用句子图结构管理对话记忆，让长对话问答更准确。
SGMem: Sentence Graph Memory for Long-Term Conversational Agents
- 将对话转为分块的句子图，跨轮次关联信息
- 在LongMemEval和LoCoMo上显著提升问答准确率
- 适合需要长期记忆的对话系统开发者
长时对话智能体需有效管理超出大语言模型上下文窗口的对话历史。现有基于事实提取或摘要的方法虽减少冗余，但难以在不同粒度的对话中组织与检索信息。我们提出SGMem（句子图记忆），将对话表示为分块单元内的句子级图，捕捉跨回合、跨轮次及会话级上下文的关联。通过结合原始对话片段与生成的记忆（如摘要、事实、洞察），SGMem为大模型提供连贯且相关的上下文以生成回复。在LongMemEval和LoCoMo上的实验表明，SGMem持续提升准确率，优于多个强基线。
原文摘要 · Abstract (English)
Long-term conversational agents require effective memory management to handle dialogue histories that exceed the context window of large language models (LLMs). Existing methods based on fact extraction or summarization reduce redundancy but struggle to organize and retrieve relevant information across different granularities of dialogue and generated memory. We introduce SGMem (Sentence Graph Memory), which represents dialogue as sentence-level graphs within chunked units, capturing associations across turn-, round-, and session-level contexts. By combining retrieved raw dialogue with generated memory such as summaries, facts and insights, SGMem supplies LLMs with coherent and relevant context for response generation. Experiments on LongMemEval and LoCoMo show that SGMem consistently improves accuracy and outperforms strong baselines in long-term conversational question answering.
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。