发表机构
Korea University（高丽大学）
机构由 AI 辅助整理，请以论文原文为准。AI 中文总结
针对终身模型编辑中知识泛化下降问题，提出结合偏好优化、回放编辑与梯度约束的GLIME方法，在保持编辑性能的同时显著提升知识泛化能力。
AI 中文摘要
知识编辑技术能够在不进行完整重训练的情况下，快速更新大型语言模型（LLM）中的特定事实知识。然而，更现实的场景需要一种终身学习框架来处理持续更新，而非一次性修改。在此类设置中，现有编辑方法常常过度拟合目标提示，显著损害编辑知识的泛化能力以及模型的通用能力。为解决这一问题，我们提出GLIME（可泛化的终身模型编辑），该方法将知识编辑与基于生成行为的偏好优化相结合。GLIME还融入了基于回放的编辑和梯度约束，以保留先前编辑过的知识。实验结果表明，GLIME在终身编辑设置中显著提升了知识泛化能力，同时保持了编辑性能和通用能力。
英文摘要
Knowledge editing enables rapid updates of specific factual knowledge in large language models (LLMs) without full retraining. However, more realistic scenarios call for a lifelong framework that handles continual updates rather than one-off modifications. In such settings, existing editing methods often overfit to target prompts, significantly degrading both the generalization of the edited knowledge and the model's general capabilities. To address this issue, we propose GLIME (Generalizable Lifelong Model Editing), which combines knowledge editing with preference optimization over generation behavior. GLIME further incorporates replay-based editing and a gradient constraint to preserve previously edited knowledge. Experimental results show that GLIME significantly improves knowledge generalization in lifelong editing settings while maintaining both editing performance and general capabilities.
CommentsAccepted to EMNLP 2026 Main Conference
