发表机构
Samsung Research America（美国三星研究院）
机构由 AI 辅助整理，请以论文原文为准。AI 中文总结
针对对话AI记忆形成阶段的关键约束，提出Gated Memory框架，通过准入门与条件丰富阶段优化事实存储，在LoCoMo-10基准上实现LLM评判准确率2.6%的相对提升。
AI 中文摘要
个性化对话AI依赖长期记忆系统，该系统从用户话语中提取事实并存储于持久向量库中。尽管在检索、去重和生命周期管理方面已取得进展，但事实首次写入存储的形成阶段几乎未得到原则性关注，我们将其视为生产系统中记忆质量的关键约束。关键上下文信号（如永久用户属性与临时情境的区分）仅存在于原始话语中，在提取生成主谓宾三元组时会不可逆丢失，下游流程无法恢复。我们提出Gated Memory（门控记忆），这是一种轻量级模块化形成框架，在对话与存储之间插入两个决策检查点：准入门（Admission Gate）在提取运行前针对完整话语上下文评估每个候选事实；条件丰富阶段（Conditional Enrichment Stage）通过带隐私约束的实体范围分类法为已准入事实提供依据。准入门仅评估当前对话轮次，同时将之前轮次作为只读参考上下文，生成结构化形成记录。已准入内容被分解为原子事实，每个事实都经过分类、标注来源（直接陈述或推断）、限定适用范围，并基于已解析的时间和地点，且遵循“不得断言上下文中不存在的实体”的约束。在话语数据具有非典型情感密度的LoCoMo-10基准上，Gated Memory在LLM评判准确率上较采用相同检索与生成机制的强基线实现了2.6%的相对提升，确立了形成质量是可衡量的记忆性能约束。
英文摘要
Personalized conversational AI relies on long-term memory systems that extract facts from user utterances and store them in persistent vector stores. Despite progress in retrieval, deduplication, and lifecycle management, the formation stage, the moment a fact is first written to storage has received almost no principled attention. We identify this as the binding constraint on memory quality in production systems. Critical contextual signals, such as the distinction between a permanent user attribute and a transient situation, exist only in the original utterance and are irreversibly lost the moment extraction produces a subject-relation-object triple. No downstream process can recover them. We propose Gated Memory, a lightweight, modular formation framework that interposes two decision checkpoints between conversation and storage: an admission gate that evaluates every candidate fact against the full utterance context before extraction runs, and a conditional enrichment stage that grounds admitted facts through an entity scope taxonomy with privacy constraints. The gate evaluates only the current exchange while using prior turns as read-only reference context, and produces a structured formation record. Admitted content is decomposed into atomic facts, each categorized, tagged with provenance (directly stated versus inferred), scoped to its condition of applicability, and grounded in resolved time and place, subject to a constraint that no entity absent from the context may be asserted. On the LoCoMo-10 benchmark with atypical emotional density in utterance data, Gated Memory achieves an overall +2.6% relative improvement in LLM-judge accuracy over a strong baseline with identical retrieval and generation, establishing formation quality as a measurable constraint on memory performance.
