多条关联知识编辑时，现有方法会互相干扰，需重新设计编辑策略。
Related Knowledge Perturbation Matters: Rethinking Multiple Pieces of Knowledge Editing in Same-Subject
- 发现同一实体的多条知识编辑会相互干扰，传统方法难以应对。
- 仅定位-编辑类方法（如ROME、MEMIT）存在相关知识扰动现象。
- 适合研究大模型知识更新机制或改进编辑方法的研究者阅读。
知识编辑已成为高效精准更新大语言模型中嵌入知识的有前景方法。本文聚焦于同一主体的知识编辑（Same-Subject Editing），即对单一实体的多个属性进行修改，以确保实体中心知识的全面与一致更新。初步观察发现：当前最先进的编辑方法在处理同一主体的多条相关知识时表现不佳。为解决传统基准中缺乏相同主体的编辑数据问题，我们提出了 $ ext{S}^2 ext{RKE}$（Same-Subject Related Knowledge Editing）基准。大量实验表明，只有主流的定位-编辑类方法（如 ROME、MEMIT）表现出“相关知识扰动”，即后续编辑会干扰先前编辑。进一步分析揭示，这些方法过度依赖主体信息，忽视其他关键因素，导致编辑效果下降。
原文摘要 · Abstract (English)
Knowledge editing has become a promising approach for efficiently and precisely updating knowledge embedded in large language models (LLMs). In this work, we focus on Same-Subject Editing, which involves modifying multiple attributes of a single entity to ensure comprehensive and consistent updates to entity-centric knowledge. Through preliminary observation, we identify a significant challenge: Current state-of-the-art editing methods struggle when tasked with editing multiple related knowledge pieces for the same subject. To address the lack of relevant editing data for identical subjects in traditional benchmarks, we introduce the $\text{S}^2\text{RKE}$(Same-Subject Related Knowledge Editing) benchmark. Our extensive experiments reveal that only mainstream locate-then-edit methods, such as ROME and MEMIT, exhibit "related knowledge perturbation," where subsequent edits interfere with earlier ones. Further analysis reveals that these methods over-rely on subject information, neglecting other critical factors, resulting in reduced editing effectiveness.
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。