LAQuer：内容基于生成中的局部化归因查询
LAQuer: Localized Attribution Queries in Content-grounded Generation
AI总结：
本文提出LAQuer任务，将生成文本中的选定片段局部化归因到对应源片段，通过提示LLM和利用内部表示构建建模框架，在多文档摘要和长文本问答上验证其显著缩短归因文本长度。
摘要：
基于内容的文本生成模型经常产生偏离其源材料的内容，需要用户验证以确保准确性。现有的归因方法将整个句子与源文档关联起来，这对于希望核查特定声明的用户来说可能过于繁重。相比之下，现有的子句级归因方法可能更精确，但未能与用户的兴趣对齐。鉴于这些局限性，我们提出了局部化归因查询（LAQuer），这是一项新任务，将生成输出中选定的文本片段定位到其对应的源文本片段，从而实现细粒度和用户导向的归因。我们比较了两种用于LAQuer任务的方法，包括提示大型语言模型（LLMs）和利用LLM内部表示。然后，我们探索了一个建模框架，将现有的归因文本生成方法扩展到LAQuer。我们在两个基于内容的文本生成任务上评估了该框架：多文档摘要（MDS）和长文本问答（LFQA）。我们的研究结果表明，LAQuer方法显著缩短了归因文本的长度。我们的贡献包括：（1）提出LAQuer任务以增强归因的可用性，（2）提出一个建模框架并对多个基线进行基准测试，（3）提出一个新的评估设置，以促进内容基于生成中局部化归因的未来研究。
英文摘要：
Grounded text generation models often produce content that deviates from their source material, requiring user verification to ensure accuracy. Existing attribution methods associate entire sentences with source documents, which can be overwhelming for users seeking to fact-check specific claims. In contrast, existing sub-sentence attribution methods may be more precise but fail to align with users' interests. In light of these limitations, we introduce Localized Attribution Queries (LAQuer), a new task that localizes selected spans of generated output to their corresponding source spans, allowing fine-grained and user-directed attribution. We compare two approaches for the LAQuer task, including prompting large language models (LLMs) and leveraging LLM internal representations. We then explore a modeling framework that extends existing attributed text generation methods to LAQuer. We evaluate this framework across two grounded text generation tasks: Multi-document Summarization (MDS) and Long-form Question Answering (LFQA). Our findings show that LAQuer methods significantly reduce the length of the attributed text. Our contributions include: (1) proposing the LAQuer task to enhance attribution usability, (2) suggesting a modeling framework and benchmarking multiple baselines, and (3) proposing a new evaluation setting to promote future research on localized attribution in content-grounded generation.
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。
