1 paper
Ruihan Yang, Fanghua Ye, Xiang We +10
Large language models (LLMs) are increasingly deployed as autonomous agents for multi-turn decision-making tasks. However, current agents typically rely on fixed cognitive patterns…