@inproceedings{zhang-etal-2026-rotation,
    title = "Rotation Control Unlearning: Quantifying and Controlling Continuous Unlearning for {LLM} with The Cognitive Rotation Space",
    author = "Zhang, Xiang  and
      Wei, Kun  and
      Yang, Xu  and
      Li, Jiahua  and
      Yan, Su  and
      Deng, Cheng",
    editor = "Liakata, Maria  and
      Moreira, Viviane P.  and
      Zhang, Jiajun  and
      Jurgens, David",
    booktitle = "Findings of the {A}ssociation for {C}omputational {L}inguistics: {ACL} 2026",
    month = jul,
    year = "2026",
    address = "San Diego, California, United States",
    publisher = "Association for Computational Linguistics",
    url = "https://preview.aclanthology.org/ingest-acl/2026.findings-acl.921/",
    pages = "18492--18510",
    ISBN = "979-8-89176-395-1",
    abstract = "As Large Language Models (LLMs) become increasingly prevalent, their security vulnerabilities have already drawn attention.Machine unlearning is introduced to seek to mitigate these risks by removing the influence of undesirable data. However, existing methods not only rely on the retained dataset to preserve model utility, but also suffer from cumulative catastrophic utility loss under continuous unlearning requests.To solve this dilemma, we propose a novel method, called Rotation Control Unlearning (RCU), which leverages the rotational salience weight of RCU to quantify and control the unlearning degree in the continuous unlearning process.The skew symmetric loss is designed to construct the existence of the cognitive rotation space, where the changes of rotational angle can simulate the continuous unlearning process.Furthermore, we design an orthogonal rotation axes regularization to enforce mutually perpendicular rotation directions for continuous unlearning requests, effectively minimizing interference and addressing cumulative catastrophic utility loss.Experiments on multiple datasets confirm that our continuous unlearning method without retained dataset achieves SOTA performance."
}