LLM Agent Skill Learning

LLM: Large Language Model

Follow this topic Ask about this topic

Momentum

45 papers in the last four weeks, up 114% on the four weeks before. 0.4% of all new papers.

Jul 13Week of Sep 28

Latest papers 271

All topics

Practice Makes Unsafe: Skill Misevolution in Self-Improving LLM Agents
Aug 13, 2026Xutao Mao, Liangjie Zhao, Xiang Zheng +1LLM Agent Skill Learning AI Agent Safety Benchmarks
City University of Hong Kong · Adelaide University
ERSkill: Evolving for Skill-Guided Adaptive Memory Retrieval
Aug 13, 2026Haolong Chen, Liang Zhang, Zhuo Li +2LLM Agent Memory LLM Agent Skill Learning
Shenzhen International Center for Industrial and Applied Mathematics · Shenzhen Research Institute of Big Data · The Chinese University of Hong Kong, Shenzhen +2
Better, Faster, Stronger: Programmatic Skill Learning Best Reduces Agent Cost
Aug 11, 2026Zixi Huang, Xiheng Wang, Andrew Wang +4LLM Agent Self-Improvement LLM Agent Skill Learning
Johns Hopkins University
SkillZip: Evaluation-Free Skill Compression for Self-Evolving Agents by Discovering Reusable Structure
Aug 11, 2026Xiaofan Bai, Hongqiang Lin, Chao Liu +4LLM Agent Skill Learning Self-Evolving Agents
Alibaba Group · Zhejiang University · Duke University
SKILLER: Language-Level Reinforcement Learning for Reusable Skill Extraction in Small Language Models
Aug 11, 2026Chenhao Dang, Siyuan Xiong, Conghui He +1RL for Language Models Small Language Models
1Shanghai Jiao Tong University · 2Shanghai Artificial Intelligence Laboratory · 3Harbin Institute of Technology, Shenzhen +1
Do Personalized Skills Help Coding Agents? An Empirical Study of Developer Interaction Histories
Aug 10, 2026Shuyan Huang, Kai Du, Andrew LanLLM Personalization Coding Agents
UMass Amherst
Bidirectional Context Self-Distillation for Reinforcement Learning of Skill-Based LLM Agents
Aug 10, 2026Tianjun Pan, Yuan Li, Hongda Wang +8LLM Agent Skill Learning Self-Distillation
Fudan University · Zhejiang University · Alibaba Group +1
SkillSentry: Reliable Skill Execution for LLM Agents via Runtime Assurance
Aug 10, 2026You Lu, Xinyu Huang, Bihuan Chen +1LLM Agent Skill Learning LLM Agents
College of Computer Science and Artificial Intelligence, China
Emotion2Skill: Model-Internal Emotion Signals for Adaptive Skill Selection and Evolution
Aug 10, 2026Bohan Lin, Hejia Geng, Xinyi Xie +5LLM Agent Self-Improvement LLM Agent Skill Learning
University of Science and Technology of China (USTC) · 2Suzhou Institute for Advanced Research, USTC · University of Oxford +2
Branch2Skill: Efficient Skill Evolution Through Reasoning Trees
Aug 9, 2026Yanwei Ren, Haotian Zhang, Likang Xiao +5LLM Agent Skill Learning Agentic Reasoning
School of Artificial Intelligence, Beihang University, Beijing, China · Hangzhou International Innovation Institute, Beihang University, Hangzhou, China · The Hong Kong Polytechnic University, Hong Kong SAR, China +3
FailForge: Distilling Procedural Competence from Persistent Failures into Code Agents
Aug 9, 2026Dongyi Lv, Fushun E, Aichen Cai +8Fine-Tuning Software Engineering Agents
Xi’an Jiaotong University · Joy Future Academy, JD
SkillSmith: Enhancing Locally Deployed Agents via Automatic Skill Construction and Evolution
Aug 8, 2026Xinle Jiang, Remy Xie, Ming TangLLM Agent Self-Improvement LLM Agent Skill Learning
Southern University of Science and Technology, China · AOE Tech Labs Ltd., Hong Kong
Self-Evolving Neuro-Symbolic Skills for Tool-Augmented Spatial Reasoning
Aug 8, 2026Shi-Yu Tian, Zhuo-Xia Wang, Xuan-Yi Zhu +6Tool Use in VLMs LLM Agent Skill Learning
National Key Laboratory for Novel Software Technology, Nanjing University, China · School of Artificial Intelligence, Nanjing University, China · College of Computer Science, Sichuan University
Reason Wide, Not Deep: Amortizing the Reasoning Premium into Distilled Skills
Aug 8, 2026Agamdeep Singh, Srishti Gautam, Priyanshu Gupta +3Efficient Language Model Reasoning LLM Agent Skill Learning
Microsoft
SkillProx: Self-Evolving Agent Skills via Proximal Textual Gradient Descent
Aug 7, 2026Mingxuan Zheng, Yujin Zhou, Chuxue Cao +6LLM Agent Self-Improvement LLM Agent Skill Learning
Hong Kong University of Science and Technology · Macau University
BONSAI: Evolvability-Guided Tree Search over Skills
Aug 7, 2026Yash Priya Shastri, Anand Eswaran, Adnan Qidwai +2Monte Carlo Tree Search LLM Agent Skill Learning
IBM Research
SkillAligner: Treating Retrieved Skills as Adaptable Drafts at Execution Time
Aug 7, 2026Qinfeng Li, Dalin He, Yuntai Bao +7LLM Agent Skill Learning LLM Agent Skill Retrieval
OmniAI Group of ZJU ACES Lab · Zhejiang University
Learning Globally Reusable Skills for Coding Agents
Aug 6, 2026Chen Yang, Jiashuo Tian, Ziqi Wang +3Continual Learning for LLM Agents Software Engineering Agents
Tianjin University · Tsinghua University
When Self-Evolution Backfires: Pre-Commit Gating against Skill Contamination in LLM Agents
Aug 6, 2026Linfang Shang, Ming Xu, Yiding Sun +4AI Agent Reliability LLM Agent Skill Learning
Tencent
SkillHEX: Improving Agent Skills via Hypothesis-Driven Autonomous Exploration and Exploitation
Aug 6, 2026Yuru Feng, Yaoqi Chen, Beidi Zhao +7Test-Time Adaptation LLM Agent Self-Improvement
Microsoft · University of California, San Diego · University of Science and Technology of China +1
Search2Skill: Skill Distillation Beyond Knowledge Boundaries Via Rubric-Based Reinforcement Learning
Aug 5, 2026Muyang Ye, Tian Lan, Feihu Jiang +10LLM Agent Skill Learning LLM Agent Skill Retrieval
Zhejiang University · Alibaba Group · Institute of Artificial Intelligence, Xiamen University +1
ContinualSkillBench: Can LLM Agents Truly Evolve Their Capabilities?
Aug 4, 2026Tianyi Guan, Yiding Wang, Haotong Yang +5Continual Learning for LLM Agents LLM Agent Evaluation
Institute for Artificial Intelligence, Peking University · 2Beijing Institute for General Artificial Intelligence
SKT: Skill-Use Training at Scale via Verified Synthetic Data Generation
Aug 3, 2026Zelin Tan, Yiqun Zhang, Hao Li +11LLM Agent Skill Learning LLM Agent Trajectory Synthesis
Shanghai Artificial Intelligence Laboratory
Hear, Invoke, and Understand: A Skill-Calling Multimodal Agent for Large Audio Language Models
Aug 3, 2026Yuwen Wang, Tian-Hao Zhang, Minghao Cai +7Tool-Augmented Language Model Agents Agent Evaluation
University of Science and Technology Beijing, China · Li Auto Inc., Beijing, China
Progressive Agent Skill Generation via Reinforcement Learning
Aug 3, 2026Junhao Shen, Zhanqiu Zhang, Yiwen Guo +1LLM Agent Skill Learning RL for LLM Agents
The Chinese University of Hong Kong · LIGHTSPEED · Independent Researcher
TraceCompiler: Skill-Guided Mining and Compilation of LLM Agent Traces into Mostly Deterministic Workflows
Aug 3, 2026Salma El Yadouni, Guanyi LiAgentic Workflows LLM-Based Program Synthesis
EPFL · EPFL, Lausanne, Switzerland · Binome Technologies +1
Self-Supervised Skill Optimization
Jul 30, 2026Siran Peng, Cuiyu Yang, Tianyu Fu +9Large Language Model-Guided Optimization LLM Agent Self-Improvement
1MAIS, CASIA · 3Mininglamp Technology · 2SAI, UCAS +2
SKILL-KD: Contrastive Skill Distillation for LLM Agents
Jul 30, 2026Qiming Shi, Yibo Dou, Jiawen Zhu +5LLM Agent Skill Learning LLM Agent Training
State Key Lab of CAD&CG, Zhejiang University · School of Software and Microelectronics, Peking University · School of Software Technology, Zhejiang University +1
SKIMIX: Multi-Agent Harness-Time Scaling with Skill Mixture for Dynamic Harness Engineering
Jul 30, 2026Jia LuoAgent Harness Optimization Multi-Agent Collaboration
Huazhong University of Science and Technology
From Scoring to Acting: Outcome-Verified Comparative Self-Distillation for LLM Agents
Jul 30, 2026Xu Xia, Jinghua Piao, Min Yang +3LLM Agent Skill Learning Self-Distillation
Southeast University · Zhongguancun Academy · Tsinghua University +2

Aug 13, 2026, cs.AIJ/K move · Enter open · S save

Practice Makes Unsafe: Skill Misevolution in Self-Improving LLM Agents

Xutao Mao, Liangjie Zhao, Xiang Zheng, Cong Wang

City University of Hong Kong · Adelaide University

LLM Agent Skill Learning AI Agent Safety Benchmarks Self-Improving Agents LLM Agent Safety

Open paper AskPDF arXiv

LLM agents increasingly save experience from past tasks as reusable skills through skill evolution. When an agent completes an unsafe task, skill evolution can keep the unsafe procedure inside an otherwise useful skill, and a later benign task can repeat the harm after the unsafe instruction is gone. We call this failure skill misevolution. It requires that the agent write the unsafe procedure into a skill and that a later task retrieve and follow that skill. We introduce SkillMisevo-Gym, which clears every task session but keeps the skill library, so each of these steps can be checked separately. We also build SkillMisevo-Bench with 525 tasks in 25 episodes, where each episode alternates malicious and related benign tasks before three benign tasks in a new session. Across 21 combinations of agents and evolution methods, all of them write unsafe skills and 15 carry the harm into the new session, so the risk comes from skill evolution itself. Evolution raises benign utility in 15 combinations but also raises malicious-task harm in 17, so ranking methods by utility alone would hide this risk. We propose SafeEvolve, which checks each skill when it is written and when a later task retrieves it. It cuts unsafe retrieval in the new session from 35.3% to 8.7% and harm from 21.3% to 4.0% while keeping benign utility during learning within 0.4 points. SafeEvolveUp cuts new-session harm by 16.0 points and keeps its utility within 1.3 points of evolution without this mitigation. As agents learn from their own experience, safety must cover the skills that carry past behavior into future tasks. Code is available at https://github.com/henrymao2004/misevolve.

Momentum

Related topics

Latest papers 271