检测、重掩、修复：面向动态上下文忠实摘要的扩散编辑
Detect, Remask, Repair: Diffusion Editing for Faithful Summarization of Evolving Contexts
AI总结：
提出DETECT-REMASK-REPAIR框架，利用掩码扩散语言模型识别并修复摘要中过时内容，在保持支持内容的同时实现局部忠实性修复，并引入StreamSum基准评估。
摘要：
现实世界事件的摘要可能随着上下文演变和新信息的到来而过时。常见的做法是从更新后的上下文生成新摘要，但完全重新生成会丢弃之前的草稿，可能掩盖变化，并且当只有少数声明不支持时可能不必要。我们研究局部忠实性修复：在保留支持内容的同时更新现有摘要中的过时片段。我们提出DETECT-REMASK-REPAIR，一个基于扩散的框架，通过掩码扩散语言模型识别、重新掩码并修复过时区域。为了评估动态上下文摘要，我们引入了StreamSum，一个合成事件时间线的基准。在DialogSum和StreamSum上的实验表明，局部扩散修复提供了一种可控的替代完全重写的方法：忠实性导向的修复改进了早期草稿，一步修复将修复成本降低到半秒以下，该框架实现了跨数据集的忠实性-速度-保留权衡。我们还发现该框架可以作为事后修正步骤，提高自回归系统的忠实性。
英文摘要：
Summaries of real-world events can become outdated as contexts evolve and new information arrives. A common response is to generate a new summary from the updated context, but full regeneration discards the previous draft, can obscure what changed, and may be unnecessary when only a few claims are unsupported. We study localized faithfulness repair: updating outdated spans in an existing summary while preserving supported content. We propose DETECT-REMASK-REPAIR, a diffusion-based framework that identifies, remasks, and repairs outdated regions with masked diffusion language models. To evaluate evolving-context summarization, we introduce StreamSum, a benchmark of synthetic event timelines. Experiments on DialogSum and StreamSum show that localized diffusion repair provides a controllable alternative to full rewriting: faithfulness-steered repair improves early drafts, one-step repair reduces repair cost to under half a second, with the framework enabling faithfulness-speed-preservation tradeoffs across datasets. We also find that the framework can provide a post-hoc correction step that improves faithfulness for autoregressive systems.
机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。
