GraphMemix: Query-Aware Evidence Forests for Long-Term Multimodal Agent Memory

2026-08-27

AI Summary

本文提出GraphMemix，通过构建查询感知证据森林解决多模态代理长期记忆组织问题，减少冗余信息并提高准确性。

Abstract

Organizing long-term memory for multimodal agents remains challenging because existing methods either suffer from expensive question-agnostic offline summaries or naive embedding similarity matching that introduces incomplete and redundant context. To address these issues, we propose GraphMemix, a combinatorial-optimization graph memory framework that models memory organization as query-aware evidence-forest construction. Specifically, our method consists of three key components:(1) candidate graph construction, which expands multi-view seed memories through schema and semantic relations to acquire query-aware original context; (2) evidence utility and activation costs, which decouples direct memory support from anchor-conditioned relation verification to suppress redundant or conflicting information; and (3) forest optimization, which jointly selects a forest-format memory context under a maximum evidence budget and its reliable relational structure. By organizing memory into a query-relevant subgraph, the method avoids substantial lifecycle cost and recovers low-similarity complementary evidence. Experimental results across four long-term multimodal memory benchmarks demonstrate significant improvements with different foundation models and establish a new Pareto frontier between accuracy and lifecycle cost.

Problem

• long-term memory
• multimodal agents
• query-agnostic summaries
• embedding similarity matching
• incomplete and redundant context

Innovation

• query-aware evidence-forest
• candidate graph construction
• evidence utility and activation costs
• forest optimization