Language Detoxification with Attribute-Discriminative Latent Space

Jin Myung Kwak, Minseon Kim, Sung Ju Hwang


Abstract
Transformer-based Language Models (LMs) have achieved impressive results on natural language understanding tasks, but they can also generate toxic text such as insults, threats, and profanity, limiting their real-world applications. To overcome this issue, a few text generation approaches aim to detoxify toxic texts using additional LMs or perturbations. However, previous methods require excessive memory, computations, and time which are serious bottlenecks in their real-world application. To address such limitations, we propose an effective yet efficient method for language detoxification using an attribute-discriminative latent space. Specifically, we project the latent space of an original Transformer LM onto a discriminative latent space that well-separates texts by their attributes using a projection block and an attribute discriminator. This allows the LM to control the text generation to be non-toxic with minimal memory and computation overhead. We validate our model, Attribute-Discriminative Language Model (ADLM) on detoxified language and dialogue generation tasks, on which our method significantly outperforms baselines both in performance and efficiency.
Anthology ID:
2023.acl-long.565
Volume:
Proceedings of the 61st Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers)
Month:
July
Year:
2023
Address:
Toronto, Canada
Editors:
Anna Rogers, Jordan Boyd-Graber, Naoaki Okazaki
Venue:
ACL
SIG:
Publisher:
Association for Computational Linguistics
Note:
Pages:
10149–10171
Language:
URL:
https://aclanthology.org/2023.acl-long.565
DOI:
10.18653/v1/2023.acl-long.565
Bibkey:
Cite (ACL):
Jin Myung Kwak, Minseon Kim, and Sung Ju Hwang. 2023. Language Detoxification with Attribute-Discriminative Latent Space. In Proceedings of the 61st Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers), pages 10149–10171, Toronto, Canada. Association for Computational Linguistics.
Cite (Informal):
Language Detoxification with Attribute-Discriminative Latent Space (Kwak et al., ACL 2023)
Copy Citation:
PDF:
https://aclanthology.org/2023.acl-long.565.pdf
Video:
 https://aclanthology.org/2023.acl-long.565.mp4