lu · Kye Emond · Kaiyu Yang · Swarat Chaudhuri · Weiran Sun · Wuyang Chen
arXiv:2510.15940v2 Announce Type: replace-cross Abstract: We present Lean Finder, a semantic search engine for Lean and mathlib that understands and aligns with the intents of mathematicians. Progress in formal theorem proving is often hindered by the difficulty of locating relevant theorems and the steep learning curve of the Lean 4 language, making advancement slow and labor-intensive. Existing Lean search engines, though helpful, rely primarily on informalizations (natural language translation of the formal statements), while largely overlooking the mismatch with real-world user queries. In contrast, we propose a user-centered semantic search tailored to the needs of mathematicians. Our approach begins by analyzing and clustering the semantics of public Lean discussions, then fine-tuning text embeddings on synthesized queries that emulate user intents. We further align Lean Finder with mathematicians' preferences using diverse feedback signals, encoding it with a rich awareness of their goals from multiple perspectives. Evaluations on real-world queries, informalized statements, and proof states demonstrate that our Lean Finder achieves over $30\%$ relative improvement compared to previous search engines and GPT-4o. In addition, Lean Finder is compatible with LLM-based theorem provers, bridging retrieval with formal reasoning. Lean Finder is available at: https://leanfinder.github.io
  •  
  • ↗️
↗️

Mantis: A Versatile Vision-Language-Action Model with Disentangled Visual Foresight

✇cs.AI, q-bio.NC updates on arXiv.org
By: Yi Yang · Xueqi Li · Yiyang Chen · Jin Song · Yihan Wang · Zipeng Xiao · Jiadi Su · You Qiaoben · Pengfei Liu · Zhijie Deng
arXiv:2511.16175v2 Announce Type: replace-cross Abstract: Recent advances in Vision-Language-Action (VLA) models demonstrate that visual signals can effectively complement sparse action supervisions. However, letting VLA directly predict high-dimensional visual states can distribute model capacity and incur prohibitive training cost, while compressing visual states into more compact supervisory signals inevitably incurs information bottlenecks. Moreover, existing methods often suffer from poor comprehension and reasoning capabilities due to the neglect of language supervision. This paper introduces Mantis, a novel framework featuring a Disentangled Visual Foresight (DVF) to tackle these issues. Specifically, Mantis decouples visual foresight prediction from the backbone with the combination of meta queries and a diffusion Transformer (DiT) head. With the current visual state provided to the DiT via a residual connection, a simple next-state prediction objective enables the meta queries to automatically capture the latent actions that delineate the visual trajectory, and hence boost the learning of explicit actions. The disentanglement reduces the burden of the VLA backbone, enabling it to maintain comprehension and reasoning capabilities through language supervision. Empirically, pretrained on human manipulation videos, robot demonstrations, and image-text pairs, Mantis achieves a 96.7% success rate on LIBERO benchmark after fine-tuning, surpassing powerful baselines while exhibiting high convergence speed. Real-world evaluations show that Mantis outperforms $\pi_{0.5}$, a leading open-source VLA model, particularly in instruction-following capability, generalization to unseen instructions, and reasoning ability. Code and weights are released to support the open-source community.
  •  
  • ↗️
There are no more articles
❌