随着大语言模型（Large Language Models, LLMs）的快速发展并取得显著成功，理解与修正其内部复杂机制已成为亟待解决的问题。近期研究尝试通过内部表征的视角解读其行为，但开发实用高效的方法以将这些表征应用于通用灵活的模型编辑仍具挑战。本研究探索如何利用表征工程的洞见指导LLM编辑，将表征传感器用作编辑预言机。我们首先明确了编辑过程中稳健可靠传感器的重要性，随后提出对抗表征工程（Adversarial Representation Engineering, ARE）框架，为概念性模型编辑提供统一且可解释的方法，且不会损害基准性能。在多项任务上的实验证明了ARE在各类模型编辑场景中的有效性。我们的代码与数据可在https://github.com/Zhang-Yihao/Adversarial-Representation-Engineering获取。
英文摘要：
Since the rapid development of Large Language Models (LLMs) has achieved remarkable success, understanding and rectifying their internal complex mechanisms has become an urgent issue. Recent research has attempted to interpret their behaviors through the lens of inner representation. However, developing practical and efficient methods for applying these representations for general and flexible model editing remains challenging. In this work, we explore how to leverage insights from representation engineering to guide the editing of LLMs by deploying a representation sensor as an editing oracle. We first identify the importance of a robust and reliable sensor during editing, then propose an Adversarial Representation Engineering (ARE) framework to provide a unified and interpretable approach for conceptual model editing without compromising baseline performance. Experiments on multiple tasks demonstrate the effectiveness of ARE in various model editing scenarios. Our code and data are available at https://github.com/Zhang-Yihao/Adversarial-Representation-Engineering.
DELMAN: Dynamic Defense Against Large Language Model Jailbreaking with Model Editing
ShanghaiTech University（上海科技大学）；The State Key Laboratory of Blockchain and Data Security, Zhejiang University（浙江大学区块链与数据安全全国重点实验室）；The CoAI group, Tsinghua University（清华大学CoAI课题组）
Instruction-Guided Editing Controls for Images and Multimedia: A Survey in LLM era
Griffith University（格里菲斯大学）；University of Bremen（不来梅大学）；Ho Chi Minh City University of Technology（胡志明市技术大学）；École Polytechnique Fédérale de Lausanne（洛桑联邦理工学院）；Hanoi University of Science and Technology（河内科技大学）；The University of Queensland（昆士兰大学）
Consistency-Aware Editing for Entity-level Unlearning in Language Models
Shanxi University, China（山西大学）；Cardiff University, UK（卡地夫大学）；Singapore University of Technology and Design, Singapore（新加坡科技设计大学）；ILCC, School of Informatics, University of Edinburgh, Edinburgh, UK（爱丁堡大学信息学院ILCC）
AdLift: Lifting Adversarial Perturbations to Safeguard 3D Gaussian Splatting Assets Against Instruction-Driven Editing
Sydney AI Centre, The University of Sydney（悉尼人工智能中心，悉尼大学）；University of Technology Sydney（技术大学悉尼）；University of Melbourne（墨尔本大学）；Hong Kong Baptist University（香港 Baptist 大学）；Mohamed bin Zayed University of Artificial Intelligence（Mohamed bin Zayed 人工智能大学）
SemanticAdv: Generating Adversarial Examples via Attribute-conditional Image Editing
The Chinese University of Hong Kong, Shenzhen（香港中文大学（深圳））；University of Michigan, Ann Arbor（密歇根大学安娜堡分校）；The Chinese University of Hong Kong（香港中文大学）；Uber ATG（优步先进技术集团）；University of Illinois Urbana-Champaign（伊利诺伊大学厄巴纳-香槟分校）
Adversarial Vulnerability Transcends Computational Paradigms: Feature Engineering Provides No Defense Against Neural Adversarial Transfer
1 Department of Information \& Computer Science, KFUPM, Dhahran, Saudi Arabia；2 Department of Computer Engineering, KFUPM, Dhahran, Saudi Arabia；3 Interdisciplinary Research Center for Finance
Towards Transferable Defense Against Malicious Image Edits
State Key Laboratory of AI Safety, Institute of Computing Technology, Chinese Academy of Sciences (CAS)（人工智能安全国家重点实验室，计算技术研究所，中国科学院）；University of China Academy of Sciences（中国科学院大学）；School of Computer Science, China University of Geosciences（中国地质大学（武汉）计算机学院）
