@inproceedings{cai-etal-2026-caption,
    title = "Caption-once, Frames-on-Demand: Visual-Need Routing for Budget-Aware Agentic Long Video Understanding",
    author = "Cai, Weitong  and
      Zhang, Hang  and
      Huang, Yukai  and
      Xie, Yiqiao  and
      Gao, Shan  and
      Deng, Jiankang  and
      Xu, Songcen  and
      Song, Jifei  and
      Zhang, Zhensong",
    editor = "Lee, Hung-Yi  and
      Dev, Sunipa  and
      Li, Jessy  and
      Elliott, Desmond",
    booktitle = "Proceedings of the 2026 Conference on Empirical Methods in Natural Language Processing",
    month = oct,
    year = "2026",
    address = "Budapest, Hungary",
    publisher = "Association for Computational Linguistics",
    url = "https://aclanthology.org/2026.emnlp-main.422/",
    pages = "9651--9676",
    ISBN = "979-8-89176-459-0"
}