用多个正向段落训练，提升稠密段落检索准确率
Improving Dense Passage Retrieval with Multiple Positive Passages
- 训练时为每个问题引入多个正向段落
- 在小批量下仍显著提升检索准确率
- 适合资源有限但追求精度的场景
基于双编码器架构的稠密段落检索（DPR）在段落检索准确率上已超越传统稀疏检索方法（如BM25）。尽管近期方法进一步提升了DPR性能，但这些模型通常在训练时仅将每个问题与一个正向段落配对，多正向段落的影响尚未被研究。本文探讨在训练中引入额外正向段落对DPR性能的影响。实验结果表明，为每个问题配备多个正向段落可持续提升检索准确率，即使使用极小批量（batch size）也能实现，支持单张GPU训练。
原文摘要 · Abstract (English)
By leveraging a dual encoder architecture, Dense Passage Retrieval (DPR) has outperformed traditional sparse retrieval algorithms such as BM25 in terms of passage retrieval accuracy. Recently proposed methods have further enhanced DPR's performance. However, these models typically pair each question with only one positive passage during training, and the effect of associating multiple positive passages has not been examined. In this paper, we explore the performance of DPR when additional positive passages are incorporated during training. Experimental results show that equipping each question with multiple positive passages consistently improves retrieval accuracy, even when using a significantly smaller batch size, which enables training on a single GPU.
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。