TagDebias: Entity and Concept Tagging for Social Bias Mitigation in Pretrained Language Models

Mehrnaz Moslemi, Amal Zouaq


Abstract
Pre-trained language models (PLMs) play a crucial role in various applications, including sensitive domains such as the hiring process. However, extensive research has unveiled that these models tend to replicate social biases present in their pre-training data, raising ethical concerns. In this study, we propose the TagDebias method, which proposes debiasing a dataset using type tags. It then proceeds to fine-tune PLMs on this debiased dataset. Experiments show that our proposed TagDebias model, when applied to a ranking task, exhibits significant improvements in bias scores.
Anthology ID:
2024.findings-naacl.101
Volume:
Findings of the Association for Computational Linguistics: NAACL 2024
Month:
June
Year:
2024
Address:
Mexico City, Mexico
Editors:
Kevin Duh, Helena Gomez, Steven Bethard
Venue:
Findings
SIG:
Publisher:
Association for Computational Linguistics
Note:
Pages:
1553–1567
Language:
URL:
https://aclanthology.org/2024.findings-naacl.101
DOI:
Bibkey:
Cite (ACL):
Mehrnaz Moslemi and Amal Zouaq. 2024. TagDebias: Entity and Concept Tagging for Social Bias Mitigation in Pretrained Language Models. In Findings of the Association for Computational Linguistics: NAACL 2024, pages 1553–1567, Mexico City, Mexico. Association for Computational Linguistics.
Cite (Informal):
TagDebias: Entity and Concept Tagging for Social Bias Mitigation in Pretrained Language Models (Moslemi & Zouaq, Findings 2024)
Copy Citation:
PDF:
https://preview.aclanthology.org/naacl24-info/2024.findings-naacl.101.pdf
Copyright:
 2024.findings-naacl.101.copyright.pdf