PairDistill: Pairwise Relevance Distillation for Dense Retrieval
Abstract: Effective information retrieval (IR) from vast datasets relies on advanced techniques to extract relevant information in response to queries. Recent advancements in dense passage retrieval (DPR) have showcased remarkable efficacy compared to traditional sparse retrieval methods. To further enhance retrieval performance, knowledge distillation techniques, often leveraging robust cross-encoder rerankers, have been extensively explored. However, existing approaches primarily distill knowledge from pointwise rerankers, which assign absolute relevance scores to documents, thus facing challenges related to inconsistent standards. This paper introduces Pairwise Relevance Distillation (PairDistill) to leverage pairwise reranking, offering fine-grained distinctions between similarly relevant documents to enrich the training of dense retrieval models.
Our experiments demonstrate that PairDistill outperforms existing methods, achieving new state-of-the-art results across multiple benchmarks. This highlights the potential of PairDistill in advancing dense retrieval techniques effectively. Our source code and trained models are released at https://anonymous.4open.science/r/pair-distill-AE1F
Paper Type: Long
Research Area: Information Retrieval and Text Mining
Research Area Keywords: passage retrieval, dense retrieval, re-ranking
Contribution Types: NLP engineering experiment
Languages Studied: English
Submission Number: 3001
Loading