Structured Episodic Event Memory

2026-01-10

arXiv.org

AI Summary

This work addresses the limitations of current large language models, whose memory mechanisms predominantly rely on static retrieval-augmented generation and struggle to capture dynamic, structured event relationships in complex reasoning. To overcome this, the paper proposes a hierarchical memory framework that integrates graph-based memory with dynamic episodic memory. Drawing on cognitive frame theory, the approach transforms interaction streams into structured Episodic Event Frames (EEFs) and reconstructs coherent narrative contexts through agent-based association fusion and a Reverse Propagation Expansion (RPE) mechanism. Evaluated on the LoCoMo and LongMemEval benchmarks, the method significantly outperforms existing baselines, demonstrating superior capability in modeling long-term memory with enhanced narrative coherence and logical consistency.

Technology Category

• Natural Language Processing: (Large) Language Models
• Machine Learning: Large Multimodal Models (LMMs)
• Cognitive Modeling & Cognitive Systems: Conceptual Inference and Reasoning

Application Category

• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs
• Search and Retrieval-Augmented AI: Large language models for search
• Graph Algorithms and Modeling for the Web: Foundation models and LLMs for Web-related graphs

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.

Problem

• memory
• Large Language Models
• structured reasoning
• autonomous agents
• episodic memory

Innovation

• Structured Episodic Event Memory
• Episodic Event Frames
• Reverse Provenance Expansion
• graph memory
• cognitive frame theory