Generative Models are Self-Watermarked: Declaring Model Authentication through Re-Generation

arXiv:2402.16889v1 Announce Type: cross
Abstract: As machine- and AI-generated content proliferates, protecting the intellectual property of generative models has become imperative, yet verifying data ownership poses formidable challenges, particularly in cases of unauthorized reuse of generated data. The challenge of verifying data ownership is further amplified by using Machine Learning as a Service (MLaaS), which often functions as a black-box system.
Our work is dedicated to detecting data reuse from even an individual sample. Traditionally, watermarking has been leveraged to …

arxiv authentication cases challenge challenges cs.ai cs.cr cs.lg data data ownership generated generative generative models intellectual property machine machine learning ownership property protecting reuse unauthorized