@inproceedings{li-etal-2026-decomposing,
    title = "Decomposing Visual Histories with Vision-Language Agents: Hierarchical Temporal Guidance for Compositional Image Generation",
    author = "Li, Lujundong  and
      Zhang, Mingxu  and
      Lai, Songning  and
      Yang, Jiayu  and
      Fan, Yuxuan  and
      Xie, Ziwei  and
      Liu, Feng  and
      Zhang, Changwang  and
      Wang, Jun  and
      Shen, Dazhong  and
      Sun, Ying",
    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.607/",
    pages = "13989--14000",
    ISBN = "979-8-89176-459-0"
}