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

TL;DR

E-mem introduces a multi-agent episodic memory framework for LLMs that enhances reasoning and reduces token costs by maintaining uncompressed, context-aware memories through hierarchical agent collaboration.

Contribution

The paper presents E-mem, a novel multi-agent episodic memory system inspired by biological engrams, improving reasoning and efficiency in LLM agents.

Findings

E-mem achieves over 54% F1 on LoCoMo benchmark.

E-mem surpasses state-of-the-art GAM by 7.75% F1.

E-mem reduces token cost by over 70%.

Abstract

The evolution of Large Language Model (LLM) agents towards System~2 reasoning, characterized by deliberative, high-precision problem-solving, requires maintaining rigorous logical integrity over extended horizons. However, prevalent memory preprocessing paradigms suffer from destructive de-contextualization. By compressing complex sequential dependencies into pre-defined structures (e.g., embeddings or graphs), these methods sever the contextual integrity essential for deep reasoning. To address this, we propose E-mem, a framework shifting from Memory Preprocessing to Episodic Context Reconstruction. Inspired by biological engrams, E-mem employs a heterogeneous hierarchical architecture where multiple assistant agents maintain uncompressed memory contexts, while a central master agent orchestrates global planning. Unlike passive retrieval, our mechanism empowers assistants to locally reason within activated segments, extracting context-aware evidence before aggregation. Evaluations on the LoCoMo benchmark demonstrate that E-mem achieves over 54\% F1, surpassing the state-of-the-art GAM by 7.75\%, while reducing token cost by over 70\%.