E-mem: Multi-agent based Episodic Context Reconstruction for LLM Agent Memory

Kaixiang Wang, Yidan Lin, Jiong Lou +3

The evolution of Large Language Model (LLM) agents towards System~2 reasoning, characterized by deliberative, high-precision problem-solving, requires maintaining rigorous logical…