@inproceedings{wagner-etal-2024-optimized,
    title = "Optimized Speculative Sampling for {GPU} Hardware Accelerators",
    author = "Wagner, Dominik  and
      Lee, Seanie  and
      Baumann, Ilja  and
      Seeberger, Philipp  and
      Riedhammer, Korbinian  and
      Bocklet, Tobias",
    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/jlcl-multiple-ingestion/2024.emnlp-main.370/",
    doi = "10.18653/v1/2024.emnlp-main.370",
    pages = "6442--6458"
}