Learning to Watermark in the Latent Space of Generative Models
Fuente:
arXiv
Saved in:
| Main Authors: | , , , , , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2026
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866914273532313600 |
|---|---|
| author | Rebuffi, Sylvestre-Alvise Tran, Tuan Lacatusu, Valeriu Fernandez, Pierre Souček, Tomáš Jovanović, Nikola Sander, Tom Elsahar, Hady Mourachko, Alexandre |
| author_facet | Rebuffi, Sylvestre-Alvise Tran, Tuan Lacatusu, Valeriu Fernandez, Pierre Souček, Tomáš Jovanović, Nikola Sander, Tom Elsahar, Hady Mourachko, Alexandre |
| contents | Existing approaches for watermarking AI-generated images often rely on post-hoc methods applied in pixel space, introducing computational overhead and potential visual artifacts. In this work, we explore latent space watermarking and introduce DistSeal, a unified approach for latent watermarking that works across both diffusion and autoregressive models. Our approach works by training post-hoc watermarking models in the latent space of generative models. We demonstrate that these latent watermarkers can be effectively distilled either into the generative model itself or into the latent decoder, enabling in-model watermarking. The resulting latent watermarks achieve competitive robustness while offering similar imperceptibility and up to 20x speedup compared to pixel-space baselines. Our experiments further reveal that distilling latent watermarkers outperforms distilling pixel-space ones, providing a solution that is both more efficient and more robust. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2601_16140 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | Learning to Watermark in the Latent Space of Generative Models Rebuffi, Sylvestre-Alvise Tran, Tuan Lacatusu, Valeriu Fernandez, Pierre Souček, Tomáš Jovanović, Nikola Sander, Tom Elsahar, Hady Mourachko, Alexandre Computer Vision and Pattern Recognition Artificial Intelligence Cryptography and Security Existing approaches for watermarking AI-generated images often rely on post-hoc methods applied in pixel space, introducing computational overhead and potential visual artifacts. In this work, we explore latent space watermarking and introduce DistSeal, a unified approach for latent watermarking that works across both diffusion and autoregressive models. Our approach works by training post-hoc watermarking models in the latent space of generative models. We demonstrate that these latent watermarkers can be effectively distilled either into the generative model itself or into the latent decoder, enabling in-model watermarking. The resulting latent watermarks achieve competitive robustness while offering similar imperceptibility and up to 20x speedup compared to pixel-space baselines. Our experiments further reveal that distilling latent watermarkers outperforms distilling pixel-space ones, providing a solution that is both more efficient and more robust. |
| title | Learning to Watermark in the Latent Space of Generative Models |
| topic | Computer Vision and Pattern Recognition Artificial Intelligence Cryptography and Security |
| url | https://arxiv.org/abs/2601.16140 |