Components of machine learning systems are not (yet) perceived as security hotspots. Secure coding practices, such as ensuring that no execution paths depend on confidential inputs, have not yet been adopted by ML developers. We initiate the study of code security of ML systems by investigating how nucleus sampling---a popular approach for generating text, used for applications such as auto-completion---unwittingly leaks texts typed by users. Our main result is that the series of nucleus sizes for many natural English word sequences is a unique fingerprint. We then show how an attacker can infer typed text by measuring these fingerprints via a suitable side channel (e.g., cache access times), explain how this attack could help de-anonymize anonymous texts, and discuss defenses.
Prompt2Fingerprint: Plug-and-Play LLM Fingerprinting via Text-to-Weight Generation
Shenzhen International Graduate School, Tsinghua University, Shenzhen, China(清华大学深圳国际研究生院,中国深圳);South China University of Technology, Guangzhou, China(华南理工大学,中国广州);Harbin Institute of Technology, Shenzhen, Shenzhen, China(哈尔滨工业大学深圳校区,中国深圳)
AttnDiff: Attention-based Differential Fingerprinting for Large Language Models
Zhejiang University of Technology(浙江工业大学);Zhejiang University(浙江大学);Binjiang Institute of Zhejiang University(浙江大学滨江研究院);Shanghai Jiao Tong University(上海交通大学);University of Science and Technology of China(中国科学技术大学)
Fingerprinting Text-to-Image Diffusion Models via Collapsed Generation
Fudan University(复旦大学);East China University of Science and Technology(华东理工大学);Alibaba Group(阿里巴巴集团);Shanghai Pudong Research Institute of Cryptology(上海浦东密码学研究所)
DNF: Dual-Layer Nested Fingerprinting for Large Language Model Intellectual Property Protection
Zhejiang University(浙江大学);Binjiang Institute of Zhejiang University(浙江大学滨江学院);Hangzhou Dianzi University(杭州电子科技大学);Hefei University of Technology(合肥工业大学)
Efficient, Robust, and Anti-Collusion Fingerprinting of Image Diffusion Models
University of Florence(佛罗伦萨大学);Shenzhen Campus of Sun Yat-sen University(中山大学深圳校区);College of Cyber Security, Jinan University(暨南大学网络空间安全学院);State Key Laboratory of Internet of Things for Smart City, University of Macau(澳门大学智慧城市物联网国家重点实验室);Department of Computer and Information Science, Faculty of Science and Technology, University of Macau(澳门大学科技学院计算机与信息科学系);University of Siena(锡耶纳大学)
LLMs Have Rhythm: Fingerprinting Large Language Models Using Inter-Token Times and Network Traffic Analysis
College of Science and Engineering (CSE), Hamad Bin Khalifa University (HBKU)(哈马德·本·卡伊夫大学科学与工程学院);Networked Systems Security Group, KTH Royal Institute of Technology -- Stockholm, Sweden(瑞典皇家理工学院网络系统安全组)
Addendum: Systematic Evaluation of Randomized Cache Designs against Cache Occupancy
Max Planck Institute for Security and Privacy(马克斯·普朗克安全与隐私研究所);Indian Institute of Technology Kharagpur(印度理工学院卡拉格普尔分校);Indian Institute of Technology Bombay(印度理工学院孟买分校)