The University of Arizona at SemEval-2021 Task 10: Applying Self-training, Active Learning and Data Augmentation to Source-free Domain Adaptation

Xin Su, Yiyun Zhao, Steven Bethard

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

5 Scopus citations

Abstract

This paper describes our systems for negation detection and time expression recognition in SemEval 2021 Task 10, Source-Free Domain Adaptation for Semantic Processing. We show that self-training, active learning and data augmentation techniques can improve the generalization ability of the model on the unlabeled target domain data without accessing source domain data. We also perform detailed ablation studies and error analyses for our time expression recognition systems to identify the source of the performance improvement and give constructive feedback on the temporal normalization annotation guidelines.

Original languageEnglish (US)
Title of host publicationSemEval 2021 - 15th International Workshop on Semantic Evaluation, Proceedings of the Workshop
EditorsAlexis Palmer, Nathan Schneider, Natalie Schluter, Guy Emerson, Aurelie Herbelot, Xiaodan Zhu
PublisherAssociation for Computational Linguistics (ACL)
Pages458-466
Number of pages9
ISBN (Electronic)9781954085701
DOIs
StatePublished - 2021
Event15th International Workshop on Semantic Evaluation, SemEval 2021 - Virtual, Bangkok, Thailand
Duration: Aug 5 2021Aug 6 2021

Publication series

NameSemEval 2021 - 15th International Workshop on Semantic Evaluation, Proceedings of the Workshop

Conference

Conference15th International Workshop on Semantic Evaluation, SemEval 2021
Country/TerritoryThailand
CityVirtual, Bangkok
Period8/5/218/6/21

ASJC Scopus subject areas

  • Computational Theory and Mathematics
  • Computer Science Applications
  • Theoretical Computer Science

Fingerprint

Dive into the research topics of 'The University of Arizona at SemEval-2021 Task 10: Applying Self-training, Active Learning and Data Augmentation to Source-free Domain Adaptation'. Together they form a unique fingerprint.

Cite this