用于编排多智能体多模态记忆整理的历时超图

中文摘要 AI 辅助

多智能体系统通过协作、工具使用、多模态推理与编排完成任务，但每个智能体受限于自身观测、上下文与资源构成的知识边界。记忆需在交互间保存并传递证据、角色特定上下文、决策、流程与经验，而非仅结果。向量记忆与图记忆将这些结构平化为嵌入或二元迹，模糊了涉及智能体、工具、文档、错误与证据的事件，限制了知识共享、追踪、复用、修订与编排。我们提出MAGE，一种基于超图的多模态数据库，设计为多智能体系统（MAS）的记忆引擎。MAGE在异构时序超图中存储智能体、消息、工具、错误、流程、文档、实体、决策与证据，将高阶协作事件保存为可复用记忆，支持决策驱动更新、角色感知检索、验证、生命周期管理与预算受限的上下文打包。通过向智能体与编排器提供知识，MAGE在不修改模型的前提下扩展了它们的知识边界。实验显示，MAGE在多种记忆基准测试中表现优于现有方法。

英文摘要

Multi-agent systems solve tasks through collaboration, tool use, multimodal reasoning, and orchestration, but each agent operates within a knowledge boundary defined by its observations, context, and resources. Memory must preserve and transfer evidence, role specific context, decisions, procedures, and experience across interactions, not only outcomes. Vector and graph memories flatten these structures into embeddings or dyadic traces, obscuring events involving agents, tools, documents, errors, and evidence. This limits knowledge sharing, tracing, reuse, revision, and orchestration. We present MAGE, a hypergraph based multimodal database designed as a memory engine for MAS. MAGE stores agents, messages, tools, errors, procedures, documents, entities, decisions, and evidence in a heterogeneous temporal hypergraph, preserving high order collaborative events as reusable memory. It supports decision driven updates, role aware retrieval, validation, lifecycle management, and budget bounded context packing. By delivering knowledge to agents and orchestrators, MAGE expands their knowledge boundaries without modifying the models. Experiments show MAGE outperforms on various memory baselines.

发表机构

• LIGHTSPEED, Singapore（光速（新加坡））
• Nanyang Technological University, Singapore（南洋理工大学）
• Emory University, Atlanta, USA（埃默里大学）

机构由 AI 辅助整理，请以论文原文为准。