SimANS: Simple Ambiguous Negatives Sampling for Dense Text Retrieval

Kun Zhou, Yeyun Gong, Xiao Liu, Wayne Xin Zhao, Yelong Shen, Anlei Dong, Jingwen Lu, Rangan Majumder, Ji-rong Wen, Nan Duan


Abstract
Sampling proper negatives from a large document pool is vital to effectively train a dense retrieval model. However, existing negative sampling strategies suffer from the uninformative or false negative problem. In this work, we empirically show that according to the measured relevance scores, the negatives ranked around the positives are generally more informative and less likely to be false negatives. Intuitively, these negatives are not too hard (may be false negatives) or too easy (uninformative). They are the ambiguous negatives and need more attention during training.Thus, we propose a simple ambiguous negatives sampling method, SimANS, which incorporates a new sampling probability distribution to sample more ambiguous negatives.Extensive experiments on four public and one industry datasets show the effectiveness of our approach.We made the code and models publicly available in https://github.com/microsoft/SimXNS.
Anthology ID:
2022.emnlp-industry.56
Volume:
Proceedings of the 2022 Conference on Empirical Methods in Natural Language Processing: Industry Track
Month:
December
Year:
2022
Address:
Abu Dhabi, UAE
Editors:
Yunyao Li, Angeliki Lazaridou
Venue:
EMNLP
SIG:
Publisher:
Association for Computational Linguistics
Note:
Pages:
548–559
Language:
URL:
https://aclanthology.org/2022.emnlp-industry.56
DOI:
10.18653/v1/2022.emnlp-industry.56
Bibkey:
Cite (ACL):
Kun Zhou, Yeyun Gong, Xiao Liu, Wayne Xin Zhao, Yelong Shen, Anlei Dong, Jingwen Lu, Rangan Majumder, Ji-rong Wen, and Nan Duan. 2022. SimANS: Simple Ambiguous Negatives Sampling for Dense Text Retrieval. In Proceedings of the 2022 Conference on Empirical Methods in Natural Language Processing: Industry Track, pages 548–559, Abu Dhabi, UAE. Association for Computational Linguistics.
Cite (Informal):
SimANS: Simple Ambiguous Negatives Sampling for Dense Text Retrieval (Zhou et al., EMNLP 2022)
Copy Citation:
PDF:
https://preview.aclanthology.org/dois-2013-emnlp/2022.emnlp-industry.56.pdf