Questions of fair use of copyright-protected content to train Large Language Models (LLMs) are being actively debated. Document-level inference has been proposed as a new task: inferring from black-box access to the trained model whether a piece of content has been seen during training. SOTA methods however rely on naturally occurring memorization of (part of) the content. While very effective against models that memorize significantly, we hypothesize--and later confirm--that they will not work against models that do not naturally memorize, e.g. medium-size 1B models. We here propose to use copyright traps, the inclusion of fictitious entries in original content, to detect the use of copyrighted materials in LLMs with a focus on models where memorization does not naturally occur. We carefully design a randomized controlled experimental setup, inserting traps into original content (books) and train a 1.3B LLM from scratch. We first validate that the use of content in our target model would be undetectable using existing methods. We then show, contrary to intuition, that even medium-length trap sentences repeated a significant number of times (100) are not detectable using existing methods. However, we show that longer sequences repeated a large number of times can be reliably detected (AUC=0.75) and used as copyright traps. Beyond copyright applications, our findings contribute to the study of LLM memorization: the randomized controlled setup enables us to draw causal relationships between memorization and certain sequence properties such as repetition in model training data and perplexity.
Digger: Detecting Copyright Content Mis-usage in Large Language Model Training
Beijing University of Posts and Telecommunications(北京邮电大学);Nanyang Technological University(南洋理工大学);Huazhong University of Science and Technology(华中科技大学);The University of New South Wales(新南威尔士大学);Harbin Institute of Technology, Shenzhen(哈尔滨工业大学(深圳))
From Essence to Defense: Adaptive Semantic-aware Watermarking for Embedding-as-a-Service Copyright Protection
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)
SoK: Large Language Model Copyright Auditing via Fingerprinting
The State Key Laboratory of Blockchain and Data Security(区块链与数据安全国家重点实验室);Hangzhou High-Tech Zone (Binjiang) Institute of Blockchain and Data Security(杭州高新技术区(滨江)区块链与数据安全研究院);Nanyang Technological University(南洋理工大学);City University of Hong Kong(香港城市大学);Sun Yat-Sen University(中山大学)
Are You Copying My Model? Protecting the Copyright of Large Language Models for EaaS via Backdoor Watermark
Beijing Jiaotong University(北京交通大学);Microsoft Research Asia(微软亚洲研究院);Sony AI(索尼人工智能);Microsoft STC Asia(微软亚洲软件技术中心);University of Science and Technology of China(中国科学技术大学)
Leave No TRACE: Black-box Detection of Copyrighted Dataset Usage in Large Language Models via Watermarking
Antiquus S. Hippocampus, Natalia Cerebro & Amelie P. Amygdale Department of Computer Science Cranberry-Lemon University Pittsburgh, PA 15213, USA(计算机科学系,Cranberry-Lemon大学);Ji Q. Ren & Yevgeny LeNet Department of Computational Neuroscience University of the Witwatersrand Joburg, South Africa(计算神经科学系,沃特瓦特斯兰大学)
ISACL: Internal State Analyzer for Copyrighted Training Data Leakage
City University of Hong Kong(香港城市大学);Microsoft(微软);University of Illinois Urbana-Champaign(伊利诺伊大学厄巴纳-香槟分校);Google LLC(谷歌公司);Arizona State University(亚利桑那州立大学);Stevens Institute of Technology(史蒂文斯理工学院)
The KL3M Data Project: Copyright-Clean Training Resources for Large Language Models
Institute for the Advancement of Legal and Ethical AI (ALEA Institute)(法律与伦理AI推进研究所(ALEA研究所));CodeX—The Stanford Center for Legal Informatics(CodeX——斯坦福法律信息学中心);Illinois Tech—Chicago Kent College of Law(伊利诺伊理工大学芝加哥肯特法学院);Center for Legal Technology & Data Science, Bucerius Law School(布塞留斯法学院法律技术与数据科学中心)
Inner-Probe: Discovering Copyright-related Data Generation in LLM Architecture
School of Computer Science and the State Key Laboratory of Multimedia Information Processing, Peking University(计算机科学系和多媒体信息处理国家重点实验室,北京大学);Institute for Artificial Intelligence, Peking University(人工智能研究院,北京大学);department of Electrical and Computer Engineering, University of California, Santa Cruz(电气与计算机工程系,加州大学圣克鲁兹分校);Alibaba DAMO Academy, Beijing(阿里巴巴达摩院,北京);College of Computing and Data Science, Nanyang Technological University(计算与数据科学学院,南洋理工大学);School of Integrated Circuits, Peking University(集成电路系,北京大学);YanXin MicroElectronics Co., Ltd.(YXME), Shanghai, China(燕芯微电子有限公司(YXME),上海,中国)
Project Alexandria: Towards Freeing Scientific Knowledge from Copyright Burdens via LLMs
LAION;Open-Ψ (Open-Sci) Collective(Open-Ψ(Open-Sci)集体);L3S Research Center, Leibniz University Hannover(汉诺威莱布尼茨大学L3S研究中心);Stanford University(斯坦福大学);TIB—Leibniz Information Centre for Science and Technology(莱布尼茨科技信息中心);Juelich Supercomputing Center(于利希超级计算中心);Technical University of Munich(慕尼黑工业大学)
A Survey of Text Watermarking in the Era of Large Language Models
Tsinghua University(清华大学);The Chinese University of Hong Kong(香港中文大学);The Hong Kong University of Science and Technology (Guangzhou)(香港科技大学(广州));Beijing University of Posts and Telecommunications(北京邮电大学);University of Illinois Chicago(芝加哥大学伊利诺伊分校)