Towards Robust Red-Green Watermarking for Autoregressive Image Generators
By: Denis Lukovnikov , Andreas Müller , Erwin Quiring and more
Potential Business Impact:
Marks AI-made pictures so you know they're fake.
In-generation watermarking for detecting and attributing generated content has recently been explored for latent diffusion models (LDMs), demonstrating high robustness. However, the use of in-generation watermarks in autoregressive (AR) image models has not been explored yet. AR models generate images by autoregressively predicting a sequence of visual tokens that are then decoded into pixels using a vector-quantized decoder. Inspired by red-green watermarks for large language models, we examine token-level watermarking schemes that bias the next-token prediction based on prior tokens. We find that a direct transfer of these schemes works in principle, but the detectability of the watermarks decreases considerably under common image perturbations. As a remedy, we propose two novel watermarking methods that rely on visual token clustering to assign similar tokens to the same set. Firstly, we investigate a training-free approach that relies on a cluster lookup table, and secondly, we finetune VAE encoders to predict token clusters directly from perturbed images. Overall, our experiments show that cluster-level watermarks improve robustness against perturbations and regeneration attacks while preserving image quality. Cluster classification further boosts watermark detectability, outperforming a set of baselines. Moreover, our methods offer fast verification runtime, comparable to lightweight post-hoc watermarking methods.
Similar Papers
Autoregressive Images Watermarking through Lexical Biasing: An Approach Resistant to Regeneration Attack
Cryptography and Security
Marks AI-made pictures so they can't be faked.
Watermarking Autoregressive Image Generation
Machine Learning (CS)
Marks AI-made pictures so we know they're fake.
A Watermark for Auto-Regressive Image Generation Models
CV and Pattern Recognition
Marks fake pictures so you know they're not real.