CFL: Causally Fair Language Models Through Token-level Attribute Controlled Generation

Rahul Madhavan, Rishabh Garg, Kahini Wadhawan, Sameep Mehta


Abstract
We propose a method to control the attributes of Language Models (LMs) for the text generation task using Causal Average Treatment Effect (ATE) scores and counterfactual augmentation. We explore this method, in the context of LM detoxification, and propose the Causally Fair Language (CFL) architecture for detoxifying pre-trained LMs in a plug-and-play manner. Our architecture is based on a Structural Causal Model (SCM) that is mathematically transparent and computationally efficient as compared with many existing detoxification techniques. We also propose several new metrics that aim to better understand the behaviour of LMs in the context of toxic text generation. Further, we achieve state of the art performance for toxic degeneration, which are computed using Real Toxicity Prompts. Our experiments show that CFL achieves such a detoxification without much impact on the model perplexity. We also show that CFL mitigates the unintended bias problem through experiments on the BOLD dataset.
Anthology ID:
2023.findings-acl.720
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:
11344–11358
Language:
URL:
https://aclanthology.org/2023.findings-acl.720
DOI:
10.18653/v1/2023.findings-acl.720
Bibkey:
Cite (ACL):
Rahul Madhavan, Rishabh Garg, Kahini Wadhawan, and Sameep Mehta. 2023. CFL: Causally Fair Language Models Through Token-level Attribute Controlled Generation. In Findings of the Association for Computational Linguistics: ACL 2023, pages 11344–11358, Toronto, Canada. Association for Computational Linguistics.
Cite (Informal):
CFL: Causally Fair Language Models Through Token-level Attribute Controlled Generation (Madhavan et al., Findings 2023)
Copy Citation:
PDF:
https://aclanthology.org/2023.findings-acl.720.pdf
Video:
 https://aclanthology.org/2023.findings-acl.720.mp4