GAM: Hierarchical Graph-based Agentic Memory for LLM Agents

TL;DR

GAM introduces a hierarchical graph-based memory system for LLM agents that balances rapid context updates with long-term knowledge retention, improving reasoning and efficiency.

Contribution

The paper proposes a novel hierarchical graph-based memory architecture that decouples encoding from consolidation to better manage evolving narratives in LLM agents.

Findings

Outperforms state-of-the-art baselines in reasoning accuracy.

Enhances context precision with graph-guided retrieval.

Demonstrates improved efficiency on LoCoMo and LongDialQA.

Abstract

To sustain coherent long-term interactions, Large Language Model (LLM) agents must navigate the tension between acquiring new information and retaining prior knowledge. Current unified stream-based memory systems facilitate context updates but remain vulnerable to interference from transient noise. Conversely, discrete structured memory architectures provide robust knowledge retention but often struggle to adapt to evolving narratives. To address this, we propose GAM, a hierarchical Graph-based Agentic Memory framework that explicitly decouples memory encoding from consolidation to effectively resolve the conflict between rapid context perception and stable knowledge retention. By isolating ongoing dialogue in an event progression graph and integrating it into a topic associative network only upon semantic shifts, our approach minimizes interference while preserving long-term consistency. Additionally, we introduce a graph-guided, multi-factor retrieval strategy to enhance context precision. Experiments on LoCoMo and LongDialQA indicate that our method consistently outperforms state-of-the-art baselines in both reasoning accuracy and efficiency.