@inproceedings{aly-etal-2025-argan,
    title = "{A}r{GAN}: {A}rabic Gender, Ability, and Nationality Dataset for Evaluating Biases in Large Language Models",
    author = "Aly, Ranwa  and
      Allam, Yara  and
      Gaber, Rana  and
      Basta, Christine",
    editor = "Fale{\'n}ska, Agnieszka  and
      Basta, Christine  and
      Costa-juss{\`a}, Marta  and
      Sta{\'n}czak, Karolina  and
      Nozza, Debora",
    booktitle = "Proceedings of the 6th Workshop on Gender Bias in Natural Language Processing (GeBNLP)",
    month = aug,
    year = "2025",
    address = "Vienna, Austria",
    publisher = "Association for Computational Linguistics",
    url = "https://preview.aclanthology.org/display_plenaries/2025.gebnlp-1.23/",
    pages = "256--267",
    ISBN = "979-8-89176-277-0",
    abstract = "Large language models (LLMs) are pretrained on substantial, unfiltered corpora, assembled from a variety of sources. This risks inheriting the deep-rooted biases that exist within them, both implicit and explicit. This is even more apparent in low-resource languages, where corpora may be prioritized by quantity over quality, potentially leading to more unchecked biases. More specifically, we address the biases present in the Arabic language in both general-purpose and Arabic-specialized architectures in three dimensions of demographics: gender, ability, and nationality. To properly assess the fairness of these models, we experiment with bias-revealing prompts and estimate the performance using existing evaluation metrics, and propose adaptations to others."
}