Image de couverture de livre

Harness Updating Is Not Harness Benefit: Disentangling Evolution Capabilities in Self-Evolving LLM Agents

Fuente: arXiv

Enregistré dans:

Détails bibliographiques
Auteurs principaux:	Lin, Minhua, Wu, Juncheng, Wang, Zijun, Shi, Zhan, Sang, Yisi, He, Bing, Liu, Zewen, Wei, Tianxin, Wu, Zongyu, Zhang, Zhiwei, Wang, Dakuo, Zhang, Xiang, Dumoulin, Benoit, Xie, Cihang, Zhou, Yuyin, Wang, Suhang, Lu, Hanqing
Format:	Preprint
Publié:	2026
Sujets:	Artificial Intelligence
Accès en ligne:	Acceder al recurso
Tags:	Ajouter un tag Pas de tags, Soyez le premier à ajouter un tag!

Documents similaires

Adaptive Auto-Harness: Sustained Self-Improvement for Agentic System Deployment on Open-Ended Task Streams
par: Liu, Zewen, et autres
Publié: (2026)

Position: Agentic Evolution is the Path to Evolving LLMs
par: Lin, Minhua, et autres
Publié: (2026)

Robustness Inspired Graph Backdoor Defense
par: Zhang, Zhiwei, et autres
Publié: (2024)

Image Corruption-Inspired Membership Inference Attacks against Large Vision-Language Models
par: Wu, Zongyu, et autres
Publié: (2025)

Are You Using Reliable Graph Prompts? Trojan Prompt Attacks on Graph Neural Networks
par: Lin, Minhua, et autres
Publié: (2024)

LLM and GNN are Complementary: Distilling LLM for Multimodal Graph Learning
par: Xu, Junjie, et autres
Publié: (2024)

ClinSeekAgent: Automating Multimodal Evidence Seeking for Agentic Clinical Reasoning
par: Wu, Juncheng, et autres
Publié: (2026)

Rethinking Graph Backdoor Attacks: A Distribution-Preserving Perspective
par: Zhang, Zhiwei, et autres
Publié: (2024)

MemMA: Coordinating the Memory Cycle through Multi-Agent Reasoning and In-Situ Self-Evolution
par: Lin, Minhua, et autres
Publié: (2026)

Bridging Source and Target Domains via Link Prediction for Unsupervised Domain Adaptation on Graphs
par: Wang, Yilong, et autres
Publié: (2025)

Universal Prompt Optimizer for Safe Text-to-Image Generation
par: Wu, Zongyu, et autres
Publié: (2024)

Decoding Time Series with LLMs: A Multi-Agent Framework for Cross-Domain Annotation
par: Lin, Minhua, et autres
Publié: (2024)

Firefly: Illuminating Large-Scale Verified Tool-Call Data Generation from Real APIs
par: Lu, Yuxuan, et autres
Publié: (2026)

PreGIP: Watermarking the Pretraining of Graph Neural Networks for Deep Intellectual Property Protection
par: Dai, Enyan, et autres
Publié: (2024)

From Seeing to Thinking: Decoupling Perception and Reasoning Improves Post-Training of Vision-Language Models
par: Wu, Juncheng, et autres
Publié: (2026)

RewardHarness: Self-Evolving Agentic Post-Training
par: Zhang, Yuxuan, et autres
Publié: (2026)

GPR: Empowering Generation with Graph-Pretrained Retriever
par: Wang, Xiaochen, et autres
Publié: (2025)

VDebugger: Harnessing Execution Feedback for Debugging Visual Programs
par: Wu, Xueqing, et autres
Publié: (2024)

AutoMedEval: Harnessing Language Models for Automatic Medical Capability Evaluation
par: Zhang, Xiechi, et autres
Publié: (2025)

Harnessing EHRs for Diffusion-based Anomaly Detection on Chest X-rays
par: Kim, Harim, et autres
Publié: (2025)

MiLoRA: Harnessing Minor Singular Components for Parameter-Efficient LLM Finetuning
par: Wang, Hanqing, et autres
Publié: (2024)

A Comprehensive Survey on Reinforcement Learning-based Agentic Search: Foundations, Roles, Optimizations, Evaluations, and Applications
par: Lin, Minhua, et autres
Publié: (2025)

Catastrophic Failure of LLM Unlearning via Quantization
par: Zhang, Zhiwei, et autres
Publié: (2024)

Chasing the Public Score: User Pressure and Evaluation Exploitation in Coding Agent Workflows
par: Chen, Hardy, et autres
Publié: (2026)

LanP: Rethinking the Impact of Language Priors in Large Vision-Language Models
par: Wu, Zongyu, et autres
Publié: (2025)

Harnessing Photonics for Machine Intelligence
par: Zhu, Hanqing, et autres
Publié: (2026)

Draw2Think: Harnessing Geometry Reasoning through Constraint Engine Interaction
par: Hu, Juncheng, et autres
Publié: (2026)

Knowledge or Reasoning? A Close Look at How LLMs Think Across Domains
par: Wu, Juncheng, et autres
Publié: (2025)

Harnessing Agentic Evolution
par: Zhang, Jiayi, et autres
Publié: (2026)

Divide-Verify-Refine: Can LLMs Self-Align with Complex Instructions?
par: Zhang, Xianren, et autres
Publié: (2024)

Counterfactual Learning on Graphs: A Survey
par: Guo, Zhimeng, et autres
Publié: (2023)

DDR: Exploiting Deep Degradation Response as Flexible Image Descriptor
par: Wu, Juncheng, et autres
Publié: (2024)

Harnessing Catalytic RNA Circuits for Construction of Artificial Signaling Pathways in Mammalian Cells
par: Chao‐Qun Wu, et autres
Publié: (2024)

STAR-1: Safer Alignment of Reasoning LLMs with 1K Data
par: Wang, Zijun, et autres
Publié: (2025)

Harnessing from Nature – Evolving Potential of Antimicrobial Peptide
par: Songhan Liu, et autres
Publié: (2025)

Harnessing Water Molecules for Strong Underwater Adhesion
par: Biaolong Ma, et autres
Publié: (2026)

Unlearning Inversion Attacks for Graph Neural Networks
par: Zhang, Jiahao, et autres
Publié: (2025)

Sibyl-AutoResearch: Autonomous Research Needs Self-Evolving Trial-and-Error Harnesses, Not Paper Generators
par: Wang, Chengcheng, et autres
Publié: (2026)

Code as Agent Harness
par: Ning, Xuying, et autres
Publié: (2026)

Your Agent, Their Asset: A Real-World Safety Analysis of OpenClaw
par: Wang, Zijun, et autres
Publié: (2026)