Large language models remain vulnerable to jailbreak attacks, yet we still lack a systematic understanding of how jailbreak success scales with attacker effort across methods, model families, and harm types. We initiate a scaling-law framework for jailbreaks by treating each attack as a compute-bounded optimization procedure and measuring progress on a shared FLOPs axis. Our systematic evaluation spans four representative jailbreak paradigms, covering optimization-based attacks, self-refinement prompting, sampling-based selection, and genetic optimization, across multiple model families and scales on a diverse set of harmful goals. We investigate scaling laws that relate attacker budget to attack success score by fitting a simple saturating exponential function to FLOPs--success trajectories, and we derive comparable efficiency summaries from the fitted curves. Empirically, prompting-based paradigms tend to be the most compute-efficient compared to optimization-based methods. To explain this gap, we cast prompt-based updates into an optimization view and show via a same-state comparison that prompt-based attacks more effectively optimize in prompt space. We also show that attacks occupy distinct success--stealthiness operating points with prompting-based methods occupying the high-success, high-stealth region. Finally, we find that vulnerability is strongly goal-dependent: harms involving misinformation are typically easier to elicit than other non-misinformation harms.
Jailbreak Attack Initializations as Extractors of Compliance Directions
Department of Computer Science, Technion - Israel Institute of Technology(技术学院计算机科学系);Department of Data and Decision Science, Technion - Israel Institute of Technology(技术学院数据与决策科学系);School of Electrical and Computer Engineering Engineering, Ben-Gurion University of the Negev(内盖夫本· Gurion大学电气与计算机工程学院)
The Art of the Jailbreak: Formulating Jailbreak Attacks for LLM Security Beyond Binary Scoring
Department of Computer Science, University of Texas at El Paso(德克萨斯大学埃尔帕索分校计算机科学系);School of Computing, Southern Illinois University Carbondale(南方伊利诺伊大学卡罗尔丹分校计算学院);University of Illinois Urbana-Champaign(伊利诺伊大学厄巴纳-香槟分校)
Immune: Improving Safety Against Jailbreaks in Multi-modal LLMs via Inference-Time Alignment
University of Maryland(马里兰大学);Indian Institute of Technology Bombay(印度班加罗尔理工学院);Princeton University(普林斯顿大学);University of Colorado Boulder(科罗拉多大学博尔德分校);Capital One(Capital One公司);University of Central Florida(佛罗里达中央大学)
The Great Contradiction Showdown: How Jailbreak and Stealth Wrestle in Vision-Language Models?
Institute of Information Science (IIS), Academia Sinica(中央研究院资讯科学研究所);National Taiwan University(国立台湾大学);National Yang Ming Chiao Tung University(国立阳明交通大学)
AudioJailbreak: Jailbreak Attacks against End-to-End Large Audio-Language Models
Wuhan University(武汉大学);Key Laboratory of System Software (Chinese Academy of Sciences)(中国科学院系统软件重点实验室);Institute of Software, Chinese Academy of Sciences(中国科学院软件研究所);State Key Laboratory of Cryptology(密码学国家重点实验室);University of Chinese Academy of Sciences(中国科学院大学);Nanjing Institute of Software Technology(南京软件技术研究所);Ant Group(蚂蚁集团);Nanyang Technological University(南洋理工大学);Zhejiang Lab(浙江实验室);Pengcheng Laboratory(鹏城实验室)
Toward Universal and Transferable Jailbreak Attacks on Vision-Language Models
School of Computing and Information Systems, The University of Melbourne, Australia(墨尔本大学计算机与信息系);School of Computing and Information Systems, Singapore Management University, Singapore(新加坡管理大学计算机与信息系);School of Information Technology, Deakin University, Australia(德肯大学信息科技系);Institute of Trustworthy Embodied AI, Fudan University, China(复旦大学可信具身人工智能研究所)