@inproceedings{mo-etal-2025-team,
    title = "Team Cantharellus at {S}em{E}val-2025 Task 3: Hallucination Span Detection with Fine Tuning on Weakly Supervised Synthetic Data",
    author = "Mo, Xinyuan  and
      Vorontsov, Nikolay  and
      Zang, Tiankai",
    editor = "Rosenthal, Sara  and
      Ros{\'a}, Aiala  and
      Ghosh, Debanjan  and
      Zampieri, Marcos",
    booktitle = "Proceedings of the 19th International Workshop on Semantic Evaluation (SemEval-2025)",
    month = jul,
    year = "2025",
    address = "Vienna, Austria",
    publisher = "Association for Computational Linguistics",
    url = "https://preview.aclanthology.org/corrections-2025-08/2025.semeval-1.226/",
    pages = "1724--1736",
    ISBN = "979-8-89176-273-2",
    abstract = "This paper describes our submission to SemEval-2025 Task-3: Mu-SHROOM, the Multilingual Shared-task on Hallucinations and Related Observable Overgeneration Mistakes, which mainly aims at detecting spans of LLM-generated text corresponding to hallucinations in multilingual and multi-model context. We explored an approach of fine-tuning pretrained language models available on Hugging Face. The results show that predictions made by a pretrained model fine-tuned on synthetic data achieve a relatively high degree of alignment with human-generated labels. We participated in 13 out of 14 available languages and reached an average ranking of 10th out of 41 participating teams, with our highest ranking reaching the top 5 place."
}