@inproceedings{yang-etal-2025-exploring-multimodal,
    title = "Exploring Multimodal Challenges in Toxic {C}hinese Detection: Taxonomy, Benchmark, and Findings",
    author = "Yang, Shujian  and
      Cui, Shiyao  and
      Hu, Chuanrui  and
      Wang, Haicheng  and
      Zhang, Tianwei  and
      Huang, Minlie  and
      Lu, Jialiang  and
      Qiu, Han",
    editor = "Che, Wanxiang  and
      Nabende, Joyce  and
      Shutova, Ekaterina  and
      Pilehvar, Mohammad Taher",
    booktitle = "Findings of the Association for Computational Linguistics: ACL 2025",
    month = jul,
    year = "2025",
    address = "Vienna, Austria",
    publisher = "Association for Computational Linguistics",
    url = "https://preview.aclanthology.org/ingestion-acl-25/2025.findings-acl.742/",
    pages = "14382--14396",
    ISBN = "979-8-89176-256-5",
    abstract = "Detecting toxic content using language models is important but challenging. While large language models (LLMs) have demonstrated strong performance in understanding Chinese, recent studies show that simple character substitutions in toxic Chinese text can easily confuse the state-of-the-art (SOTA) LLMs. In this paper, we highlight the multimodal nature of Chinese language as a key challenge for deploying LLMs in toxic Chinese detection. First, we propose a taxonomy of 3 perturbation strategies and 8 specific approaches in toxic Chinese content. Then, we curate a dataset based on this taxonomy, and benchmark 9 SOTA LLMs (from both the US and China) to assess if they can detect perturbed toxic Chinese text. Additionally, we explore cost-effective enhancement solutions like in-context learning (ICL) and supervised fine-tuning (SFT). Our results reveal two important findings. (1) LLMs are less capable of detecting perturbed multimodal Chinese toxic contents. (2) ICL or SFT with a small number of perturbed examples may cause the LLMs ``overcorrect'': misidentify many normal Chinese contents as toxic."
}