AI RESEARCH

Your Agent May Misevolve: Emergent Risks in Self-evolving LLM Agents

arXiv CS.LG

ArXi:2509.26354v2 Announce Type: replace-cross Advances in Large Language Models (LLMs) have enabled a new class of self-evolving agents that autonomously improve through interaction with the environment, nstrating strong capabilities. However, self-evolution also