Mandol：一种用于长期对话的聚合式智能体记忆系统
Mandol: An Agglomerative Agent Memory System for Long-Term Conversations
AI总结：
提出Mandol聚合式记忆系统，通过统一内存原生架构整合碎片化记忆，结合层次化记忆模型、聚合语义数据结构与量化查询机制，在长期对话基准上取得最佳准确率并实现5.4倍检索加速。
摘要：
长期对话智能体需要记忆和查询跨会话、多类型且具有复杂关联的信息。现有的智能体记忆系统依赖异构的向量和图数据库，导致记忆信息碎片化并产生高跨数据库I/O延迟。在检索方面，常见的RAG风格方法容易引入噪声、遗漏相关线索且缺乏令牌预算控制，降低了LLM的准确性和效率。\n我们提出Mandol，一种聚合式记忆系统，将碎片化的记忆表示和存储整合到统一的内存原生架构中。其核心组件包括：(1) 层次化记忆模型，将记忆组织为基础层（表示原始记忆信息）和高层抽象层（将基础记忆聚合成可追踪的抽象记忆），两者统一表示为结构化语义图；(2) 聚合语义数据结构，结合SemanticMap和SemanticGraph，原生融合键值、向量和图结构，并提供统一混合检索算子以消除跨数据库I/O；(3) 量化查询机制，具有查询自适应路由、量化去噪与冲突解决以及令牌约束的上下文生成，检索过程中无需LLM参与。在两个广泛使用的长期对话基准LoCoMo和LongMemEval上的实验表明，Mandol在代表性智能体记忆系统中取得了最佳整体准确率。在性能比较中，Mandol在10 QPS并发负载下实现了5.4倍的检索加速和4.8倍的插入加速，同时在消费级硬件上保持低延迟。
英文摘要：
Long-term conversational agents need to remember and query cross-session, multi-typed information with complex correlations. Existing agent memory systems rely on heterogeneous vector and graph databases, which fragment memory information and cause high cross-database I/O latency. For retrieval, common RAG-style methods tend to introduce noise, miss correlated clues, and lack token budget control, degrading LLM accuracy and efficiency. We propose Mandol, an agglomerative memory system that consolidates fragmented memory representations and storage into a unified memory-native architecture. Its core components include: (1) a hierarchical memory model that organizes memory into a basic layer representing raw memory information and a high-level abstract layer that agglomerates basic memories into traceable abstract memories, both uniformly represented as structured semantic graphs; (2) an agglomerative semantic data structure combining SemanticMap and SemanticGraph, which natively fuses key-value, vector, and graph structures and provides unified hybrid retrieval operators to eliminate cross-database I/O; and (3) a quantitative query mechanism with query-adaptive routing, quantitative denoising and conflict resolution, and token-constrained context generation, all without involving LLMs during retrieval. Experiments on two widely used long-term conversation benchmarks, LoCoMo and LongMemEval, show that Mandol achieves the best overall accuracy among representative agent memory systems. For performance comparison, Mandol also obtains a 5.4x retrieval speedup and a 4.8x insertion speedup under 10 QPS concurrent load, while still maintaining low latency on consumer-grade hardware.
机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。
