作者位次 #2/8ICML 2026Poster
Large Language Model Agents Are Not Always Faithful Self-Evolvers
Weixiang Zhao · Yingshuo Wang · Yichen Zhang · Yang Deng · Yanyan Zhao · Wanxiang Che · Bing Qin · Ting Liu
智能体与工具OpenReview
首次获取该候选人的完整档案及 StartLux 评估共 20 积分；每位用户单独解锁，之后查看免费。主动刷新 20 积分。
实时查找电子邮箱，费用不超过20积分；未找到邮箱不扣费。
Weixiang Zhao · Yingshuo Wang · Yichen Zhang · Yang Deng · Yanyan Zhao · Wanxiang Che · Bing Qin · Ting Liu
Weixiang Zhao · Yichen Zhang · Yingshuo Wang · Yang Deng · Yanyan Zhao · Xuda Zhi · Yongbo Huang · Hao He · Wanxiang Che · Bing Qin · Ting Liu
Experience-driven self-evolution has emerged as a promising paradigm for improving the autonomy of large language model agents, yet its reliance on self-curated experience introduces underexplored safety risks. In this study, we investigate how experience accumulation and utilization in self-evol...