Structured Episodic Event Memory

TL;DR

This paper introduces Structured Episodic Event Memory (SEEM), a hierarchical memory framework for LLMs that enhances reasoning and narrative coherence by integrating graph and episodic memories grounded in cognitive theory.

Contribution

SEEM is a novel hierarchical memory architecture combining graph and episodic memories, with mechanisms for reconstructing coherent narratives, inspired by cognitive frame theory.

Findings

SEEM outperforms baselines on LoCoMo and LongMemEval benchmarks.

SEEM improves narrative coherence and logical consistency in agents.

Experimental results validate the effectiveness of the hierarchical memory design.

Abstract

Current approaches to memory in Large Language Models (LLMs) predominantly rely on static Retrieval-Augmented Generation (RAG), which often results in scattered retrieval and fails to capture the structural dependencies required for complex reasoning. For autonomous agents, these passive and flat architectures lack the cognitive organization necessary to model the dynamic and associative nature of long-term interaction. To address this, we propose Structured Episodic Event Memory (SEEM), a hierarchical framework that synergizes a graph memory layer for relational facts with a dynamic episodic memory layer for narrative progression. Grounded in cognitive frame theory, SEEM transforms interaction streams into structured Episodic Event Frames (EEFs) anchored by precise provenance pointers. Furthermore, we introduce an agentic associative fusion and Reverse Provenance Expansion (RPE) mechanism to reconstruct coherent narrative contexts from fragmented evidence. Experimental results on the LoCoMo and LongMemEval benchmarks demonstrate that SEEM significantly outperforms baselines, enabling agents to maintain superior narrative coherence and logical consistency.