@inproceedings{li-etal-2026-self-evolving,
    title = "Self-Evolving Vision Language Models for Structured Visual Understanding",
    author = "Li, Zongxia  and
      Wu, Xiaomin  and
      Huang, Chengsong  and
      Du, Hongyang  and
      Wu, Xiyang  and
      Yu, Lantao  and
      He, Yicheng  and
      Liu, Zhichao  and
      Zhang, Jiarui  and
      Liu, Fuxiao  and
      Xie, Jing",
    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.1685/",
    pages = "38332--38347",
    ISBN = "979-8-89176-459-0"
}