用重排技术提升财报问答准确率，显著减少错误答案。
Enhancing Financial Report Question-Answering: A Retrieval-Augmented Generation System with Reranking Analysis
- 结合全文与语义检索，再用交叉编码器重排结果。
- 重排后正确率提升至49.0%，错误率从35.3%降至22.5%。
- 适合金融分析、智能投研人员参考使用。
财务分析师在处理超过100页的10-K年报时面临巨大挑战。本文提出一种用于回答标普500公司财报问题的检索增强生成（RAG）系统，并评估神经重排对性能的影响。该流程采用混合搜索策略，结合全文检索与语义检索，随后可选地通过交叉编码器模型进行重排。我们在FinDER基准数据集上进行系统评估，包含1,500个查询，分五个实验组。结果表明，重排显著提升答案质量：得分8及以上正确率达49.0%，较无重排时的33.5%提高15.5个百分点；完全错误答案的误差率由35.3%下降至22.5%。研究强调了重排在金融RAG系统中的关键作用，并通过现代语言模型与优化检索策略实现了对基线方法的性能超越。
原文摘要 · Abstract (English)
Financial analysts face significant challenges extracting information from lengthy 10-K reports, which often exceed 100 pages. This paper presents a Retrieval-Augmented Generation (RAG) system designed to answer questions about S&P 500 financial reports and evaluates the impact of neural reranking on system performance. Our pipeline employs hybrid search combining full-text and semantic retrieval, followed by an optional reranking stage using a cross-encoder model. We conduct systematic evaluation using the FinDER benchmark dataset, comprising 1,500 queries across five experimental groups. Results demonstrate that reranking significantly improves answer quality, achieving 49.0 percent correctness for scores of 8 or above compared to 33.5 percent without reranking, representing a 15.5 percentage point improvement. Additionally, the error rate for completely incorrect answers decreases from 35.3 percent to 22.5 percent. Our findings emphasize the critical role of reranking in financial RAG systems and demonstrate performance improvements over baseline methods through modern language models and refined retrieval strategies.
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。