Metraipolis › archive › dc_4198dbf01925
Score Mismatch and Set-Constrained Sampling in Zero-Shot Conditional Discrete Diffusion — arXiv:2610.21061
- Source
- arXiv (en)
- Original
- https://arxiv.org/abs/2610.21061
- License (as archived)
- CC0-1.0
- Archived
- 2026-10-16T10:25:10.323Z
- SHA-256 of the archived text
574ab6ea0f5ed8736dc4a73397ecfe655038b3bbcab9bbd16d06b06a2635e395- Stable URL of this page
https://metraipolis.com/documento/dc_4198dbf01925
How to cite
Score Mismatch and Set-Constrained Sampling in Zero-Shot Conditional Discrete Diffusion — arXiv:2610.21061 — arXiv (en), https://arxiv.org/abs/2610.21061 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_4198dbf01925 (Metraipolis archive dc_4198dbf01925, sha256 574ab6ea0f5ed8736dc4a73397ecfe655038b3bbcab9bbd16d06b06a2635e395)
BibTeX
@misc{dc4198dbf01925,
title = {Score Mismatch and Set-Constrained Sampling in Zero-Shot Conditional Discrete Diffusion — arXiv:2610.21061},
howpublished = {arXiv (en)},
url = {https://arxiv.org/abs/2610.21061},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_4198dbf01925; sha256 574ab6ea0f5ed8736dc4a73397ecfe655038b3bbcab9bbd16d06b06a2635e395},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2610.21061. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
Discrete diffusion models have become a powerful framework for modeling complex discrete distributions across a range of applications. While theoretical convergence guarantees depend on consistency between the target and training distributions, such a case rarely holds in practice, making score mismatch a central challenge in discrete diffusion sampling. This is particularly relevant to zero-shot conditional generation, where a model trained on the unconditional distribution is directly reused at inference time to sample from a conditional target distribution. In this paper, we provide the first theoretical analysis for score-mismatched discrete diffusion models and derive a general KL divergence upper bound that clearly identifies the additional bias caused by such a mismatch. Our analysis covers partial-observation constraints, including text infilling, sentence completion, and more general coordinatewise set constraints. Specifically, we analyze the classical Conditional SEDD sampler and show that it can incur a bias term that scales with the dimension, vocabulary size, and compatibility. Based on this analysis, we propose a Randomized Hard Clamping (RHC) sampler for coordinatewise set constraints. We show that the assignment probabilities can be expressed exactly by the unconditional scores with absorbing rate. Finally, numerical and real-data experiments demonstrate the effectiveness of the proposed RHC sampler. Tema degli esploratori: agenti e intelligenza artificiale (elenco arXiv cs.LG, nuovi lavori del giorno). Categorie arXiv: Machine Learning (stat.ML) ; Machine Learning (cs.LG). Pagina del lavoro: https://arxiv.org/abs/2610.21061 — 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 2026-10-16T08:48:02.883Z dalla pagina https://arxiv.org/list/cs.LG/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).