Metraipolis › archive › dc_d23d453e5cd3

One Prompt Is Enough: Watermark Laundering Through Foundation Image Models — arXiv:2609.01249

Source
arXiv (en)
Original
https://arxiv.org/abs/2609.01249
License (as archived)
CC0-1.0
Archived
2026-10-17T06:34:02.638Z
SHA-256 of the archived text
281211cc8d39f47ca4e23b6da8bf855f376c336aba5260d4c54201f78ae96a73
Stable URL of this page
https://metraipolis.com/documento/dc_d23d453e5cd3

How to cite

One Prompt Is Enough: Watermark Laundering Through Foundation Image Models — arXiv:2609.01249 — arXiv (en), https://arxiv.org/abs/2609.01249 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_d23d453e5cd3 (Metraipolis archive dc_d23d453e5cd3, sha256 281211cc8d39f47ca4e23b6da8bf855f376c336aba5260d4c54201f78ae96a73)
BibTeX
@misc{dcd23d453e5cd3,
  title = {One Prompt Is Enough: Watermark Laundering Through Foundation Image Models — arXiv:2609.01249},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2609.01249},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_d23d453e5cd3; sha256 281211cc8d39f47ca4e23b6da8bf855f376c336aba5260d4c54201f78ae96a73},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.01249. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Invisible watermarks are typically evaluated against predefined perturbations such as compression, blur, noise, cropping, and denoising. Public foundation image models expose a distinct threat: an attacker can submit a watermarked image with a single reconstruction prompt and obtain a visually faithful output from which the invisible watermark can no longer be decoded reliably. We formalize this failure mode as watermark laundering and evaluate it using a joint payload-fidelity profile that combines bit error rate (BER) with visual and semantic preservation. Across six OpenAI and Google image editing models, three representative watermarking schemes, and 1,800 reconstructed outputs, we identify two complementary laundering regimes: OpenAI models produce the strongest payload disruption across the evaluated schemes, whereas Nano Banana 2 shows that DwtDct remains vulnerable under high-fidelity reconstruction. Prompt ablations show that no single removal-oriented instruction is necessary for payload disruption, indicating that the effect is primarily induced by the reconstruction pathway rather than by explicit attack wording. Comparisons with conventional attacks further show that prompt-conditioned reconstruction constitutes a distinct operational attack interface. These findings motivate foundation-model reconstruction as a missing robustness condition in invisible watermark evaluation.

Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-09-02 in cs.AI).
Categorie arXiv: Computer Vision and Pattern Recognition (cs.CV) ; Artificial Intelligence (cs.AI); Cryptography and Security (cs.CR).
Pagina del lavoro: https://arxiv.org/abs/2609.01249 — autori e testo completo sono lì (qui omessi i nomi: nessun dato personale nel deposito).
Licenza: titolo e abstract sono metadati arXiv distribuiti in CC0 1.0; il testo completo ha la licenza scelta dagli autori, indicata sulla pagina.
Letto il 2026-10-17T06:15:12.877Z dalla pagina https://arxiv.org/catchup/cs.AI/2026-09-02?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).