针对金融文档重复内容导致检索失效，提出分层检索与证据清洗框架
Hierarchical Retrieval with Evidence Curation for Open-Domain Financial Question Answering on Standardized Documents
- 分层检索先选文档再挑段落，减少相似文本干扰
- 自动剔除无关段落，必要时生成新查询补全信息
- 专为标准化金融文件设计，适合金融问答场景
基于检索增强生成（RAG）的大语言模型在金融领域广泛用于知识密集型任务。然而，标准化文档（如美国证监会申报文件）具有相似格式、重复性模板和结构化表格，导致传统RAG方法误检近似重复文本，引发冗余检索，影响准确性和完整性。为此，本文提出分层检索与证据清洗（HiREC）框架：首先分层检索，先定位相关文档，再从中筛选最相关段落；随后通过证据清洗去除无关内容，必要时自动生成补充查询以获取缺失信息。为评估该方法，我们构建并发布大规模开放域金融问答基准LOFin，包含145,897份SEC文件和1,595个问答对。代码与数据已开源于https://github.com/deep-over/LOFin-bench-HiREC。
原文摘要 · Abstract (English)
Retrieval-augmented generation (RAG) based large language models (LLMs) are widely used in finance for their excellent performance on knowledge-intensive tasks. However, standardized documents (e.g., SEC filing) share similar formats such as repetitive boilerplate texts, and similar table structures. This similarity forces traditional RAG methods to misidentify near-duplicate text, leading to duplicate retrieval that undermines accuracy and completeness. To address these issues, we propose the Hierarchical Retrieval with Evidence Curation (HiREC) framework. Our approach first performs hierarchical retrieval to reduce confusion among similar texts. It first retrieve related documents and then selects the most relevant passages from the documents. The evidence curation process removes irrelevant passages. When necessary, it automatically generates complementary queries to collect missing information. To evaluate our approach, we construct and release a Large-scale Open-domain Financial (LOFin) question answering benchmark that includes 145,897 SEC documents and 1,595 question-answer pairs. Our code and data are available at https://github.com/deep-over/LOFin-bench-HiREC.
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。