让大模型能持续更新知识而不跑偏，一次改千条仍保持性能。
O-Edit: Orthogonal Subspace Editing for Language Model Sequential Editing
- 用正交空间隔离每次知识更新方向，减少干扰
- 支持上千次编辑，性能比现有方法高4.2倍
- 无需重训，小额外参数即可实现长期维护
大型语言模型在预训练中获取知识，但随时间推移可能过时或错误，需后期更新。现有知识编辑技术多仅支持单次编辑，多次编辑后易导致整体性能下降，难以满足连续更新需求。为此，我们提出正交子空间编辑（O-Edit），通过正交化每次知识更新方向，最小化连续编辑间的干扰，并降低新编辑对无关知识的影响。该方法无需重放历史编辑数据，可逐次处理知识更新。可在主流大模型上执行数千次编辑，平均性能提升达现有方法的4.2倍，同时有效保持下游任务表现，且额外参数开销极小。
原文摘要 · Abstract (English)
Large language models (LLMs) acquire knowledge during pre-training, but over time, this knowledge may become incorrect or outdated, necessitating updates after training. Knowledge editing techniques address this issue without the need for costly re-training. However, most existing methods are designed for single edits, and as the number of edits increases, they often cause a decline in the model's overall performance, posing significant challenges for sequential editing. To overcome this, we propose Orthogonal Subspace Editing, O-Edit. This algorithm orthogonalizes the direction of each knowledge update, minimizing interference between successive updates and reducing the impact of new updates on unrelated knowledge. Our approach does not require replaying previously edited data and processes each edit knowledge on time. It can perform thousands of edits on mainstream LLMs, achieving an average performance improvement that is 4.2 times better than existing methods while effectively preserving the model's performance on downstream tasks, all with minimal additional parameter overhead.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。