CAPSTONE: Curriculum Sampling for Dense Retrieval with Document Expansion
Xingwei He, Yeyun Gong, A-Long Jin, Hang Zhang, Anlei Dong, Jian Jiao, Siu Yiu, Nan Duan
Abstract
The dual-encoder has become the de facto architecture for dense retrieval. Typically, it computes the latent representations of the query and document independently, thus failing to fully capture the interactions between the query and document. To alleviate this, recent research has focused on obtaining query-informed document representations. During training, it expands the document with a real query, but during inference, it replaces the real query with a generated one. This inconsistency between training and inference causes the dense retrieval model to prioritize query information while disregarding the document when computing the document representation. Consequently, it performs even worse than the vanilla dense retrieval model because its performance heavily relies on the relevance between the generated queries and the real query. In this paper, we propose a curriculum sampling strategy that utilizes pseudo queries during training and progressively enhances the relevance between the generated query and the real query. By doing so, the retrieval model learns to extend its attention from the document alone to both the document and query, resulting in high-quality query-informed document representations. Experimental results on both in-domain and out-of-domain datasets demonstrate that our approach outperforms previous dense retrieval models.- Anthology ID:
- 2023.emnlp-main.651
- Volume:
- Proceedings of the 2023 Conference on Empirical Methods in Natural Language Processing
- Month:
- December
- Year:
- 2023
- Address:
- Singapore
- Editors:
- Houda Bouamor, Juan Pino, Kalika Bali
- Venue:
- EMNLP
- SIG:
- Publisher:
- Association for Computational Linguistics
- Note:
- Pages:
- 10531–10541
- Language:
- URL:
- https://aclanthology.org/2023.emnlp-main.651
- DOI:
- 10.18653/v1/2023.emnlp-main.651
- Cite (ACL):
- Xingwei He, Yeyun Gong, A-Long Jin, Hang Zhang, Anlei Dong, Jian Jiao, Siu Yiu, and Nan Duan. 2023. CAPSTONE: Curriculum Sampling for Dense Retrieval with Document Expansion. In Proceedings of the 2023 Conference on Empirical Methods in Natural Language Processing, pages 10531–10541, Singapore. Association for Computational Linguistics.
- Cite (Informal):
- CAPSTONE: Curriculum Sampling for Dense Retrieval with Document Expansion (He et al., EMNLP 2023)
- PDF:
- https://preview.aclanthology.org/improve-issue-templates/2023.emnlp-main.651.pdf