Pseudo-Relevance for Enhancing Document Representation
Jihyuk Kim, Seung-won Hwang, Seoho Song, Hyeseon Ko, Young-In Song
Abstract
This paper studies how to enhance the document representation for the bi-encoder approach in dense document retrieval. The bi-encoder, separately encoding a query and a document as a single vector, is favored for high efficiency in large-scale information retrieval, compared to more effective but complex architectures. To combine the strength of the two, the multi-vector representation of documents for bi-encoder, such as ColBERT preserving all token embeddings, has been widely adopted. Our contribution is to reduce the size of the multi-vector representation, without compromising the effectiveness, supervised by query logs. Our proposed solution decreases the latency and the memory footprint, up to 8- and 3-fold, validated on MSMARCO and real-world search query logs.- Anthology ID:
- 2022.emnlp-main.800
- Volume:
- Proceedings of the 2022 Conference on Empirical Methods in Natural Language Processing
- Month:
- December
- Year:
- 2022
- Address:
- Abu Dhabi, United Arab Emirates
- Venue:
- EMNLP
- SIG:
- Publisher:
- Association for Computational Linguistics
- Note:
- Pages:
- 11639–11652
- Language:
- URL:
- https://aclanthology.org/2022.emnlp-main.800
- DOI:
- Cite (ACL):
- Jihyuk Kim, Seung-won Hwang, Seoho Song, Hyeseon Ko, and Young-In Song. 2022. Pseudo-Relevance for Enhancing Document Representation. In Proceedings of the 2022 Conference on Empirical Methods in Natural Language Processing, pages 11639–11652, Abu Dhabi, United Arab Emirates. Association for Computational Linguistics.
- Cite (Informal):
- Pseudo-Relevance for Enhancing Document Representation (Kim et al., EMNLP 2022)
- PDF:
- https://preview.aclanthology.org/ingestion-script-update/2022.emnlp-main.800.pdf