@inproceedings{he-etal-2025-dont,
    title = "Don{'}t Half-listen: Capturing Key-part Information in Continual Instruction Tuning",
    author = "He, Yongquan  and
      Zhang, Wenyuan  and
      Huang, Xuancheng  and
      Zhang, Peng  and
      Meng, Lingxun  and
      Zhou, Xiang  and
      Zeng, Ke  and
      Cai, Xunliang",
    editor = "Che, Wanxiang  and
      Nabende, Joyce  and
      Shutova, Ekaterina  and
      Pilehvar, Mohammad Taher",
    booktitle = "Proceedings of the 63rd Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers)",
    month = jul,
    year = "2025",
    address = "Vienna, Austria",
    publisher = "Association for Computational Linguistics",
    url = "https://preview.aclanthology.org/landing_page/2025.acl-long.1153/",
    pages = "23649--23668",
    ISBN = "979-8-89176-251-0",
    abstract = "Instruction tuning for large language models (LLMs) can drive them to produce results consistent with human goals in specific downstream tasks. However, the process of continual instruction tuning (CIT) for LLMs may bring about the catastrophic forgetting (CF) problem, where previously learned abilities are degraded. Recent methods try to alleviate the CF problem by modifying models or replaying data, which may only remember the surface-level pattern of instructions and get confused on held-out tasks. In this paper, we propose a novel continual instruction tuning method based on Key-part Information Gain (KPIG). Our method computes the information gain on masked parts to dynamically replay data and refine the training objective, which enables LLMs to capture task-aware information relevant to the correct response and alleviate overfitting to general descriptions in instructions. In addition, we propose two metrics, P-score and V-score, to measure the generalization and instruction-following abilities of LLMs. Experiments demonstrate our method achieves superior performance on both seen and held-out tasks."
}