Autoregressive Images Watermarking through Lexical Biasing: An Approach Resistant to Regeneration Attack

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Hui, Siqi, Song, Yiren, Zhou, Sanping, Deng, Ye, Huang, Wenli, Wang, Jinjun
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866908389416632320
author Hui, Siqi
Song, Yiren
Zhou, Sanping
Deng, Ye
Huang, Wenli
Wang, Jinjun
author_facet Hui, Siqi
Song, Yiren
Zhou, Sanping
Deng, Ye
Huang, Wenli
Wang, Jinjun
contents Autoregressive (AR) image generation models have gained increasing attention for their breakthroughs in synthesis quality, highlighting the need for robust watermarking to prevent misuse. However, existing in-generation watermarking techniques are primarily designed for diffusion models, where watermarks are embedded within diffusion latent states. This design poses significant challenges for direct adaptation to AR models, which generate images sequentially through token prediction. Moreover, diffusion-based regeneration attacks can effectively erase such watermarks by perturbing diffusion latent states. To address these challenges, we propose Lexical Bias Watermarking (LBW), a novel framework designed for AR models that resists regeneration attacks. LBW embeds watermarks directly into token maps by biasing token selection toward a predefined green list during generation. This approach ensures seamless integration with existing AR models and extends naturally to post-hoc watermarking. To increase the security against white-box attacks, instead of using a single green list, the green list for each image is randomly sampled from a pool of green lists. Watermark detection is performed via quantization and statistical analysis of the token distribution. Extensive experiments demonstrate that LBW achieves superior watermark robustness, particularly in resisting regeneration attacks.
format Preprint
id arxiv_https___arxiv_org_abs_2506_01011
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Autoregressive Images Watermarking through Lexical Biasing: An Approach Resistant to Regeneration Attack
Hui, Siqi
Song, Yiren
Zhou, Sanping
Deng, Ye
Huang, Wenli
Wang, Jinjun
Cryptography and Security
Autoregressive (AR) image generation models have gained increasing attention for their breakthroughs in synthesis quality, highlighting the need for robust watermarking to prevent misuse. However, existing in-generation watermarking techniques are primarily designed for diffusion models, where watermarks are embedded within diffusion latent states. This design poses significant challenges for direct adaptation to AR models, which generate images sequentially through token prediction. Moreover, diffusion-based regeneration attacks can effectively erase such watermarks by perturbing diffusion latent states. To address these challenges, we propose Lexical Bias Watermarking (LBW), a novel framework designed for AR models that resists regeneration attacks. LBW embeds watermarks directly into token maps by biasing token selection toward a predefined green list during generation. This approach ensures seamless integration with existing AR models and extends naturally to post-hoc watermarking. To increase the security against white-box attacks, instead of using a single green list, the green list for each image is randomly sampled from a pool of green lists. Watermark detection is performed via quantization and statistical analysis of the token distribution. Extensive experiments demonstrate that LBW achieves superior watermark robustness, particularly in resisting regeneration attacks.
title Autoregressive Images Watermarking through Lexical Biasing: An Approach Resistant to Regeneration Attack
topic Cryptography and Security
url https://arxiv.org/abs/2506.01011