@inproceedings{haturusinghe-etal-2025-subasa,
    title = "Subasa - Adapting Language Models for Low-resourced Offensive Language Detection in {S}inhala",
    author = "Haturusinghe, Shanilka  and
      Weerasooriya, Tharindu Cyril  and
      Homan, Christopher M  and
      Zampieri, Marcos  and
      Liyanage, Sidath Ravindra",
    editor = "Ebrahimi, Abteen  and
      Haider, Samar  and
      Liu, Emmy  and
      Haider, Sammar  and
      Leonor Pacheco, Maria  and
      Wein, Shira",
    booktitle = "Proceedings of the 2025 Conference of the Nations of the Americas Chapter of the Association for Computational Linguistics: Human Language Technologies (Volume 4: Student Research Workshop)",
    month = apr,
    year = "2025",
    address = "Albuquerque, USA",
    publisher = "Association for Computational Linguistics",
    url = "https://preview.aclanthology.org/fix-sig-urls/2025.naacl-srw.26/",
    pages = "260--270",
    ISBN = "979-8-89176-192-6",
    abstract = "Accurate detection of offensive language is essential for a number of applications related to social media safety. There is a sharp contrast in performance in this task between low and high-resource languages. In this paper, we adapt fine-tuning strategies that have not been previously explored for Sinhala in the downstream task of offensive language detection. Using this approach, we introduce four models: ``Subasa-XLM-R'', which incorporates an intermediate Pre-Finetuning step using Masked Rationale Prediction. Two variants of ``Subasa-Llama'' and ``Subasa-Mistral'', are fine-tuned versions of Llama (3.2) and Mistral (v0.3), respectively, with a task-specific strategy. We evaluate our models on the SOLD benchmark dataset for Sinhala offensive language detection. All our models outperform existing baselines. Subasa-XLM-R achieves the highest Macro F1 score (0.84) surpassing state-of-the-art large language models like GPT-4o when evaluated on the same SOLD benchmark dataset under zero-shot settings. The models and code are publicly available."
}