arXiv cs.AI· Siyuan Liu (Fudan University, Meituan Longcat Team), Fan Yu (Fudan University, Meituan Longcat Team), Dongyu Ru (Meituan Longcat Team), Yizhu Liu (Meituan Longcat Team), Yifan Yang (Meituan Longcat Team), Xuezhi Cao (Meituan Longcat Team), Xunliang Cai (Meituan Longcat Team), Yixin Cao (Fudan University)·· 3 小时前AI 评分30
DENSE：将智能体轨迹蒸馏为基于证据的捷径树以实现自我精炼
DENSE: Distilling Agent Trajectories into Evidence-Grounded Shortcut Trees for Self-Refinement
AI 导读
DENSE（Distilling Evidence from Nested Subtask Executions）将智能体执行轨迹组织为嵌套捷径树，在无外部结果监督下把冗余、失败的执行转化为可复用的任务求解反馈。
来源：arXiv cs.AI · arxiv.org