Breaking Down the Defenses: A Comparative Survey of Attacks on Large Language Models
University of Illinois Urbana-Champaign（伊利诺伊大学厄巴纳-香槟分校）；University of California, Los Angeles（加利福尼亚大学洛杉矶分校）；University of Texas at Arlington（德克萨斯大学阿灵顿分校）；Georgia Institute of Technology（佐治亚理工学院）；Stanford University（斯坦福大学）；Amazon GenAI（亚马逊生成式人工智能部门）
Large Language Models (LLMs) have become a cornerstone in the field of Natural Language Processing (NLP), offering transformative capabilities in understanding and generating human-like text. However, with their rising prominence, the security and vulnerability aspects of these models have garnered significant attention. This paper presents a comprehensive survey of the various forms of attacks targeting LLMs, discussing the nature and mechanisms of these attacks, their potential impacts, and current defense strategies. We delve into topics such as adversarial attacks that aim to manipulate model outputs, data poisoning that affects model training, and privacy concerns related to training data exploitation. The paper also explores the effectiveness of different attack methodologies, the resilience of LLMs against these attacks, and the implications for model integrity and user trust. By examining the latest research, we provide insights into the current landscape of LLM vulnerabilities and defense mechanisms. Our objective is to offer a nuanced understanding of LLM attacks, foster awareness within the AI community, and inspire robust solutions to mitigate these risks in future developments.
Merging Triggers, Breaking Backdoors: Defensive Poisoning for Instruction-Tuned Language Models
Graduate School of Artificial Intelligence, POSTECH, Republic of Korea（POSTECH人工智能研究生院，韩国）；Department of Computer Science and Engineering, POSTECH, Republic of Korea（POSTECH计算机科学与工程系，韩国）
FRACTURED-SORRY-Bench: Framework for Revealing Attacks in Conversational Turns Undermining Refusal Efficacy and Defenses over SORRY-Bench (Automated Multi-shot Jailbreaks)
A Survey on Proactive Defense Strategies Against Misinformation in Large Language Models
The Hong Kong University of Science and Technology (Guangzhou)（香港科技大学（广州））；The Hong Kong University of Science and Technology（香港科技大学）；Harbin Institute of Technology（哈尔滨工业大学）；Ant Group, Alibaba（蚂蚁集团，阿里巴巴）；Northeast Forest University（东北林业大学）
Feint and Attack: Attention-Based Strategies for Jailbreaking and Protecting LLMs
Beijing University of Posts and Telecommunications（北京邮电大学）；Hangzhou Dianzi University（杭州电子科技大学）；Beijing Academy of Artificial Intelligence（北京人工智能研究院）；Fujian Cancer Hospital（福建癌症医院）
University of California（加利福尼亚大学）；Lawrence Livermore National Lab（劳伦斯利弗莫尔国家实验室）
我们研究了找到一个通用的（图像不可知的）扰动来欺骗机器学习（ML）分类器（例如，神经网络，决策树）在硬标签黑盒设置。白盒环境中对抗性ML的最新工作（模型参数已知）已经表明，许多最先进的图像分类器容易受到普遍对抗性扰动的影响：一种固定的人类无法感知的扰动，当添加到任何图像时，会导致其被错误分类的概率很高Kurakin et al.[2016]，Szegedy et al. [2013]，Chen et al. [2017 a]，Carlini和Wagner [2017]。本文考虑了一个更实际和具有挑战性的问题，找到这样的普遍扰动在一个模糊的（或黑盒）设置。更具体地说，我们使用零阶优化算法来找到这样一个普遍的对抗扰动时，没有模型信息被揭示，除了攻击者可以进行查询，以探测分类。我们进一步放松的假设，查询的输出是连续值的置信度分数的所有类，并考虑输出是一个硬标签的决定的情况下。令人惊讶的是，我们发现，即使在这些极其模糊的制度中，最先进的ML分类器也可以通过向任何自然图像添加一个人类无法感知的图像扰动来以非常高的概率被愚弄。在硬标签黑盒设置中令人惊讶的普遍扰动的存在引起了严重的安全问题，因为存在一个普遍的噪声向量，对手可能会利用它来破坏大多数自然图像上的分类器。
Enhancing Model Defense Against Jailbreaks with Proactive Safety Reasoning
School of Computing（计算学院）；National University of Singapore（新加坡国立大学）；School of Computer Science and Engineering（计算机科学与工程学院）；Nanyang Technological University（南洋理工大学）；Department of Computing（计算系）；The Hong Kong Polytechnic University（香港理工大学）
Privacy-preserving Universal Adversarial Defense for Black-box Models
School of Cyber Science and Engineering, Wuhan University（武汉大学网络科学与工程学院）；City University of Hong Kong (CityU)（香港城市大学）；College of Computing and Data Science, Nanyang Technological University（南洋理工大学计算与数据科学学院）
SoK: Benchmarking Poisoning Attacks and Defenses in Federated Learning
Shanghai Jiao Tong University（上海交通大学）；Hong Kong University of Science and Technology (Guangzhou)（香港科技大学（广州））；Tsinghua University（清华大学）；Jinan University（暨南大学）
Breaking the Secret: Economic Interventions for Combating Collusion in Embodied Multi-Agent Systems
College of Computer Science and Technology, Taiyuan University of Technology（太原理工大学计算机科学与技术学院）；State Key Laboratory of Internet Architecture, Tsinghua University（清华大学互联网架构国家重点实验室）；China Mobile Research Institute（中国移动研究院）；College of Artificial Intelligence, Taiyuan University of Technology（太原理工大学人工智能学院）；Shanxi Key Laboratory of Industrial Internet Security（山西省工业互联网安全重点实验室）；Institute of Information Engineering, Chinese Academy of Sciences（中国科学院信息工程研究所）
