TriSampler: A Better Negative Sampling Principle for Dense Retrieval
Zhen Yang, Zhou Shao, Yuxiao Dong, Jie Tang
Abstract
Negative sampling stands as a pivotal technique in dense retrieval, essential for training effective retrieval models and significantly impacting retrieval performance. While existing negative sampling methods have made commendable progress by leveraging hard negatives, a comprehensive guiding principle for constructing negative candidates and designing negative sampling distributions is still lacking. To bridge this gap, we embark on a theoretical analysis of negative sampling in dense retrieval. This exploration culminates in the unveiling of the quasi-triangular principle, a novel framework that elucidates the triangular-like interplay between query, positive document, and negative document. Fueled by this guiding principle, we introduce TriSampler, a straightforward yet highly effective negative sampling method. The keypoint of TriSampler lies in its ability to selectively sample more informative negatives within a prescribed constrained region. Experimental evaluation show that TriSampler consistently attains superior retrieval performance across a diverse of representative retrieval models.
BibTeX
@article{Yang_Shao_Dong_Tang_2024, title={TriSampler: A Better Negative Sampling Principle for Dense Retrieval}, volume={38}, url={https://ojs.aaai.org/index.php/AAAI/article/view/28779}, DOI={10.1609/aaai.v38i8.28779}, abstractNote={Negative sampling stands as a pivotal technique in dense retrieval, essential for training effective retrieval models and significantly impacting retrieval performance. While existing negative sampling methods have made commendable progress by leveraging hard negatives, a comprehensive guiding principle for constructing negative candidates and designing negative sampling distributions is still lacking. To bridge this gap, we embark on a theoretical analysis of negative sampling in dense retrieval. This exploration culminates in the unveiling of the quasi-triangular principle, a novel framework that elucidates the triangular-like interplay between query, positive document, and negative document. Fueled by this guiding principle, we introduce TriSampler, a straightforward yet highly effective negative sampling method. The keypoint of TriSampler lies in its ability to selectively sample more informative negatives within a prescribed constrained region. Experimental evaluation show that TriSampler consistently attains superior retrieval performance across a diverse of representative retrieval models.}, number={8}, journal={Proceedings of the AAAI Conference on Artificial Intelligence}, author={Yang, Zhen and Shao, Zhou and Dong, Yuxiao and Tang, Jie}, year={2024}, month={Mar.}, pages={9269-9277} }