@inproceedings{faisal-etal-2025-dialectal,
    title = "Dialectal Toxicity Detection: Evaluating {LLM}-as-a-Judge Consistency Across Language Varieties",
    author = "Faisal, Fahim  and
      Rahman, Md Mushfiqur  and
      Anastasopoulos, Antonios",
    editor = "Christodoulopoulos, Christos  and
      Chakraborty, Tanmoy  and
      Rose, Carolyn  and
      Peng, Violet",
    booktitle = "Findings of the Association for Computational Linguistics: EMNLP 2025",
    month = nov,
    year = "2025",
    address = "Suzhou, China",
    publisher = "Association for Computational Linguistics",
    url = "https://preview.aclanthology.org/author-page-yu-wang-polytechnic/2025.findings-emnlp.664/",
    doi = "10.18653/v1/2025.findings-emnlp.664",
    pages = "12429--12452",
    ISBN = "979-8-89176-335-7",
    abstract = "There has been little systematic study on how dialectal differences affect toxicity detection by modern LLMs. Furthermore, although using LLMs as evaluators ({``}LLM-as-a-judge'') is a growing research area, their sensitivity to dialectal nuances is still underexplored and requires more focused attention. In this paper, we address these gaps through a comprehensive toxicity evaluation of LLMs across diverse dialects. We create a multi-dialect dataset through synthetic transformations and human-assisted translations, covering 10 language clusters and 60 varieties. We then evaluate five LLMs on their ability to assess toxicity, measuring multilingual, dialectal, and LLM-human consistency. Our findings show that LLMs are sensitive to both dialectal shifts and low-resource multilingual variation, though the most persistent challenge remains aligning their predictions with human judgments."
}