基于查询语义特性的稠密文本检索模型  

Dense passage retrieval model based on query semantic characteristics

作　　者：赵铁柱 林伦凯 杨秋鸿 Zhao Tiezhu;Lin Lunkai;Yang Qiuhong(School of Computer Science&Technology,Dongguan University of Technology,Dongguan Guangdong 523808,China;School of Artificial Intelligence,Dongguan City University,Dongguan Guangdong 523419,China)

机构地区：[1]东莞理工学院计算机科学与技术学院,广东东莞523808 [2]东莞城市学院人工智能学院,广东东莞523419

出　　处：《计算机应用研究》2024年第5期1388-1393,共6页Application Research of Computers

基　　金：广东省普通高校重点领域专项资助项目(2021ZDZX3007);东莞市社会发展科技资助项目(20231800936732);东莞城市学院青年教师发展基金资助项目(2022QJY005Z)。

摘　　要：针对现有稠密文本检索模型(dense passage retrieval,DPR)存在的负采样效率低、易产生过拟合等问题,提出了一种基于查询语义特性的稠密文本检索模型(Q-DPR)。首先,针对模型的负采样过程,提出了一种基于近邻查询的负采样方法。该方法通过检索近邻查询,快速地构建高质量的负相关样本,以降低模型的训练成本。其次,针对模型易产生过拟合的问题,提出了一种基于对比学习的查询自监督方法。该方法通过建立查询间的自监督对比损失,缓解模型对训练标签的过拟合,从而提升模型的检索准确性。Q-DPR在面向开放领域问答的大型数据集MSMARCO上表现优异,取得了0.348的平均倒数排名以及0.975的召回率。实验结果证明,该模型成功地降低了训练的开销,同时也提升了检索的性能。Addressing the issues of low negative sampling efficiency and tendency towards overfitting in existing dense passage retrieval(DPR)models,this paper proposed a DPR model based on query semantic characteristics(Q-DPR).Firstly,it introduced a negative sampling method based on neighbor queries for the negative sampling process.This method constructed high-quality negative samples rapidly by retrieving neighboring queries,thereby reducing the training costs.Secondly,to mitigate overfitting,it proposed a query self-supervised method based on contrastive learning.This method alleviated overfitting to training labels by establishing a self-supervised contrastive loss among queries,thereby enhancing retrieval accuracy.Q-DPR performed exceptionally well on the large-scale MSMARCO dataset for open-domain question answering,achieving a mean reciprocal rank of 0.348 and a recall rate of 0.975.Experimental results demonstrate that this model successfully reduces trai-ning overhead while also improving retrieval performance.

关 键 词：查询 稠密文本检索 近邻 对比学习 自监督 

分 类 号：TP391[自动化与计算机技术—计算机应用技术]