@inproceedings{son-etal-2024-prefixing,
    title = "Prefixing Attention Sinks can Mitigate Activation Outliers for Large Language Model Quantization",
    author = "Son, Seungwoo  and
      Park, Wonpyo  and
      Han, Woohyun  and
      Kim, Kyuyeun  and
      Lee, Jaeho",
    editor = "Al-Onaizan, Yaser  and
      Bansal, Mohit  and
      Chen, Yun-Nung",
    booktitle = "Proceedings of the 2024 Conference on Empirical Methods in Natural Language Processing",
    month = nov,
    year = "2024",
    address = "Miami, Florida, USA",
    publisher = "Association for Computational Linguistics",
    url = "https://preview.aclanthology.org/landing_page/2024.emnlp-main.134/",
    doi = "10.18653/v1/2024.emnlp-main.134",
    pages = "2242--2252"
}