@inproceedings{kim-etal-2025-tom,
    title = "{WHEN} {TOM} {EATS} {KIMCHI}: Evaluating Cultural Awareness of Multimodal Large Language Models in Cultural Mixture Contexts",
    author = "Kim, Jun Seong  and
      Thu, Kyaw Ye  and
      Ismayilzada, Javad  and
      Park, Junyeong  and
      Kim, Eunsu  and
      Ahmad, Huzama  and
      An, Na Min  and
      Thorne, James  and
      Oh, Alice",
    editor = "Prabhakaran, Vinodkumar  and
      Dev, Sunipa  and
      Benotti, Luciana  and
      Hershcovich, Daniel  and
      Cao, Yong  and
      Zhou, Li  and
      Cabello, Laura  and
      Adebara, Ife",
    booktitle = "Proceedings of the 3rd Workshop on Cross-Cultural Considerations in NLP (C3NLP 2025)",
    month = may,
    year = "2025",
    address = "Albuquerque, New Mexico",
    publisher = "Association for Computational Linguistics",
    url = "https://preview.aclanthology.org/fix-sig-urls/2025.c3nlp-1.11/",
    pages = "143--154",
    ISBN = "979-8-89176-237-4",
    abstract = "In a highly globalized world, it is important for multi-modal large language models (MLLMs) to recognize and respond correctly to mixed-cultural inputs.For example, a model should correctly identify kimchi (Korean food) in an image both when an Asian woman is eating it, as well as an African man is eating it.However, current MLLMs show an over-reliance on the visual features of the person, leading to misclassification of the entities. To examine the robustness of MLLMs to different ethnicity, we introduce MIXCUBE, a cross-cultural bias benchmark, and study elements from five countries and four ethnicities. Our findings reveal that MLLMs achieve both higher accuracy and lower sensitivity to such perturbation for high-resource cultures, but not for low-resource cultures. GPT-4o, the best-performing model overall, shows up to 58{\%} difference in accuracy between the original and perturbed cultural settings in low-resource cultures"
}