@inproceedings{liu-etal-2026-harve,
    title = "{HARVE}: Hacking-Aware Reward-Head Vector Editing for Robust Reward Models",
    author = "Liu, Shuang  and
      Bo, Yuxuan  and
      Zhao, Qiuyang  and
      Huang, Caiyue  and
      Chen, Xiaorong  and
      Liu, Yanguang  and
      Du, Mengnan",
    editor = "Lee, Hung-Yi  and
      Dev, Sunipa  and
      Li, Jessy  and
      Elliott, Desmond",
    booktitle = "Proceedings of the 2026 Conference on Empirical Methods in Natural Language Processing",
    month = oct,
    year = "2026",
    address = "Budapest, Hungary",
    publisher = "Association for Computational Linguistics",
    url = "https://aclanthology.org/2026.emnlp-main.1095/",
    pages = "25127--25150",
    ISBN = "979-8-89176-459-0"
}