作者位次 #9/12ACL 2026Findings· Findings of ACL
Reasoning as Gradient: Scaling MLE Agents Beyond Tree Search
Yifei Zhang · Xu Yang · Xiao Yang · Bowen Xian · Qizheng Li · Shikai Fang · Jingyuan Li · Tmall Group of Alibaba) · Minrui Xu · Yuge Zhang · Weiqing Liu · Jiang Bian
摘要
LLM-based agents for machine learning engineering (MLE) predominantly rely on tree search, a form of gradient-free optimization that uses scalar validation scores to rank candidates. As LLM reasoning capabilities improve, exhaustive enumeration becomes increasingly inefficient compared to directe...