A4 Refereed article in a conference publication
Explaining Classes through Stable Word Attributions
Authors: Rönnqvist Samuel, Myntti Amanda, Kyröläinen Aki-Juhani, Ginter Filip, Laippala Veronika
Editors: Smaranda Muresan, Preslav Nakov, Aline Villavicencio
Conference name: Annual Meeting of the Association for Computational Linguistics
Publication year: 2022
Journal: Annual Meeting of the Association for Computational Linguistics
Book title : The 60th Annual Meeting of the Association for Computational Linguistics: Findings of ACL 2022
Journal name in source: FINDINGS OF THE ASSOCIATION FOR COMPUTATIONAL LINGUISTICS (ACL 2022)
Series title: Annual Meeting of the Association for Computational Linguistics
Volume: 60
First page : 1063
Last page: 1074
Number of pages: 12
ISBN: 978-1-955917-25-4
DOI: https://doi.org/10.18653/v1/2022.findings-acl.85
Web address : https://aclanthology.org/2022.findings-acl.85
Self-archived copy’s web address: https://research.utu.fi/converis/portal/detail/Publication/176874206
Input saliency methods have recently become a popular tool for explaining predictions of deep learning models in NLP. Nevertheless, there has been little work investigating methods for aggregating prediction-level explanations to the class level, nor has a framework for evaluating such class explanations been established. We explore explanations based on XLM-R and the Integrated Gradients input attribution method, and propose 1) the Stable Attribution Class Explanation method (SACX) to extract keyword lists of classes in text classification tasks, and 2) a framework for the systematic evaluation of the keyword lists. We find that explanations of individual predictions are prone to noise, but that stable explanations can be effectively identified through repeated training and explanation. We evaluate on web register data and show that the class explanations are linguistically meaningful and distinguishing of the classes.
Downloadable publication This is an electronic reprint of the original article. |  |
