Model editing has emerged as a practical approach for mitigating factual errors and outdated knowledge in large language models (LLMs). Among existing methods, the Locate-and-Edit (L&E) paradigm is the dominant framework: it locates MLP parameters implicated in expressing a target fact, and then performs a localized update to rewrite that fact. However, long sequences of edits often trigger abrupt model collapse in L&E beyond a critical point. We empirically identify a strong correlation between collapse and explosive growth of edited MLP weight norms, and formally prove that commonly used L&E update rules can induce exponential norm growth across sequential edits in the absence of explicit norm control. To address this issue, we propose Norm-Anchor Scaling NAS, a plug-and-play norm-constrained strategy. Across extensive experiments, NAS delays the collapse point of representative L&E algorithms by more than 4 times and yields a 72.2% average relative gain in editing performance, requiring only a single additional line of code and incurring negligible computational overhead.
Constraining Sequential Model Editing with Editing Anchor Compression
National Engineering Research Center of Speech and Language Information Processing, University of Science and Technology of China（中国科学技术大学语音及语言信息处理国家工程研究中心）；Zhejiang University（浙江大学）；University of California, Los Angeles（加利福尼亚大学洛杉矶分校）
The Labyrinth and the Thread: Rethinking Regularizations in Sequential Knowledge Editing for Large Language Models
Bosch Center for Artificial Intelligence (BCAI)（博世人工智能中心（BCAI））；Bosch (China) Investment Ltd.（博世（中国）投资有限公司）；School of Statistics, East China Normal University（东华大学统计学院）
Spectral Characterization and Mitigation of Sequential Knowledge Editing Collapse
Shandong University（山东大学）；School of Computer Science, Beijing University of Posts and Telecommunications（北京邮电大学计算机学院）；Tsinghua Shenzhen International Graduate School, Tsinghua University（清华大学深圳国际研究生院）
Virginia Tech（弗吉尼亚理工大学）；Duke University（杜克大学）；Adobe Research（奥多比研究院）；Waseda University（早稻田大学）；Meta AI；The Pennsylvania State University（宾夕法尼亚州立大学）
Memory Editing (ME) has emerged as an efficient method to modify erroneous facts or inject new facts into Large Language Models (LLMs). Two mainstream ME methods exist: parameter-modifying ME and parameter-preserving ME (integrating extra modules while preserving original parameters). Regrettably, previous studies on ME evaluation have two critical limitations: (i) evaluating LLMs with single edit only, neglecting the need for continuous editing, and (ii) evaluations focusing solely on basic factual triples, overlooking broader LLM capabilities like logical reasoning and reading understanding.…
SCAN: Sparse Circuit Anchor Interpretable Neuron for Lifelong Knowledge Editing
New Laboratory of Pattern Recognition (NLPR), State Key Laboratory of Multimodal Artificial Intelligence Systems (MAIS), Institute of Automation, Chinese Academy of Sciences（模式识别新实验室（NLPR）、多模态人工智能系统国家重点实验室（MAIS）、自动化研究所、中国科学院）；Cuiying Honors College, Lanzhou University（cuiying荣誉学院、兰州大学）；Research Institute of Intelligent Complex Systems, Fudan University（智能复杂系统研究院、复旦大学）；Zhongguancun Academy（中关村学院）
Harbin Institute of Technology, Shenzhen, China.（哈尔滨工业大学（深圳））；Peng Cheng Laboratory, China.（鹏城实验室）；Huazhong University of Science and Technology, China（华中科技大学）
Potential and Challenges of Model Editing for Social Debiasing
Zhejiang University School of Engineering（浙江大学工程学院）；Westlake University（西湖大学）；Institute of Advanced Technology, Westlake Institute for Advanced Study（西湖高等研究院先进技术研究所）
The Butterfly Effect of Model Editing: Few Edits Can Trigger Large Language Models Collapse
CAS Key Laboratory of AI Safety, Institute of Computing Technology, Chinese Academy of Sciences（中国科学院计算技术研究所人工智能安全重点实验室）；University of Chinese Academy of Sciences（中国科学院大学）；Baidu Inc.（百度公司）
REPAIR: Robust Editing via Progressive Adaptive Intervention and Reintegration
Engineering Center for Space Utilization of Chinese Academy of Sciences（中国科学院空间利用工程技术中心）；University of Chinese Academy of Sciences（中国科学院大学）；School of Computer Science and Engineering（计算机科学与工程学院）；Northeastern University（东北大学）；New York University（纽约大学）；School of Computing（计算机学院）；National University of Singapore（新加坡国立大学）；The Chinese University of Hong Kong（香港中文大学）；Technical University of Munich（慕尼黑技术大学）
CodeEditorBench: Evaluating Code Editing Capability of Large Language Models
Multimodal Art Projection Research Community（多模态艺术投影研究社区）；University of Waterloo（滑铁卢大学）；HKUST（香港科技大学）；University of Manchester（曼彻斯特大学）；Tongji University（同济大学）；Vector Institute（矢量研究所）
TimeROME-DLM: Temporal Causal Tracing and Low-Rank Inference-Time Knowledge Editing for Masked Diffusion Language Models
Shanghai Jiao Tong University（上海交通大学）；Nanyang Technological University（南洋理工大学）；National University of Singapore（新加坡国立大学）；University of Science and Technology of China（中国科学技术大学）
ProtTeX: Structure-In-Context Reasoning and Editing of Proteins with Large Language Models
Changping Laboratory（昌平实验室）；School of Computer Science and Technology, Soochow University（苏州大学计算机科学与技术学院）；College of Chemistry and Molecular Engineering, Peking University（北京大学化学与分子工程学院）
GenRecEdit: Adapting Model Editing for Generative Recommendation with Cold-Start Items
Gaoling School of Artificial Intelligence, Renmin University of China（中国人民大学全球化人工智能学院）；School of Information Technology and Management, University of International Business and Economics（国际经济贸易大学信息科学技术与管理学院）
EvoEdit: Evolving Null-space Alignment for Robust and Efficient Knowledge Editing
Institute for Clarity in Documentation（文档清晰度研究所）；Inria Paris-Rocquencourt（法国国家信息与自动化研究所巴黎-罗康库尔中心）；Rajiv Gandhi University（拉吉夫·甘地大学）；Tsinghua University（清华大学）；Palmer Research Laboratories（帕默研究实验室）
Searching for novel molecular compounds with desired properties is an important problem in drug discovery. Many existing frameworks generate molecules one atom at a time. We instead propose a flexible editing paradigm that generates molecules using learned molecular fragments--meaningful substructures of molecules. To do so, we train a variational autoencoder (VAE) to encode molecular fragments in a coherent latent space, which we then utilize as a vocabulary for editing molecules to explore the complex chemical property space. Equipped with the learned fragment vocabulary, we propose Fragment…
Edit Less, Achieve More: Dynamic Sparse Neuron Masking for Lifelong Knowledge Editing in LLMs
Key Lab of Intell. Info. Process., Inst. of Comput. Tech., CAS（智能信息处理重点实验室，计算技术研究所，中国科学院）；University of Chinese Academy of Sciences（中国科学院大学）；Agriculture Information Institute, CAAS（农业信息研究所，中国农业科学院）
The Ohio State University（俄亥俄州立大学）；Carnegie Mellon University（卡内基梅隆大学）
While most neural generative models generate outputs in a single pass, the human creative process is usually one of iterative building and refinement. Recent work has proposed models of editing processes, but these mostly focus on editing sequential data and/or only model a single editing pass. In this paper, we present a generic model for incremental editing of structured data (i.e., "structural edits"). Particularly, we focus on tree-structured data, taking abstract syntax trees of computer programs as our canonical example. Our editor learns to iteratively generate tree edits (e.g., deletin…
LadderEdit: Edit-Level Residual Compression for Memory-Efficient Lifelong Editing of LLMs
Yale University（耶鲁大学）；Washington University in St. Louis（圣路易斯华盛顿大学）；Icahn School of Medicine at Mount Sinai（西奈山伊坎医学院）；Arizona State University（亚利桑那州立大学）
The medoid of a set of n points is the point in the set that minimizes the sum of distances to other points. It can be determined exactly in O(n^2) time by computing the distances between all pairs of points. Previous works show that one can significantly reduce the number of distance computations needed by adaptively querying distances. The resulting randomized algorithm is obtained by a direct conversion of the computation problem to a multi-armed bandit statistical inference problem. In this work, we show that we can better exploit the structure of the underlying computation problem by modi…
