作者位次 #4/9ACL 2026Poster· Natural Language Processing
DPEPO : Diverse Parallel Exploration Policy Optimization for LLM -based Agents
JunShuo Zhang · Chengrui Huang · Feng Guo · Zihan Li · Ke Shi · Menghua Jiang · Jiguo Yu · Shuo Shang · Shen Gao
摘要
Large language model (LLM) agents that follow the sequential “reason-then-act” paradigm have achieved superior performance in many complex tasks. However, these methods suffer from limited exploration and incomplete environmental understanding, as they interact with only a single environment per...