@inproceedings{yu-etal-2025-text,
    title = "Text Detoxification: Data Efficiency, Semantic Preservation and Model Generalization",
    author = "Yu, Jing  and
      Zhao, Yibo  and
      Zhu, Jiapeng  and
      Shao, Wenming  and
      Pang, Bo  and
      Zhang, Zhao  and
      Li, Xiang",
    editor = "Christodoulopoulos, Christos  and
      Chakraborty, Tanmoy  and
      Rose, Carolyn  and
      Peng, Violet",
    booktitle = "Proceedings of the 2025 Conference on Empirical Methods in Natural Language Processing",
    month = nov,
    year = "2025",
    address = "Suzhou, China",
    publisher = "Association for Computational Linguistics",
    url = "https://preview.aclanthology.org/ingest-emnlp/2025.emnlp-main.1636/",
    pages = "32160--32174",
    ISBN = "979-8-89176-332-6",
    abstract = "The widespread dissemination of toxic content on social media poses a serious threat to both online environments and public discourse, highlighting the urgent need for detoxification methods that effectively remove toxicity while preserving the original semantics.However, existing approaches often struggle to simultaneously achieve strong detoxification performance, semantic preservation, and robustness to out-of-distribution data. Moreover, they typically rely on costly, manually annotated parallel corpora while showing poor data efficiency.To address these challenges, we propose GEM, a two-stage training framework that jointly optimizes Model Generalization, Data Efficiency, and Semantic Preservation.We first perform supervised fine-tuning on a small set of high-quality, filtered parallel data to establish a strong initialization. Then, we leverage unlabeled toxic inputs and a custom-designed reward model to train the LLM using Group Relative Policy Optimization.Experimental results demonstrate that our method effectively mitigates the trade-offs faced by previous work, achieving state-of-the-art performance with improved generalization and significantly reduced dependence on annotated data. Our code is available at https://github.com/allacnobug/Detoxification-of-Text."
}