解决大模型持续编辑时知识冲突导致的性能下降问题
On the Superimposed Noise Accumulation Problem in Sequential Knowledge Editing of Large Language Models
- 提出动态正交约束方法，减少多次编辑引发的知识冲突
- 在长期编辑下成功率提升16.8%，优于最强基线
- 适合需要频繁更新知识的大模型应用开发者
序列化知识编辑技术旨在以低成本持续更新大语言模型的知识，避免生成过时或错误信息。然而，现有方法在长期编辑后编辑成功率显著下降。通过理论分析与实验发现，随着编辑次数增加，模型输出逐渐偏离目标，导致成功率降低，我们称此为叠加噪声累积问题。进一步分析表明，该问题源于无关知识的错误激活及激活知识间的冲突。基于此，提出DeltaEdit方法，通过动态正交约束策略减少知识冲突。实验表明，DeltaEdit显著降低了叠加噪声，在长期编辑场景下相比最强基线提升了16.8%的编辑性能。
原文摘要 · Abstract (English)
Sequential knowledge editing techniques aim to continuously update knowledge in large language models at low cost, preventing models from generating outdated or incorrect information. However, existing sequential editing methods suffer from a significant decline in editing success rates after long-term editing. Through theoretical analysis and experiments, our findings reveal that as the number of edits increases, the model's output increasingly deviates from the desired target, leading to a drop in editing success rates. We refer to this issue as the superimposed noise accumulation problem. Our further analysis demonstrates that the problem is related to the erroneous activation of irrelevant knowledge and conflicts between activated knowledge. Based on this analysis, a method named DeltaEdit is proposed that reduces conflicts between knowledge through dynamic orthogonal constraint strategies. Experiments show that DeltaEdit significantly reduces superimposed noise, achieving a 16.8% improvement in editing performance over the strongest baseline.
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。