构造基准与评估协议，分别评估"记忆是否包含解决 gap 的证据"与"生成摘要是否反映它"。提出 ReMEMBER 缺失证据记忆框架：基于未解决的窗口依赖做条件检索，在固定预算下把检索到的 chunk 精炼为证据密集记忆，供窗口摘要使用。
章节结构（全文标题提取）： 1 Introduction 2 Related Work 3 Streaming Dialogue Summarization: Formulation and Evaluation 4 Methodology: ReMEMBER 5 Evaluation 6 Conclusion · 3.1 Problem Formulation · 3.2 Benchmark Construction · 3.3 Evaluation Metrics · 4.1 Gap-Conditioned Evidence Retrieval · 4.2 Gap-Conditioned Chunk Refinement · 4.3 Integration with Summary Generation · 5.1 Experimental Setup · 5.2 Gap-Resolving Evidence Recovery · 5.3 Impact on Streaming Summaries · 5.4 Gap-Level Evidence Use
📊 结果
在历史长达 160K token 的对话上，ReMEMBER 在相同预算下显著提升记忆召回与 gap 分辨率完整性，优于记忆构建基线。
📝 原文摘要
▶ 原文摘要 Abstract
Users of modern platforms repeatedly need summaries of recent dialogue, but the window rarely contains enough context to be interpreted on its own. We formalize this setting as streaming dialogue summarization, where a system must summarize a current window using selective memory from an unbounded history under a fixed budget. We show that the central challenge is not how much history is accessed, but whether memory recovers the evidence that the current window presupposes. We construct a benchmark and evaluation protocol that separately assesses whether memory contains gap-resolving evidence and whether the generated summary reflects it. We propose ReMEMBER, a missing-evidence memory framework that conditions retrieval on unresolved window dependencies and refines retrieved chunks into evidence-dense memory under a fixed budget. Experiments on dialogues with histories up to 160K tokens show that ReMEMBER improves memory recall and gap-resolution completeness over memory construction baselines under the same budget.