@inproceedings{lu-etal-2026-one-size,
    title = "One Size Does Not Fit All: Token-Wise Adaptive Compression for {KV} Cache",
    author = "Lu, Liming  and
      Qiu, Kaixi  and
      Zhou, Jiayu  and
      Kai, Jushi  and
      Zhang, Haoyan  and
      Wang, Jingzhi  and
      Wang, Huanyu  and
      Jiang, Bo  and
      Leng, Jingwen  and
      He, Ziwei  and
      Lin, Zhouhan",
    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.524/",
    pages = "12100--12114",
    ISBN = "979-8-89176-459-0"
}