Agentic Copyright Watermarking against Adversarial Evidence Forgery with Purification-Agnostic Curriculum Proxy Learning
This work addresses copyright protection for AI models, which is crucial for developers facing threats from illegal distribution, but it appears incremental as it builds on existing watermarking techniques with specific enhancements.
The paper tackles the problem of protecting AI model ownership against unauthorized use by developing a model watermarking protocol and defenses against adversarial evidence forgery, resulting in improved security, reliability, and performance of watermarked models as demonstrated experimentally.
With the proliferation of AI agents in various domains, protecting the ownership of AI models has become crucial due to the significant investment in their development. Unauthorized use and illegal distribution of these models pose serious threats to intellectual property, necessitating effective copyright protection measures. Model watermarking has emerged as a key technique to address this issue, embedding ownership information within models to assert rightful ownership during copyright disputes. This paper presents several contributions to model watermarking: a self-authenticating black-box watermarking protocol using hash techniques, a study on evidence forgery attacks using adversarial perturbations, a proposed defense involving a purification step to counter adversarial attacks, and a purification-agnostic curriculum proxy learning method to enhance watermark robustness and model performance. Experimental results demonstrate the effectiveness of these approaches in improving the security, reliability, and performance of watermarked models.