通过结构化记忆减少大模型过度个性化问题
Mitigating Over-Personalization in LLMs via Structured Memory
- 将用户记忆按领域分组呈现，避免信息混淆
- 在七种模型上平均降低8.8%的跨域泄露
- 无需修改模型即可提升对话真实性
对话助手越来越依赖长期记忆实现跨会话个性化。但当用户信息被引入模型上下文时，可能在不相关场景中产生不当影响。本文研究了两类问题：跨领域泄露（不同生活领域记忆相互干扰）和记忆诱导顺从性（模型更倾向于附和用户观点而非客观回应）。针对此，提出在推理阶段采用结构化记忆呈现方式，将记忆按领域分区，而非以无序列表注入。在PersistBench上的实验表明，该方法在不改变模型或记忆内容的前提下，显著降低跨领域泄露，最强方案平均减少8.8%的泄露率，同时保持原有功能效用。
原文摘要 · Abstract (English)
Conversational assistants increasingly rely on persistent long-term memory to personalize responses across sessions. However, when stored user information is reintroduced into the model context, it can also influence responses in inappropriate or unrelated settings. We study two such failure modes in memory-augmented LLMs: cross-domain leakage, where memories from one life domain affect responses in another, and memory-induced sycophancy, where stored user beliefs make models more likely to agree with the user rather than respond truthfully. We apply a simple inference-time modification to how memories are presented to the model, without changing the model or the memory contents. Across seven models on PersistBench, we compare the commonly used all-in context format, where memories are injected as an unstructured list, with structured formats that partition memories by domain. This simple modification consistently reduces cross-domain leakage while preserving utility, with our strongest method reducing leakage by $8.8\%$ on average relative to the baseline.
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。