Importance-awareness masking network for robust document retrieval

Junping Liu, Jiaqi He, Xinrong Hu, Wangli Yang, Jie Yang, Yi Guo

Research output: Chapter in Book / Conference PaperConference Paperpeer-review

Abstract

In this paper, we introduce the IMPortance-awaReness maskIng NeTwork (IMPRINT), a novel approach to enhance the robustness of document retrieval systems against query variations, particularly those containing mis-spellings. Unlike previous models that treat all query components (words/features) equally, IMPRINT prioritizes the most important components while masking out less relevant ones. Specifically, we propose a Mutual Information-based measure to quantify component importance and integrate it into a dynamic masking mechanism that adjusts the retention probability of each component. Our method is evaluated on a combination of three benchmark datasets and three types of query variations. The experimental results show substantial performance gains compared to state-of-the-art models, achieving an average improvement of 1.2 absolute MRR@10 points in retrieval accuracy.
Original languageEnglish
Title of host publicationProceedings of 2025 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP 2025), April 6 - April 11, 2025, Hyderabad, India
EditorsBhaskar D Rao, Isabel Trancoso, Gaurav Sharma, Neelesh B. Mehta
Place of PublicationU.S.
PublisherIEEE
Number of pages5
ISBN (Electronic)9798350368741
ISBN (Print)9798350368741
DOIs
Publication statusPublished - 2025
EventICASSP (Conference) - Hyderabad, India
Duration: 6 Apr 202511 Apr 2025

Publication series

NameICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings
ISSN (Print)1520-6149

Conference

ConferenceICASSP (Conference)
Country/TerritoryIndia
CityHyderabad
Period6/04/2511/04/25

Bibliographical note

Publisher Copyright:
© 2025 IEEE.

Keywords

  • Document Retrieval
  • Model Robustness
  • Query Variations
  • Word/Feature Importance

Fingerprint

Dive into the research topics of 'Importance-awareness masking network for robust document retrieval'. Together they form a unique fingerprint.

Cite this