利用文档结构提升问答系统检索、重排序与生成效果
Structures Facilitate Retrieve, Rerank, and Generate
- 通过对比同章节段落优化表示，增强检索能力
- 基于文档结构分组候选，融合上下文信息重排序
- 结合子图上下文生成更连贯的回答，支持中英文
文档驱动对话系统（DGDS）依赖外部文档回答领域特定问题。现有方法通常将文档拆分为独立段落进行检索和生成，未能有效利用文档内部结构信息，也缺乏足够的上下文支持知识选择与回复生成。本文提出SF-Re2G系统，系统性解决上述问题：首先，通过对比同一章节的段落来优化段落表示，提升检索性能；其次，构建结构增强型重排序器，利用同一对话轮次的多个依据段落往往位于相近位置的特性，将检索候选按文档结构分组为子图，并结合组内信息重新评分；最后，选取的段落用于生成回复时，考虑子图上下文以提升生成质量。在两个DGDS数据集上的实验验证了该方法在中英文场景下的有效性。
原文摘要 · Abstract (English)
Document-grounded dialogue systems (DGDS) utilize knowledge from external documents to answer domain-specific user questions. Existing solutions typically divide documents into independent passages for retrieval and response generation. This approach, however, neither makes good use of structural information within documents nor provides enough (document) context for knowledge selection and responses. This paper proposes SF-Re2G to address such issues systematically. Firstly, we seek to improve a passage representation by contrasting it with others of the same section, thus improving the retrieval performance. Secondly, a structure-enhanced reranker is built, leveraging the fact that multiple grounding passages of one dialog turn tend to be in the same neighborhood. Specifically, candidates from the retrieval are grouped into subgraphs according to the document structure. The reranker will rescore the candidate integrating its group information. Finally, the chosen passages are used for responses, taking into account the subgraph context for better generation. Experimental results on two DGDS datasets validate our method for both Chinese and English.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。