arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 2609.26634cs.CL

知识拉取请求:面向持续文档撰写的框架

Knowledge Pull Requests for Continual Document Authoring

发表机构约翰霍普金斯大学
查看机构详情
  • Johns Hopkins University(约翰霍普金斯大学)

机构由 AI 辅助整理,请以论文原文为准。

Alexander Martin, Benjamin Van Durme

首次发表
浏览论文内容

中文总结 AI 辅助

提出知识拉取请求(KPRs)框架,通过提取、过滤、路由声明并标记冲突,实现可解释的持续文档修订,在维基百科跨语言修订和RAGTIME报告中优于重写或重新生成,并提升问答依据。

中文摘要 AI 辅助

我们提出了知识拉取请求(KPRs),这是一个用于持续文档撰写的框架,使得每一次修改都可解释。文档需要随着新知识从其他来源、语言或时代出现而不断修订,但现有方法要么在编辑时不考虑知识的变化,要么从头重新生成。KPR通过提取声明、过滤并将其路由到各个章节,并标记与现有内容的冲突,将新知识整合到文档中,生成一个变更日志,该日志将知识的变化(声明提案)与文本的变化(文档差异)分开。我们在跨语言的维基百科修订以及RAGTIME上的查询驱动报告更新中评估了KPRs。与从来源重写或从头重新生成相比,KPRs整合了更多信息,并更好地保留了现有内容,同时每个生成的令牌添加的信息最多。此外,与使用搜索的前沿模型相比,KPR修订的文章在问答任务中提供了更好的依据,因为搜索无法揭示仅在其他语言中记录的知识。

英文摘要

We introduce Knowledge Pull Requests (KPRs), a framework for continual document authoring that makes each change interpretable. Documents require ongoing revision as new knowledge surfaces from other sources, languages, or times, but existing approaches either edit with no account of what knowledge changed or regenerate from scratch. A KPR integrates new knowledge into a document by extracting claims, filtering and routing them to sections, and flagging conflicts with existing content, producing a ChangeLog that separates what knowledge changes (claim proposal) from how the text changes (document diff). We evaluate KPRs on revising Wikipedia across languages and updating query-driven reports on RAGTIME. KPRs integrate more information and better preserve existing content than rewriting from sources or regenerating from scratch, while adding the most information per token generated. A KPR-revised article also grounds question answering better than a frontier model with search, which does not surface knowledge documented only in other languages.

补充信息

↑