Metraipolis › archive › dc_22ce4ce04d9e
SAC-Copula: Quality-Preserving Watermarking for Diffusion Language Models via Smooth Correlated Gumbel Fields — arXiv:2608.20839
- Source
- arXiv (en)
- Original
- https://arxiv.org/abs/2608.20839
- License (as archived)
- CC0-1.0
- Archived
- 2026-10-18T10:18:25.101Z
- SHA-256 of the archived text
b324c71e379f9a4e46827ff507544a19b6e8fee021845ca84cb353b8d6b736eb- Stable URL of this page
https://metraipolis.com/documento/dc_22ce4ce04d9e
How to cite
SAC-Copula: Quality-Preserving Watermarking for Diffusion Language Models via Smooth Correlated Gumbel Fields — arXiv:2608.20839 — arXiv (en), https://arxiv.org/abs/2608.20839 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_22ce4ce04d9e (Metraipolis archive dc_22ce4ce04d9e, sha256 b324c71e379f9a4e46827ff507544a19b6e8fee021845ca84cb353b8d6b736eb)
BibTeX
@misc{dc22ce4ce04d9e,
title = {SAC-Copula: Quality-Preserving Watermarking for Diffusion Language Models via Smooth Correlated Gumbel Fields — arXiv:2608.20839},
howpublished = {arXiv (en)},
url = {https://arxiv.org/abs/2608.20839},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_22ce4ce04d9e; sha256 b324c71e379f9a4e46827ff507544a19b6e8fee021845ca84cb353b8d6b736eb},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2608.20839. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
Watermarking diffusion language models (DLMs) requires mechanisms compatible with iterative parallel unmasking rather than autoregressive decoding. Existing sampling-based watermarking methods typically inject position-wise i.i.d. perturbations, which can be poorly aligned with DLM decoding dynamics and degrade generation quality. We propose SAC-Copula, a quality-preserving watermarking method for DLMs based on smooth, locally correlated Gumbel perturbation fields constructed via a Gaussian copula. We further develop a SAC-aware detector using covariance-aware filtering and native-sample calibration. Mechanism-level analysis shows that local correlation reduces latent perturbation roughness and better matches iterative refinement dynamics. Experiments on LLaDA show that SAC-Copula achieves a favorable quality-detectability trade-off compared with existing baselines. In particular, further evaluations on Dream-7B and additional datasets show that SAC-Copula substantially improves PPL tail stability over the i.i.d. Gumbel baseline, while maintaining strong low-FPR detectability and competitive overall generation quality. Additional token-edit stress tests further assess watermark robustness under controlled synchronization drift. Code is available at this https URL . Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-08-24 in cs.CL). Categorie arXiv: Computation and Language (cs.CL) ; Cryptography and Security (cs.CR); Machine Learning (cs.LG). Pagina del lavoro: https://arxiv.org/abs/2608.20839 — 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 2026-10-18T10:17:37.585Z dalla pagina https://arxiv.org/catchup/cs.CL/2026-08-24?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).