Evaluating Durability: Benchmark Insights into Multimodal Watermarking

TL;DR

This paper assesses the robustness of multimodal watermarks in image and text models against real-world corruptions, emphasizing the importance of durability for copyright and authenticity verification.

Contribution

It provides the first comprehensive evaluation of watermark robustness in multimodal content under real-world disturbances, guiding future improvements.

Findings

Watermarks show varying robustness to different corruptions.

Robustness is crucial for reliable copyright verification.

Results highlight the need for developing more resilient watermarking methods.

Abstract

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/}.