Evaluating Durability: Benchmark Insights into Multimodal Watermarking
This work addresses the need for reliable watermarking in multimodal AI systems, but it appears incremental as it focuses on benchmarking existing techniques rather than introducing new methods.
The study evaluated the robustness of watermarks in content generated by image and text models against real-world corruptions and perturbations, aiming to highlight the importance of durability for copyright and authenticity in multimodal applications.
With the development of large models, watermarks are increasingly employed to assert copyright, verify authenticity, or monitor content distribution. As applications become more multimodal, the utility of watermarking techniques becomes even more critical. The effectiveness and reliability of these watermarks largely depend on their robustness to various disturbances. However, the robustness of these watermarks in real-world scenarios, particularly under perturbations and corruption, is not well understood. To highlight the significance of robustness in watermarking techniques, our study evaluated the robustness of watermarked content generated by image and text generation models against common real-world image corruptions and text perturbations. Our results could pave the way for the development of more robust watermarking techniques in the future. Our project website can be found at \url{https://mmwatermark-robustness.github.io/}.