@inproceedings{huang-etal-2025-longsafety,
    title = "{L}ong{S}afety: Enhance Safety for Long-Context {LLM}s",
    author = "Huang, Mianqiu  and
      Liu, Xiaoran  and
      Zhou, Shaojun  and
      Zhang, Mozhi  and
      Guo, Qipeng  and
      Li, Linyang  and
      Wang, Pengyu  and
      Gao, Yang  and
      Tan, Chenkun  and
      Li, Linlin  and
      Liu, Qun  and
      Zhou, Yaqian  and
      Qiu, Xipeng  and
      Huang, Xuanjing",
    editor = "Novikova, Jekaterina",
    booktitle = "Proceedings of the The First Workshop on LLM Security (LLMSEC)",
    month = aug,
    year = "2025",
    address = "Vienna, Austria",
    publisher = "Association for Computational Linguistics",
    url = "https://preview.aclanthology.org/corrections-2025-08/2025.llmsec-1.4/",
    pages = "26--47",
    ISBN = "979-8-89176-279-4",
    abstract = "Recent advancements in model architectures and length extrapolation techniques have significantly extended the context length of large language models (LLMs), paving the way for their application in increasingly complex tasks. However, despite the growing capabilities of long-context LLMs, the safety issues in long-context scenarios remain underexplored. While safety alignment in short context has been widely studied, the safety concerns of long-context LLMs have not been adequately addressed. In this work, we introduce {\$}{\{}textbf{\{}LongSafety{\}}{\}}{\$}, a comprehensive safety alignment dataset for long-context LLMs, containing 10 tasks and 17k samples, with an average length of 40.9k tokens. Our experiments demonstrate that training with LongSafety can enhance long-context safety performance while enhancing short-context safety and preserving general capabilities. Furthermore, we demonstrate that long-context safety does not equal long-context alignment with short-context safety data and LongSafety has generalizing capabilities in context length and long-context safety scenarios."
}