用图结构记忆让智能体跨会话保持连贯，解决传统方法信息断裂或中心节点过载问题。
GAAMA: Graph Augmented Associative Memory for Agents
- 构建四类节点五种边的语义图，通过概念节点实现跨主题关联
- 在多轮对话数据集上比基线提升4.2个百分点，长对话中优势更明显
- 适合需要长期记忆与个性化交互的智能体系统，如客服、助手
能够跨多轮会话与用户交互的AI智能体需要持久的长期记忆以维持一致且个性化的行为。现有方法或依赖扁平化检索增强生成（RAG），导致记忆间结构关系丢失；或采用以实体为中心的知识图谱，但在对话数据中易出现巨型中心节点，削弱基于图的相关性传播。本文提出GAAMA，一种面向智能体的图增强关联记忆机制，通过三步流程构建概念中介的知识图：（1）保留原始会话片段，（2）利用大模型提取原子事实与主题级概念节点，（3）合成高阶反思内容。最终图谱包含四类节点（会话、事实、反思、概念）和五种边类型，概念节点提供跨领域遍历路径，避免实体中心设计的巨型节点问题。检索结合余弦相似度的k近邻搜索与边类型感知的个性化页面排名（PPR），通过加性评分函数融合。进一步提出GRAFT（通过增强事实与拓扑修复图谱）作为后检索修正层，诊断检索失败并精准修复知识图谱。在LoCoMo-10（1,540个问题，10次多轮对话）上，GAAMA获得79.1%平均奖励，较调优后的RAG基线提升4.2个百分点，为最强对比方法。在MemoryArena上，GAAMA在三个任务中均优于全上下文基线——群体旅行(+0.4~pp)、网络购物(+3.4~pp)、渐进式搜索(+0.7~pp)，且优势随对话长度单调增长。尤为显著的是，GAAMA在所有类别中表现稳定，每项均达到最佳对手水平，而所有竞争对手至少在一个类别中性能下降。
原文摘要 · Abstract (English)
AI agents that interact with users across multiple sessions require persistent long-term memory to maintain coherent, personalized behavior. Current approaches either rely on flat retrieval-augmented generation (RAG), which loses structural relationships among memories, or use entity-centric knowledge graphs that suffer from mega-hub effects in conversational data, diluting graph-based relevance propagation. We propose GAAMA, a graph-augmented associative memory for agents that constructs a concept-mediated knowledge graph through a three-step pipeline: (1)verbatim episode preservation, (2)LLM-based extraction of atomic facts and topic-level concept nodes, and (3)synthesis of higher-order reflections. The resulting graph uses four node types (episode, fact, reflection, concept) connected by five structural edge types, with concept nodes providing cross-cutting traversal paths that avoid the mega-hub problem of entity-centric designs. Retrieval combines cosine-similarity-based k-nearest neighbor search with edge-type-aware Personalized PageRank (PPR) through an additive scoring function. We further introduce GRAFT (Graph Repair by Augmenting Facts & Topology), a post-retrieval corrective layer that diagnoses retrieval failures and surgically repairs the knowledge graph. On LoCoMo-10 (1,540 questions, 10 multi-session conversations), GAAMA achieves 79.1% mean reward, a +4.2~pp improvement over a tuned RAG baseline, the strongest comparator. On MemoryArena, GAAMA outperforms full-context baselines across three tasks - Group Travel (+0.4~pp), Web Shopping (+3.4~pp), and Progressive Search (+0.7~pp) - with advantages growing monotonically with dialogue length. Notably, GAAMA delivers consistent performance across all categories, matching the best competing method in each, whereas every competitor degrades in at least one category.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。