← smac.pub home

Doc2Query--: When Less is More

Best Short Paper pdf bibtex code poster demo short conference paper

Authors: Mitko Gospodinov, Sean MacAvaney, Craig Macdonald

Appeared in: Proceedings of the 45th European Conference on Information Retrieval Research (ECIR 2023)

DOI 10.1007/978-3-031-28238-6_31 DBLP conf/ecir/GospodinovMM23 arXiv 2301.03266 Google Scholar 7wWfoDgAAAAJ:k_IJM867U9cC Semantic Scholar 7f67b84eb13dc6397800a62c4d436c62effd210c smac.pub ecir2023-doc2querymm


Doc2Query --- the process of expanding the content of a document before indexing using a sequence-to-sequence model --- has emerged as a prominent technique for improving the first-stage retrieval effectiveness of search engines. However, sequence-to-sequence models are known to be prone to "hallucinating" content that is not present in the source text. We find that Doc2Query is indeed prone to hallucination, which ultimately harms retrieval effectiveness and inflates the index size. In this work, we explore techniques for filtering out these harmful queries prior to indexing. We find that using a relevance model to remove poor-performing queries can improve the retrieval effectiveness of Doc2Query by up to 16%, while simultaneously reducing mean query execution time by 23% and cutting the index size by 33%.

BibTeX @inproceedings{gospodinov:ecir2023-doc2querymm, author = {Gospodinov, Mitko and MacAvaney, Sean and Macdonald, Craig}, title = {Doc2Query--: When Less is More}, booktitle = {Proceedings of the 45th European Conference on Information Retrieval Research}, year = {2023}, url = {https://arxiv.org/abs/2301.03266}, doi = {10.1007/978-3-031-28238-6_31} }