EMNLP 2024main0 citations

XDetox: Text Detoxification with Token-Level Toxicity Explanations

Beomseok Lee, Hyunwoo Kim, Keon Kim, Yong Suk Choi

Abstract

Methods for mitigating toxic content through masking and infilling often overlook the decision-making process, leading to either insufficient or excessive modifications of toxic tokens. To address this challenge, we propose XDetox, a novel method that integrates token-level toxicity explanations with the masking and infilling detoxification process. We utilized this approach with two strategies to enhance the performance of detoxification. First, identifying toxic tokens to improve the quality of masking. Second, selecting the regenerated sentence by re-ranking the least toxic sentence among candidates. Our experimental results show state-of-the-art performance across four datasets compared to existing detoxification methods. Furthermore, human evaluations indicate that our method outperforms baselines in both fluency and toxicity reduction. These results demonstrate the effectiveness of our method in text detoxification.

BibTeX
@inproceedings{lee-etal-2024-xdetox,
    title = "{XD}etox: Text Detoxification with Token-Level Toxicity Explanations",
    author = "Lee, Beomseok  and
      Kim, Hyunwoo  and
      Kim, Keon  and
      Choi, Yong Suk",
    editor = "Al-Onaizan, Yaser  and
      Bansal, Mohit  and
      Chen, Yun-Nung",
    booktitle = "Proceedings of the 2024 Conference on Empirical Methods in Natural Language Processing",
    month = nov,
    year = "2024",
    address = "Miami, Florida, USA",
    publisher = "Association for Computational Linguistics",
    url = "https://aclanthology.org/2024.emnlp-main.848/",
    doi = "10.18653/v1/2024.emnlp-main.848",
    pages = "15215--15226"
}
XDetox: Text Detoxification with Token-Level Toxicity Explanations · EMNLP 2024