Self-Evolving Agents

Follow this topic Ask about this topic

Self-evolving agents are systems that use experience from task execution to change how they solve future tasks, rather than relying only on a fixed model or hand-designed workflow. Common approaches store and refine reusable skills or memories, optimize prompts and agent harnesses, train policies with reinforcement learning, or co-evolve task generators, environments, and teams of agents. These methods are tested in settings such as web and GUI use, coding, embodied robotics, scientific workflows, and reasoning benchmarks, with feedback ranging from execution outcomes to verifiers and paired validation. A central challenge is making improvements transfer to new tasks without forgetting earlier abilities or introducing unsafe behavior; reliable evaluation, update verification, and long-term monitoring remain open problems.

Latest papers

427 papers104 papers in the last four weeks, up 148% on the four weeks before. 0.6% of all new papers.

Governed Individuation: Cryptographically Decoupling an Agent's Learning from Its Authority
Jul 6, 2026Xue Qin, Simin Luan, Cong Yang +1AI Agent Security AI Agent Safety
School of Software, Harbin Institute of Technology, Harbin, China · School of Computer Science and Technology, Harbin Institute of Technology, Harbin, China · School of Future Science and Engineering, Soochow University, Suzhou, China
Towards Self-Evolving Agents: A Human-Inspired Adaptive Exploration-Exploitation Framework for Genetic Network Programming
Jul 4, 2026Ali Kohan, Mohamad Roshanzamir, Roohallah Alizadehsani +1Exploration-Exploitation Tradeoff Genetic Programming
Department of Computer Engineering, Faculty of Engineering, Fasa University, Fasa 74617-81189, Iran · Institute for Intelligent Systems Research and Innovation (IISRI), Deakin University, Geelong, Australia · Center of Excellence in Precision Medicine and Digital Health, Department of Physiology, Faculty of Dentistry, Chulalongkorn University, Bangkok 10330, Thailand. +2
Harness-Aware Self-Evolving: Co-Evolving Model Weights, Harness, and Task Solutions
Jul 4, 2026Haochen Luo, Yi Huang, Sichun Luo +5Agentic RL Agent Harness Optimization
School of Computing and Data Science, The University of Hong Kong · Grace Investment Machine · Jiutian Research, China Mobile
EvoPolicyGym: Evaluating Autonomous Policy Evolution in Interactive Environments
Jul 2, 2026Zhilin Wang, Han Song, Runzhe Zhan +13RL Benchmarks AI Agent Benchmarks
University of Science and Technology of China · The Chinese University of Hong Kong · University of Macau +5
Self-Evolving Agents with Anytime-Valid Certificates
Jul 1, 2026Biswa SenguptaSoftware Engineering Agents LLM Agent Self-Improvement
LLM Suite Team, JPMorgan Chase & Co.
A Self-Evolving Agentic System for Automated Generation and Execution of Biological Protocols
Jun 30, 2026Yankai Jiang, Weiting Tang, Haoran Sun +11Scientific Code Generation Laboratory Automation
1Shanghai Artificial Intelligence Laboratory, Shanghai, China · 2Genoria AI, Shenzhen, China
From Detecting Agency to Doing Work: Self-Caused Credit Builds a Durable Behavioral Self in a Minimal Spiking Agent
Jun 29, 2026Haoliang HanCredit Assignment in RL Credit Assignment
Institute of Biomedical Strategy, China Pharmaceutical University
SAGA: Scene-Aware, Goal-Evolving Agents for Long-Horizon Strategy Game Planning
Jun 29, 2026Tianyu Jin, Shuo Chen, Yida Wang +6Multi-Agent LLM Systems Game-Playing Agents
Beijing University of Posts and Telecommunications · Beijing Institute for General Artificial Intelligence (BIGAI)
AI Trading's Alpha Singularity: Emergent Market Reasoning through Agent-to-Agent Self-Evolution
Jun 28, 2026Yuqi Li, Siyuan Liu, Bingjun LiuFinance Emergent Behavior in Multi-Agent Systems
Panda AI
Agentic Hardware Design as Repository-Level Code Evolution
Jun 26, 2026Cunxi Yu, Chenhui Deng, Nathaniel Pinckney +1Software Engineering Electronic Design Automation
NVIDIA Research
Escaping the Self-Confirmation Trap: An Execute-Distill-Verify Paradigm for Agentic Experience Learning
Jun 23, 2026Shiding Zhu, Yudi Qi, Yajie Wang +6LLM Agents Self-Evolving Agents
Zhejiang University · Shenzhen International Graduate School, Tsinghua University · Independent Researcher +3
Metis: Bridging Text and Code Memory for Self-Evolving Agents
Jun 23, 2026Zijie Dai, Siuhin He, Hui Li +9Agent Memory LLM Agent Memory
The Chinese University of Hong Kong · Huawei · Wuhan University
Safety in Self-Evolving LLM Agent Systems: Threats, Amplification, and Case Studies
Jun 22, 2026Ruixiao Lin, Xinhao Deng, Qingming Li +12LLM Agent Security Self-Evolving Agents
1Zhejiang University · 2Ant Group · 3Tsinghua University +3
Self-Evolving Cognitive Framework via Causal World Modeling for Embodied Scientific Intelligence
Jun 21, 2026Yi Yu, Tetsunari InamuraCausal World Models Embodied AI
Graduate School of Advanced Science and Engineering, Hiroshima University, Japan · Advanced Intelligence and Robotics Research Center, Brain Science Institute, Tamagawa University, Japan
Recursive Self-Evolving Agents via Held-Out Selection
Jun 17, 2026Michael Nguyen, Quoc Nguyen, Paul VuongLLM Agent Evaluation LLM Agent Workflow Optimization
School of Information Technology Monash University Malaysia
Skill-MAS: Evolving Meta-Skill for Automatic Multi-Agent Systems
Jun 17, 2026Hehai Lin, Qi Yang, Chengwei QinLLM Agent Skill Learning Multi-Agent Systems
The Hong Kong University of Science and Technology (Guangzhou) · Ant Group
EvolveNav: Proactive Preflection and Self-Evolving Memory for Zero-Shot Object Goal Navigation
Jun 16, 2026Qi Chai, Wenhao Shen, Nanjie Yao +5Object-Goal Navigation Memory-Augmented Agents
XGRIDS Shenzhen, China
OPD-Evolver: Cultivating Holistic Agent Evolver via On-Policy Distillation
Jun 16, 2026Guibin Zhang, Xun Xu, Yanwei Yue +4LLM Agent Memory Memory-Augmented Agents
LV-NUS Lab · FDU · PKU +1
SEAGym: An Evaluation Environment for Self-Evolving LLM Agents
Jun 16, 2026Congjie Zheng, Chuanyi Xue, Bin Liang +2LLM Agent Evaluation AI Agent Benchmarks
Department of Automation, Tsinghua University, Beijing, 100084, China · Beijing National Research Center for Information Science and Technology (BNRist), Tsinghua University, Beijing, 100084, China
APEX: Adaptive Principle EXtraction A Three-Layer Self-Evolution Framework for Production AI Agents
Jun 13, 2026Ya-Chuan Chen, Tien-Jen Lai, Hsiang-Wei HuAgentic Workflows LLM Agent Self-Improvement
Grace AI Technology
HarnessX: A Composable, Adaptive, and Evolvable Agent Harness Foundry
Jun 12, 2026Tingyang Chen, Shuo Lu, Kang Zhao +11LLM Agent Orchestration Agent Harness Optimization
Darwin Agent Team
ProPlay: Procedural World Models for Self-Evolving LLM Agents
Jun 11, 2026Yijun Ma, Zehong Wang, Yiyang Li +5World Models LLM Agent Planning
University of Notre Dame · University of Connecticut
Hybrid Open-Ended Tri-Evolution Makes Better Deep Researcher
Jun 10, 2026Hongming Piao, Chi Liu, Mengzhuo Chen +5Deep Research Agents LLM Agent Training
IQuest Research · Zhejiang University
Role-Agent: Bootstrapping LLM Agents via Dual-Role Evolution
Jun 9, 2026Xucong Wang, Ziyu Ma, Shidong Yang +4LLM Agent Self-Improvement LLM Agent Training
University of Science and Technology of China · AMAP, Alibaba Group
Beyond Static Evaluation: Co-Evolutionary Mechanisms for LLM-Driven Strategy Evolution in Adversarial Games
Jun 9, 2026Haoran Li, Zengle Ge, Ziyang Zhang +10Game-Playing Agents Evolutionary Optimization
University of Chinese Academy of Sciences · Famou Agent Team, Baidu AI Cloud · The University of Sydney, Australia +3
From 0-to-1 to 1-to-N: Reproducible Engineering Evidence for MetaAI Recursive Self-Design
Jun 8, 2026Dun Li, Jiatao Li, Hongzhi LiLLM Agent Self-Improvement Self-Improving Agents
Department of Computing, The Hong Kong Polytechnic University, Hong Kong SAR, China · Merchant Marine College, Shanghai Maritime University, Shanghai 201306, China · College of Big Data and Artificial Intelligence, Chizhou University, Chizhou 247100, China
Experience Makes Skillful: Enabling Generalizable Medical Agent Reasoning via Self-Evolving Skill Memory
Jun 8, 2026Haoran Sun, Wenjie Li, Yujie Zhang +8Continual Learning for LLM Agents LLM Agent Memory
Fudan University · Shanghai Artificial Intelligence Laboratory · University of Science and Technology of China +1
Self-Evolving Scientific Agent Designs Physically Reasoned White-Box Fluid Control
Jun 7, 2026Boai Sun, Wenjin Guo, Zongmin Yu +1Robotic Control Continuous Control
National University of Singapore, Singapore.
PACE: Anytime-Valid Acceptance Tests for Self-Evolving Agents
Jun 6, 2026Zayx ShawnSequential Hypothesis Testing LLM Agent Reliability
Independent Researcher
Socratic-SWE: Self-Evolving Coding Agents via Trace-Derived Agent Skills
Jun 5, 2026Chuan Xiao, Zhengbo Jiao, Shaobo Wang +5Software Engineering Agents LLM Agent Skill Learning
2Shanghai Jiao Tong University · 1AI Data, Alibaba Group

Jul 6, 2026J/K move · Enter open · S save

Governed Individuation: Cryptographically Decoupling an Agent's Learning from Its Authority

Xue Qin, Simin Luan, Cong Yang, Zhijun Li

School of Software, Harbin Institute of Technology, Harbin, China · School of Computer Science and Technology, Harbin Institute of Technology, Harbin, China · School of Future Science and Engineering, Soochow University, Suzhou, China

AI Agent Security AI Agent Safety Runtime Enforcement for AI Agents Capability-Based Security Access Control Self-Evolving Agents

Open paper AskPDF arXiv

Autonomous agents are moving from sandboxed text generators to operators of code, data, and physical infrastructure, and they increasingly learn while deployed. This reopens a question that alignment techniques answer only probabilistically: after an agent has adapted in the field, is the running system still confined to what its operator authorised? Here we show that confinement can be guaranteed as an invariant of the agent's execution architecture rather than a probabilistic outcome of its training. Governed individuation binds an agent at boot to a cryptographically frozen identity digest, and routes every action through a gate defined over the semantic effect of the action rather than its name. We prove that no amount of learning, skill acquisition, or self-induced governance abstraction can widen the agent's permitted authority without an operator-signed change to its identity; the guarantee holds even when the agent induces its own safety principle and that principle is wrong. Empirically, in an open-ended tool-use benchmark where a large action space rules out name-based blocking, ungoverned software agents under reward pressure attempt to tamper with their own evaluation at a task-dependent rate that reaches every run on the hardest task, whereas the gate reduces executed forbidden effects to zero as a verified property of the construction while preserving task success. An adversarial evaluation of monitors of increasing semantic depth shows false-allows falling from 75% (name-based gating) to zero (dynamic effect tracing), and refusal history transfers compliance to held-out red-line families. Trust in a deployed learning agent shifts from a wager on its continued alignment to a check anyone can run at boot.