@inproceedings{nguyen-etal-2026-beyond-caption,
    title = "Beyond Caption Loss: Generative Distillation for Modality Alignment in Vision-Language Models",
    author = "Nguyen, Nam Kha  and
      Zhang, Wei Emma  and
      Dai, Xiang  and
      Sim, Mong Yuan  and
      Chen, Weitong",
    editor = "Lee, Hung-Yi  and
      Dev, Sunipa  and
      Li, Jessy  and
      Elliott, Desmond",
    booktitle = "Findings of the {A}ssociation for {C}omputational {L}inguistics: {EMNLP} 2026",
    month = oct,
    year = "2026",
    address = "Budapest, Hungary",
    publisher = "Association for Computational Linguistics",
    url = "https://aclanthology.org/2026.findings-emnlp.1876/",
    pages = "40021--40034",
    ISBN = "979-8-89176-462-0"
}