On Transferability of Bias Mitigation Effects in Language Model Fine-Tuning

Xisen Jin, Francesco Barbieri, Brendan Kennedy, Aida Mostafazadeh Davani, Leonardo Neves, Xiang Ren


Abstract
Fine-tuned language models have been shown to exhibit biases against protected groups in a host of modeling tasks such as text classification and coreference resolution. Previous works focus on detecting these biases, reducing bias in data representations, and using auxiliary training objectives to mitigate bias during fine-tuning. Although these techniques achieve bias reduction for the task and domain at hand, the effects of bias mitigation may not directly transfer to new tasks, requiring additional data collection and customized annotation of sensitive attributes, and re-evaluation of appropriate fairness metrics. We explore the feasibility and benefits of upstream bias mitigation (UBM) for reducing bias on downstream tasks, by first applying bias mitigation to an upstream model through fine-tuning and subsequently using it for downstream fine-tuning. We find, in extensive experiments across hate speech detection, toxicity detection and coreference resolution tasks over various bias factors, that the effects of UBM are indeed transferable to new downstream tasks or domains via fine-tuning, creating less biased downstream models than directly fine-tuning on the downstream task or transferring from a vanilla upstream model. Though challenges remain, we show that UBM promises more efficient and accessible bias mitigation in LM fine-tuning.
Anthology ID:
2021.naacl-main.296
Volume:
Proceedings of the 2021 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies
Month:
June
Year:
2021
Address:
Online
Editors:
Kristina Toutanova, Anna Rumshisky, Luke Zettlemoyer, Dilek Hakkani-Tur, Iz Beltagy, Steven Bethard, Ryan Cotterell, Tanmoy Chakraborty, Yichao Zhou
Venue:
NAACL
SIG:
Publisher:
Association for Computational Linguistics
Note:
Pages:
3770–3783
Language:
URL:
https://aclanthology.org/2021.naacl-main.296
DOI:
10.18653/v1/2021.naacl-main.296
Bibkey:
Cite (ACL):
Xisen Jin, Francesco Barbieri, Brendan Kennedy, Aida Mostafazadeh Davani, Leonardo Neves, and Xiang Ren. 2021. On Transferability of Bias Mitigation Effects in Language Model Fine-Tuning. In Proceedings of the 2021 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, pages 3770–3783, Online. Association for Computational Linguistics.
Cite (Informal):
On Transferability of Bias Mitigation Effects in Language Model Fine-Tuning (Jin et al., NAACL 2021)
Copy Citation:
PDF:
https://aclanthology.org/2021.naacl-main.296.pdf
Video:
 https://aclanthology.org/2021.naacl-main.296.mp4
Data
Hate Speech