Pseudo-Relevance Feedback for Multiple Representation Dense Retrieval

Wang, X., Macdonald, C. , Tonellotto, N. and Ounis, I. (2021) Pseudo-Relevance Feedback for Multiple Representation Dense Retrieval. In: 7th ACM SIGIR International Conference on the Theory of Information Retrieval (ICTIR 2021), 11 Jul 2021, pp. 297-306. ISBN 9781450386111 (doi: 10.1145/3471158.3472250)

[img] Text
244350.pdf - Accepted Version

736kB

Abstract

Pseudo-relevance feedback mechanisms, from Rocchio to the relevance models, have shown the usefulness of expanding and reweighting the users' initial queries using information occurring in an initial set of retrieved documents, known as the pseudo-relevant set. Recently, dense retrieval -- through the use of neural contextual language models such as BERT for analysing the documents' and queries' contents and computing their relevance scores -- has shown a promising performance on several information retrieval tasks still relying on the traditional inverted index for identifying documents relevant to a query. Two different dense retrieval families have emerged: the use of single embedded representations for each passage and query (e.g. using BERT's [CLS] token), or via multiple representations (e.g. using an embedding for each token of the query and document). In this work, we conduct the first study into the potential for multiple representation dense retrieval to be enhanced using pseudo-relevance feedback. In particular, based on the pseudo-relevant set of documents identified using a first-pass dense retrieval, we extract representative feedback embeddings (using KMeans clustering) -- while ensuring that these embeddings discriminate among passages (based on IDF) -- which are then added to the query representation. These additional feedback embeddings are shown to both enhance the effectiveness of a reranking as well as an additional dense retrieval operation. Indeed, experiments on the MSMARCO passage ranking dataset show that MAP can be improved by upto 26% on the TREC 2019 query set and 10% on the TREC 2020 query set by the application of our proposed ColBERT-PRF method on a ColBERT dense retrieval approach.

Item Type:Conference Proceedings
Additional Information:Nicola Tonellotto was partially supported by the Italian Ministry of Education and Research (MIUR) in the framework of the CrossLab project (Departments of Excellence). Xiao Wang acknowledges support by the China Scholarship Council (CSC) from the Ministry of Education of P.R. China. Craig Macdonald and Iadh Ounis acknowledge EPSRC grant EP/ R018634/1: Closed-Loop Data Science for Complex, Computationally- & Data-Intensive Analytics.
Status:Published
Refereed:Yes
Glasgow Author(s) Enlighten ID:Macdonald, Professor Craig and Ounis, Professor Iadh and Wang, Ms Xiao and Tonellotto, Dr Nicola
Authors: Wang, X., Macdonald, C., Tonellotto, N., and Ounis, I.
College/School:College of Science and Engineering > School of Computing Science
ISBN:9781450386111
Copyright Holders:Copyright © 2021 Association for Computing Machinery
First Published:First published in ICTIR '21: Proceedings of the 2021 ACM SIGIR International Conference on Theory of Information Retrieval
Publisher Policy:Reproduced in accordance with the publisher copyright policy
Related URLs:

University Staff: Request a correction | Enlighten Editors: Update this record

Project CodeAward NoProject NamePrincipal InvestigatorFunder's NameFunder RefLead Dept
300982Exploiting Closed-Loop Aspects in Computationally and Data Intensive AnalyticsRoderick Murray-SmithEngineering and Physical Sciences Research Council (EPSRC)EP/R018634/1Computing Science