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)
Amit Levi, Rom Himelstein, Yaniv Nemcovsky, Avi Mendelson, Chaim Baskin
Safety-aligned LLMs respond to prompts with either compliance or refusal, each corresponding to distinct directions in the model's activation space. Recent works show that initializing attacks via self-transfer from other prompts significantly enhances their performance. However, the underlying mechanisms of these initializations remain unclear, and attacks utilize arbitrary or hand-picked initializations. This work presents that each gradient-based jailbreak attack and subsequent initialization gradually converge to a single compliance direction that suppresses refusal, thereby enabling an efficient transition from refusal to compliance. Based on this insight, we propose CRI, an initialization framework that aims to project unseen prompts further along compliance directions. We demonstrate our approach on multiple attacks, models, and datasets, achieving an increased attack success rate (ASR) and reduced computational overhead, highlighting the fragility of safety-aligned LLMs. A reference implementation is available at: https://amit1221levi.github.io/CRI-Jailbreak-Init-LLMs-evaluation.
Behind Harmful Compliance: Behavioral and Mechanistic Divergence Across LLM Jailbreaks
Department of Computer Science(计算机科学系);Luddy School of Informatics, Computing, and Engineering(信息学、计算与工程学院);Indiana University Bloomington(印第安纳大学布卢明顿分校)
ForgeDAN: An Evolutionary Framework for Jailbreaking Aligned Large Language Models
iFLYTEK Security Laboratory(iFLYTEK安全实验室);Anhui SparkShield Intelligent Technology Co., Ltd.(安徽SparkShield智能科技有限公司);Peking University(北京大学);School of Automation, University of Electronic Science and Technology of China(电子科技大学自动化学院);Northwest University(西北大学);University of New South Wales(新南威尔士大学);National Computer Network Emergency Response Technical Team/Coordination Center of China(中国国家计算机网络应急技术处置协调中心)
Prefill-level Jailbreak: A Black-Box Risk Analysis of Large Language Models
Institute of Information Engineering, Chinese Academy of Sciences(中国科学院信息工程研究所);School of Cyber Security, University of Chinese Academy of Sciences(中国科学院大学网络安全学院);Sinochem Energy High-Tech Co., Ltd.(中石化能源高科技有限公司)
Bleeding Pathways: Vanishing Discriminability in LLM Hidden States Fuels Jailbreak Attacks
Institute of Information Engineering, Chinese Academy of Sciences(中国科学院信息工程研究所);School of Cyber Security, University of Chinese Academy of Sciences(中国科学院大学网络安全学院);Ant Group(蚂蚁集团)
Nanjing University(南京大学);Microsoft Research Asia(微软亚洲研究院);Tsinghua University(清华大学);Queen Mary University of London(伦敦玛丽女王大学);Pennsylvania State University(宾夕法尼亚州立大学);NEC Laboratories America(美国 NEC 实验室)
TRAP: Targeted Random Adversarial Prompt Honeypot for Black-Box Identification
Parameter Lab(参数实验室);IT University of Copenhagen(哥本哈根信息技术大学);Pioneer Centre for Artificial Intelligence(人工智能先锋中心);NAVER AI Lab(NAVER人工智能实验室);University of Tübingen(蒂宾根大学)