Robustifying multi-hop question answering through pseudo-evidentiality training

Kyungjae Lee, Seung Won Hwang, Sang Eun Han, Dohyeon Lee

Research output: Chapter in Book/Report/Conference proceedingConference contribution

6 Citations (Scopus)

Abstract

This paper studies the bias problem of multihop question answering models, of answering correctly without correct reasoning. One way to robustify these models is by supervising to not only answer right, but also with right reasoning chains. An existing direction is to annotate reasoning chains to train models, requiring expensive additional annotations. In contrast, we propose a new approach to learn evidentiality, deciding whether the answer prediction is supported by correct evidences, without such annotations. Instead, we compare counterfactual changes in answer confidence with and without evidence sentences, to generate “pseudo-evidentiality” annotations. We validate our proposed model on an original set and challenge set in HotpotQA, showing that our method is accurate and robust in multi-hop reasoning.

Original languageEnglish
Title of host publicationACL-IJCNLP 2021 - 59th Annual Meeting of the Association for Computational Linguistics and the 11th International Joint Conference on Natural Language Processing, Proceedings of the Conference
PublisherAssociation for Computational Linguistics (ACL)
Pages6110-6119
Number of pages10
ISBN (Electronic)9781954085527
Publication statusPublished - 2021
EventJoint Conference of the 59th Annual Meeting of the Association for Computational Linguistics and the 11th International Joint Conference on Natural Language Processing, ACL-IJCNLP 2021 - Virtual, Online
Duration: 2021 Aug 12021 Aug 6

Publication series

NameACL-IJCNLP 2021 - 59th Annual Meeting of the Association for Computational Linguistics and the 11th International Joint Conference on Natural Language Processing, Proceedings of the Conference

Conference

ConferenceJoint Conference of the 59th Annual Meeting of the Association for Computational Linguistics and the 11th International Joint Conference on Natural Language Processing, ACL-IJCNLP 2021
CityVirtual, Online
Period21/8/121/8/6

Bibliographical note

Publisher Copyright:
© 2021 Association for Computational Linguistics

All Science Journal Classification (ASJC) codes

  • Software
  • Computational Theory and Mathematics
  • Linguistics and Language
  • Language and Linguistics

Fingerprint

Dive into the research topics of 'Robustifying multi-hop question answering through pseudo-evidentiality training'. Together they form a unique fingerprint.

Cite this