@inproceedings{zhen-etal-2025-taming,
    title = "Taming the Titans: A Survey of Efficient {LLM} Inference Serving",
    author = "Zhen, Ranran  and
      Li, Juntao  and
      Ji, Yixin  and
      Yang, Zhenlin  and
      Liu, Tong  and
      Xia, Qingrong  and
      Duan, Xinyu  and
      Wang, Zhefeng  and
      Huai, Baoxing  and
      Zhang, Min",
    editor = "Flek, Lucie  and
      Narayan, Shashi  and
      Phương, L{\^e} Hồng  and
      Pei, Jiahuan",
    booktitle = "Proceedings of the 18th International Natural Language Generation Conference",
    month = oct,
    year = "2025",
    address = "Hanoi, Vietnam",
    publisher = "Association for Computational Linguistics",
    url = "https://preview.aclanthology.org/ingest-luhme/2025.inlg-main.32/",
    pages = "522--541"
}