@inproceedings{xiong-etal-2026-uniattn,
    title = "{U}ni{A}ttn: Reducing Inference Costs via Softmax Unification for Post-Training {LLM}s",
    author = "Xiong, Yizhe  and
      Huang, Wei  and
      Ye, Xin  and
      Chen, Hui  and
      Lin, Zijia  and
      Lian, Haoran  and
      Su, Zhenpeng  and
      Han, Jungong  and
      Ding, Guiguang",
    editor = "Lee, Hung-Yi  and
      Dev, Sunipa  and
      Li, Jessy  and
      Elliott, Desmond",
    booktitle = "Findings of the {A}ssociation for {C}omputational {L}inguistics: {EMNLP} 2026",
    month = oct,
    year = "2026",
    address = "Budapest, Hungary",
    publisher = "Association for Computational Linguistics",
    url = "https://aclanthology.org/2026.findings-emnlp.2025/",
    pages = "43072--43091",
    ISBN = "979-8-89176-462-0"
}