@inproceedings{zhang-etal-2026-spectral,
    title = "Spectral Characterization and Mitigation of Sequential Knowledge Editing Collapse",
    author = "Zhang, Chi  and
      Zhang, Mengqi  and
      Ye, Xiaotian  and
      Cheng, Runxi  and
      Zhou, Zisheng  and
      Zhou, Ying  and
      Ren, Pengjie  and
      Chen, Zhumin",
    editor = "Liakata, Maria  and
      Moreira, Viviane P.  and
      Zhang, Jiajun  and
      Jurgens, David",
    booktitle = "Proceedings of the 64th Annual Meeting of the {A}ssociation for {C}omputational {L}inguistics (Volume 1: Long Papers)",
    month = jul,
    year = "2026",
    address = "San Diego, California, United States",
    publisher = "Association for Computational Linguistics",
    url = "https://preview.aclanthology.org/ingest-acl/2026.acl-long.1384/",
    pages = "30009--30032",
    ISBN = "979-8-89176-390-6",
    abstract = "Sequential knowledge editing in large language models often causes catastrophic collapse of the model{'}s general abilities, especially for parameter-modifying methods. Existing approaches mitigate this issue through heuristic constraints on parameter updates, the mechanisms underlying such degradation remain insufficiently understood. In this work, we present a systematic spectral analysis of sequential knowledge editing and show that a model{'}s general abilities are closely associated with dominant singular directions of pretrained weight matrices. These directions are highly sensitive to perturbations and are progressively disrupted by repeated edits, closely tracking the collapse in both editing efficacy and general performance. Building on this insight, we propose REVIVE, a plug-and-play framework that prevents model collapse by explicitly preserving this dominant subspace. REVIVE analyzes parameter updates in the spectral basis of the original weights and filters out components that would interfere with the dominant subspace. Extensive experiments across multiple models and benchmarks show that REVIVE consistently improves editing efficacy while substantially preserving general abilities under long-horizon sequential editing, including extreme settings with up to 20,000 edits."
}