Trade-Offs Between Fairness and Privacy in Language Modeling

Cleo Matzken, Steffen Eger, Ivan Habernal


Abstract
Protecting privacy in contemporary NLP models is gaining in importance. So does the need to mitigate social biases of such models. But can we have both at the same time? Existing research suggests that privacy preservation comes at the price of worsening biases in classification tasks. In this paper, we explore the extent to which this tradeoff really holds when we incorporate both privacy preservation and de-biasing techniques into training text generation models. How does improving the model along one dimension affect the other dimension as well as the utility of the model? We conduct an extensive set of experiments that include bias detection, privacy attacks, language modeling, and performance on downstream tasks.
Anthology ID:
2023.findings-acl.434
Volume:
Findings of the Association for Computational Linguistics: ACL 2023
Month:
July
Year:
2023
Address:
Toronto, Canada
Editors:
Anna Rogers, Jordan Boyd-Graber, Naoaki Okazaki
Venue:
Findings
SIG:
Publisher:
Association for Computational Linguistics
Note:
Pages:
6948–6969
Language:
URL:
https://aclanthology.org/2023.findings-acl.434
DOI:
10.18653/v1/2023.findings-acl.434
Bibkey:
Cite (ACL):
Cleo Matzken, Steffen Eger, and Ivan Habernal. 2023. Trade-Offs Between Fairness and Privacy in Language Modeling. In Findings of the Association for Computational Linguistics: ACL 2023, pages 6948–6969, Toronto, Canada. Association for Computational Linguistics.
Cite (Informal):
Trade-Offs Between Fairness and Privacy in Language Modeling (Matzken et al., Findings 2023)
Copy Citation:
PDF:
https://aclanthology.org/2023.findings-acl.434.pdf