To help enforce data-protection regulations such as GDPR and detect unauthorized uses of personal data, we develop a new \emph{model auditing} technique that helps users check if their data was used to train a machine learning model. We focus on auditing deep-learning models that generate natural-language text, including word prediction and dialog generation. These models are at the core of popular online services and are often trained on personal data such as users' messages, searches, chats, and comments.
We design and evaluate a black-box auditing method that can detect, with very few queries to a model, if a particular user's texts were used to train it (among thousands of other users). We empirically show that our method can successfully audit well-generalized models that are not overfitted to the training data. We also analyze how text-generation models memorize word sequences and explain why this memorization makes them amenable to auditing.
TROVE: A Challenge for Fine-Grained Text Provenance via Source Sentence Tracing and Relationship Classification
State Key Laboratory of Multimodal Artificial Intelligence Systems, Institute of Automation, CAS(多模态人工智能系统国家重点实验室,自动化研究所);School of Artificial Intelligence, University of Chinese Academy of Sciences(中国科学院大学人工智能学院);Fanyu AI Laboratory, Zhongke Fanyu Technology Co., Ltd(中科创始人工智能实验室,中科创始科技有限公司)
vCause: Efficient and Verifiable Causality Analysis for Cloud-based Endpoint Auditing
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空间安全学院);University of Electronic Science and Technology of China(电子科技大学);Qi An Xin Technology Group Inc.(奇安信科技集团股份有限公司)
MemCatalyst: Amplifying Data Auditing on Vision-Language Models via Data Poisoning
School of Computer Science and Technology, Beijing Institute of Technology(北京理工大学计算机科学与技术学院);School of Computer Science and Engineering, Northeastern University(东北大学计算机科学与工程学院);Faculty of Computer Science, Dalhousie University(达尔豪斯大学计算机科学学院);King Abdullah University of Science and Technology(阿卜杜拉国王科技大学)
ArtistAuditor: Auditing Artist Style Pirate in Text-to-Image Generation Models
Xi’an Jiaotong University(西安交通大学);Zhejiang University(浙江大学);The Chinese University of Hong Kong(香港中文大学);Vrije Universiteit Amsterdam(阿姆斯特丹自由大学);Hangzhou Dianzi University(杭州电子科技大学)
Membership Inference for Contrastive Pre-training Models with Text-only PII Queries
Beijing Electronic Science and Technology Institute(北京电子科技研究所);Institute of Artificial Intelligence, China Telecom (TeleAI)(中国电信人工智能研究院(TeleAI));Nanyang Technological University(南洋理工大学);Nanjing University(南京大学);School of Computing and Information Technology, Great Bay University(广东大湾区大学计算机与信息科技学院)