@inproceedings{li-etal-2026-toxitrace,
    title = "{T}oxi{T}race: Gradient-Aligned Training for Explainable {C}hinese Toxicity Detection",
    author = "Li, Boyang  and
      Shou, Hongzhe  and
      Liang, Yuanyuan  and
      Zhang, JingBin  and
      Zhou, Fang",
    editor = "Liakata, Maria  and
      Moreira, Viviane P.  and
      Zhang, Jiajun  and
      Jurgens, David",
    booktitle = "Findings of the {A}ssociation for {C}omputational {L}inguistics: {ACL} 2026",
    month = jul,
    year = "2026",
    address = "San Diego, California, United States",
    publisher = "Association for Computational Linguistics",
    url = "https://preview.aclanthology.org/ingest-acl/2026.findings-acl.354/",
    pages = "7121--7138",
    ISBN = "979-8-89176-395-1",
    abstract = "Existing Chinese toxic content detection methods mainly target sentence-level classification but often fail to provide readable and contiguous toxic evidence spans. We propose ToxiTrace, an explainability-oriented method for BERT-style encoders with three components: (1) CuSA, which refines encoder-derived saliency cues into fine-grained toxic spans with lightweight LLM guidance; (2) GCLoss, a gradient-constrained objective that concentrates token-level saliency on toxic evidence while suppressing irrelevant activations; and (3) ARCL, which constructs sample-specific contrastive reasoning pairs to sharpen the semantic boundary between toxic and non-toxic content. Experiments show that ToxiTrace improves classification accuracy and toxic span extraction while preserving efficient encoder-based inference and producing more coherent, human-readable explanations. The core training code is available at https://github.com/ZhouF-ECNU/ToxiTrace."
}