@article{zhu-etal-2024-survey-model,
    title = "A Survey on Model Compression for Large Language Models",
    author = "Zhu, Xunyu  and
      Li, Jian  and
      Liu, Yong  and
      Ma, Can  and
      Wang, Weiping",
    journal = "Transactions of the Association for Computational Linguistics",
    volume = "12",
    year = "2024",
    address = "Cambridge, MA",
    publisher = "MIT Press",
    url = "https://preview.aclanthology.org/fix-sig-urls/2024.tacl-1.85/",
    doi = "10.1162/tacl_a_00704",
    pages = "1556--1577"
}