为大模型设计五层记忆架构，提升对话连贯性与个性化。
MMAG: Mixed Memory-Augmented Generation for Large Language Models Applications
- 构建五层记忆体系：对话、长期用户、事件、感知上下文与短期工作记忆。
- 在Heero对话代理中验证，加密长期记忆显著提升用户参与度和留存率。
- 适用于需要长期交互的智能助手、客服系统等场景。
大型语言模型（LLMs）在单次提示下能生成连贯文本，但在持续交互中难以保持相关性、个性化和连贯性。人类交流依赖多种记忆形式，包括回忆过往对话、适应个人特征和情境。本文提出混合记忆增强生成（MMAG）框架，将记忆组织为五个相互作用的层级：对话记忆、长期用户记忆、情景与事件关联记忆、感知与上下文感知记忆、短期工作记忆。借鉴认知心理学，将各层级映射到技术组件，并设计协调、优先级与冲突解决策略。通过在Heero对话代理中的实现验证，加密的长期生物信息和对话历史已提升用户参与度与留存率。讨论了存储、检索、隐私与延迟等实现挑战，指出开放问题。MMAG为构建更连贯、主动且符合人类需求的记忆丰富型语言代理提供了基础。
原文摘要 · Abstract (English)
Large Language Models (LLMs) excel at generating coherent text within a single prompt but fall short in sustaining relevance, personalization, and continuity across extended interactions. Human communication, however, relies on multiple forms of memory, from recalling past conversations to adapting to personal traits and situational context. This paper introduces the Mixed Memory-Augmented Generation (MMAG) pattern, a framework that organizes memory for LLM-based agents into five interacting layers: conversational, long-term user, episodic and event-linked, sensory and context-aware, and short-term working memory. Drawing inspiration from cognitive psychology, we map these layers to technical components and outline strategies for coordination, prioritization, and conflict resolution. We demonstrate the approach through its implementation in the Heero conversational agent, where encrypted long-term bios and conversational history already improve engagement and retention. We further discuss implementation concerns around storage, retrieval, privacy, and latency, and highlight open challenges. MMAG provides a foundation for building memory-rich language agents that are more coherent, proactive, and aligned with human needs.
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。