More Robust Dense Retrieval with Contrastive Dual Learning

TL;DR

This paper introduces DANCE, a contrastive dual learning approach that enhances dense retrieval by learning fine-grained query representations, resulting in a more uniform embedding space and improved retrieval performance.

Contribution

DANCE is a novel training paradigm that jointly optimizes query and document embeddings, improving the quality of query representations for dense retrieval.

Findings

DANCE achieves superior ranking performance on MS MARCO.

It produces a more uniform and discriminative embedding space.

Enhanced query representations lead to better retrieval accuracy.

Abstract

Dense retrieval conducts text retrieval in the embedding space and has shown many advantages compared to sparse retrieval. Existing dense retrievers optimize representations of queries and documents with contrastive training and map them to the embedding space. The embedding space is optimized by aligning the matched query-document pairs and pushing the negative documents away from the query. However, in such training paradigm, the queries are only optimized to align to the documents and are coarsely positioned, leading to an anisotropic query embedding space. In this paper, we analyze the embedding space distributions and propose an effective training paradigm, Contrastive Dual Learning for Approximate Nearest Neighbor (DANCE) to learn fine-grained query representations for dense retrieval. DANCE incorporates an additional dual training object of query retrieval, inspired by the classic information retrieval training axiom, query likelihood. With contrastive learning, the dual training object of DANCE learns more tailored representations for queries and documents to keep the embedding space smooth and uniform, thriving on the ranking performance of DANCE on the MS MARCO document retrieval task. Different from ANCE that only optimized with the document retrieval task, DANCE concentrates the query embeddings closer to document representations while making the document distribution more discriminative. Such concentrated query embedding distribution assigns more uniform negative sampling probabilities to queries and helps to sufficiently optimize query representations in the query retrieval task. Our codes are released at https://github.com/thunlp/DANCE.