发表机构
Graduate School of Science and Technology, University of Tsukuba; Center for Artificial Intelligence Research, Tsukuba Institute for Advanced Research, University of Tsukuba; Institute of Systems and Information Engineering, University of Tsukuba（筑波大学科学技术研究生院; 筑波大学筑波高级研究院人工智能研究中心; 筑波大学系统与信息工程研究院）
机构由 AI 辅助整理，请以论文原文为准。AI 中文总结
FedAlphaEdit是首个在单一零空间原则下对齐本地编辑与服务器端合并规则的协同知识编辑框架，可让无法共享原始编辑数据的机构共同维护近似集中式编辑效果的共享大语言模型。
AI 中文摘要
多个机构各自拥有私有知识编辑内容，希望在不共享原始编辑请求的情况下将其整合到单个大语言模型中。AlphaEdit等零空间约束编辑方法从数学上保证每次更新不会破坏无关知识，CollabEdit等协同框架可在不共享数据的情况下聚合多个客户端的编辑内容，将两者结合看似简单，但本文表明这种简单结合存在结构性缺陷并找出了原因。基于该分析，本文提出FedAlphaEdit，这是首个在单一零空间原则下对齐本地编辑与服务器端合并规则以保留现有知识的协同知识编辑框架。FedAlphaEdit基于零空间对齐合并，客户端共享投影统计量，服务器在单次理想化假设下可证明能恢复在一处编辑所有内容的结果。实验表明，该方法修复了缺陷，在两类架构上同时将编辑成功率和知识保留率提升至接近集中式编辑的水平，使医院、金融公司等无法共享原始编辑数据的机构能共同维护一个近似于在一处编辑所有事实的共享模型。
英文摘要
Multiple institutions may each hold their own private knowledge edits and wish to integrate them into a single large language model without sharing raw edit requests. Null-space-constrained editing methods such as AlphaEdit mathematically guarantee that each update leaves unrelated knowledge intact, while collaborative frameworks such as CollabEdit aggregate edits from multiple clients without data sharing. Combining the two appears trivial. However, we show that this naive combination fails structurally, and we identify its cause. Guided by this analysis, we propose FedAlphaEdit. To our knowledge, this is the first collaborative knowledge editing framework that aligns both local editing and the server-side merging rule under a single null-space principle for preserving existing knowledge. FedAlphaEdit builds on null-space-aligned merging, in which clients share projected statistics and the server provably recovers the result of editing everything in one place under a one-shot idealization. Empirically, the proposed method repairs the collapse and brings edit success and preservation simultaneously close to the level of centralized editing across two architecture families. FedAlphaEdit thus lets institutions that cannot share raw edit data, such as hospitals and financial firms, jointly maintain a shared model that closely approximates editing all facts in one place.
Comments12 pages, 2 figures, 8 tables. Includes appendices (proofs, implementation reconciliation, experimental details). Code: https://github.com/soutasuga/FedAlphaEdit
