作者位次 #7/9ACL 2026Poster· Natural Language Processing
Reinforcement Learning for Self-Improving Agent with Skill Library
Jiongxiao Wang · Qiaojing Yan · Yawei Wang · Yijun Tian · Soumya Smruti Mishra · Zhichao Xu · Megha Gandhi · Panpan Xu · Lin Lee Cheong
摘要
Large Language Model (LLM)-based agents have demonstrated remarkable capabilities in complex reasoning and multi-turn interactions but struggle to continuously improve and adapt when deployed in new environments. One promising approach is implementing skill libraries that allow agents to learn, v...