Evaluating Dataset Watermarking for Fine-tuning Traceability of Customized Diffusion Models: A Comprehensive Benchmark and Removal Approach
This addresses copyright and security risks for users of customized diffusion models, but is incremental as it builds on existing watermarking techniques.
The paper tackled the problem of evaluating dataset watermarking for tracing fine-tuned diffusion models, establishing a comprehensive benchmark that revealed existing methods perform well in universality and transmissibility but lack robustness in real-world scenarios, and proposed a removal method that fully eliminates watermarks without affecting fine-tuning.
Recent fine-tuning techniques for diffusion models enable them to reproduce specific image sets, such as particular faces or artistic styles, but also introduce copyright and security risks. Dataset watermarking has been proposed to ensure traceability by embedding imperceptible watermarks into training images, which remain detectable in outputs even after fine-tuning. However, current methods lack a unified evaluation framework. To address this, this paper establishes a general threat model and introduces a comprehensive evaluation framework encompassing Universality, Transmissibility, and Robustness. Experiments show that existing methods perform well in universality and transmissibility, and exhibit some robustness against common image processing operations, yet still fall short under real-world threat scenarios. To reveal these vulnerabilities, the paper further proposes a practical watermark removal method that fully eliminates dataset watermarks without affecting fine-tuning, highlighting a key challenge for future research.