English

CAPSTONE: Curriculum Sampling for Dense Retrieval with Document Expansion

Computation and Language 2023-10-31 v2

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.

Keywords

Cite

@article{arxiv.2212.09114,
  title  = {CAPSTONE: Curriculum Sampling for Dense Retrieval with Document Expansion},
  author = {Xingwei He and Yeyun Gong and A-Long Jin and Hang Zhang and Anlei Dong and Jian Jiao and Siu Ming Yiu and Nan Duan},
  journal= {arXiv preprint arXiv:2212.09114},
  year   = {2023}
}

Comments

Accetpted to EMNLP 2023

R2 v1 2026-06-28T07:41:02.471Z