From One to Many: Expanding the Scope of Toxicity Mitigation in Language Models

Beyza Ermis, Luiza Pozzobon, Sara Hooker, Patrick Lewis

Abstract
To date, toxicity mitigation in language models has almost entirely been focused on single-language settings. As language models embrace multilingual capabilities, it’s crucial our safety measures keep pace. Recognizing this research gap, our approach expands the scope of conventional toxicity mitigation to address the complexities presented by multiple languages. In the absence of sufficient annotated datasets across languages, we employ translated data to evaluate and enhance our mitigation techniques. We also compare finetuning mitigation approaches against retrieval-augmented techniques under both static and continual toxicity mitigation scenarios. This allows us to examine the effects of translation quality and the cross-lingual transfer on toxicity mitigation. We also explore how model size and data quantity affect the success of these mitigation efforts. Covering nine languages, our study represents a broad array of linguistic families and levels of resource availability, ranging from high to mid-resource languages. Through comprehensive experiments, we provide insights into the complexities of multilingual toxicity mitigation, offering valuable insights and paving the way for future research in this increasingly important field.
Anthology ID:
2024.findings-acl.893
Volume:
Findings of the Association for Computational Linguistics: ACL 2024
Month:
August
Year:
2024
Address:
Bangkok, Thailand
Editors:
Lun-Wei Ku, Andre Martins, Vivek Srikumar
Venue:
Findings
SIG:
Publisher:
Association for Computational Linguistics
Note:
Pages:
15041–15058
Language:
URL:
https://aclanthology.org/2024.findings-acl.893
DOI:
10.18653/v1/2024.findings-acl.893
Bibkey:
Cite (ACL):
Beyza Ermis, Luiza Pozzobon, Sara Hooker, and Patrick Lewis. 2024. From One to Many: Expanding the Scope of Toxicity Mitigation in Language Models. In Findings of the Association for Computational Linguistics: ACL 2024, pages 15041–15058, Bangkok, Thailand. Association for Computational Linguistics.
Cite (Informal):
From One to Many: Expanding the Scope of Toxicity Mitigation in Language Models (Ermis et al., Findings 2024)
Copy Citation:
PDF:
https://preview.aclanthology.org/autopr/2024.findings-acl.893.pdf
PDF Search Fix metadata