Mnemis: Dual-Route Retrieval on Hierarchical Graphs for Long-Term LLM Memory

TL;DR

Mnemis introduces a dual-route memory retrieval framework for LLMs that combines similarity search with hierarchical, top-down reasoning to improve long-term memory access.

Contribution

It presents a novel hierarchical graph-based memory system integrating System-1 and System-2 retrieval mechanisms, enhancing relevance and coverage.

Findings

Achieves state-of-the-art scores on LoCoMo and LongMemEval-S benchmarks.

Effectively combines similarity search with hierarchical reasoning.

Outperforms existing memory retrieval methods.

Abstract

AI Memory, specifically how models organizes and retrieves historical messages, becomes increasingly valuable to Large Language Models (LLMs), yet existing methods (RAG and Graph-RAG) primarily retrieve memory through similarity-based mechanisms. While efficient, such System-1-style retrieval struggles with scenarios that require global reasoning or comprehensive coverage of all relevant information. In this work, We propose Mnemis, a novel memory framework that integrates System-1 similarity search with a complementary System-2 mechanism, termed Global Selection. Mnemis organizes memory into a base graph for similarity retrieval and a hierarchical graph that enables top-down, deliberate traversal over semantic hierarchies. By combining the complementary strength from both retrieval routes, Mnemis retrieves memory items that are both semantically and structurally relevant. Mnemis achieves state-of-the-art performance across all compared methods on long-term memory benchmarks, scoring 93.9 on LoCoMo and 91.6 on LongMemEval-S using GPT-4.1-mini.