Metraipolis › archive › dc_bf641cdd8548
Restrained Diffusion Bridges: Targeted Denoiser Training and Refinement — arXiv:2611.02170
- Source
- arXiv (en)
- Original
- https://arxiv.org/abs/2611.02170
- License (as archived)
- CC0-1.0
- Archived
- 2026-11-03T18:37:12.829Z
- SHA-256 of the archived text
293d57a91d55bd5cba98902217ad1eadc792fb2d23f345de814dab85f58558df- Stable URL of this page
https://metraipolis.com/documento/dc_bf641cdd8548
How to cite
Restrained Diffusion Bridges: Targeted Denoiser Training and Refinement — arXiv:2611.02170 — arXiv (en), https://arxiv.org/abs/2611.02170 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_bf641cdd8548 (Metraipolis archive dc_bf641cdd8548, sha256 293d57a91d55bd5cba98902217ad1eadc792fb2d23f345de814dab85f58558df)
BibTeX
@misc{dcbf641cdd8548,
title = {Restrained Diffusion Bridges: Targeted Denoiser Training and Refinement — arXiv:2611.02170},
howpublished = {arXiv (en)},
url = {https://arxiv.org/abs/2611.02170},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_bf641cdd8548; sha256 293d57a91d55bd5cba98902217ad1eadc792fb2d23f345de814dab85f58558df},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2611.02170. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
Diffusion models are often inaccurate in critical, data-sparse regions of the target distribution. Sampling from restrained distributions, biased distributions used in classical enhanced sampling to concentrate data near chosen configurations, can control where training data are collected but biases the learned distribution. We establish that this bias cancels exactly when forward noising from restrained data is conditioned on the restrained configuration at an appropriately chosen intermediate time. This allows a pretrained model fine-tuned on restrained data to achieve high accuracy in any chosen region of the target distribution, regardless of its probability, using standard denoising or flow matching objectives without importance weights. When no target data are available, Gibbs sampling enables inference from a model trained exclusively on restrained data. Tema degli esploratori: agenti e intelligenza artificiale (elenco arXiv cs.LG, nuovi lavori del giorno). Categorie arXiv: Machine Learning (cs.LG) ; Chemical Physics (physics.chem-ph). Pagina del lavoro: https://arxiv.org/abs/2611.02170 — 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 2026-11-03T18:12:54.549Z dalla pagina https://arxiv.org/list/cs.LG/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).