通过实体结构索引提升长文档问答准确率，解决跨实体证据定位难题。
EnSI-RAG: Entity-Structure-Indexed Retrieval-Augmented Generation for Long-Document Question Answering
- 构建以实体为中心的结构化索引，分离证据查找与答案生成
- 在Loong和Oolong数据集上达到78.24%平均准确率，领先基线6.62分
- 适合需要多跳推理的复杂长文档问答任务
长文档问答仍具挑战性，因相关证据可能跨越多个实体及其关系。现有检索增强生成（RAG）方法通常将文档切分为原始片段并基于嵌入相似性检索，当片段边界割裂实体与支持证据，或问题需跨文档多跳推理时性能下降。本文提出EnSI-RAG（实体-结构索引检索增强生成），构建与查询无关的实体中心索引。每个记录（e, t, k, v）包含实体e、类型t、语义类别k（属性、关系、方面）及值v，同时保留对原始文本段落的链接。查询时，这些记录作为检索依据，大模型将检索到的段落合成最终答案。该设计实现证据定位与答案生成解耦，并保持可追溯的来源证据。在Loong和Oolong数据集上，EnSI-RAG平均准确率达78.24%，相较基准分数提升6.62分，验证了其有效性。代码已开源。
原文摘要 · Abstract (English)
Question answering (QA) over long, connected documents remains challenging because relevant evidence may span multiple entities and their relationships. Existing retrieval-augmented generation (RAG) methods typically index documents as raw chunks and retrieve them through embedding similarity. Their performance degrades when chunk boundaries separate entities from supporting evidence or when a question requires multi-hop reasoning across the corpus. We propose EnSI-RAG (Entity-Structure-Indexed Retrieval-Augmented Generation), a framework that constructs a query-independent, entity-centered index. Each record (e, t, k, v) represents an entity e, its type t, a semantic category k in {property, relation, aspect}, and a value v, while retaining links to the original source passages. At query time, these records serve as retrieval handles, and an LLM synthesizes the retrieved passages into the final answer. This design separates evidence localization from answer synthesis while preserving traceable source evidence. Across Loong and Oolong, EnSI-RAG achieves an average accuracy of 78.24. Relative to the published baseline scores used as references, this is 6.62 points higher, suggesting its effectiveness across these settings. The code is available at https://github.com/RamonMeng/EnSI-RAG.
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。