Metraipolis › archive › dc_04ca7ef2ba75
D-DOIT: Training-free Adaptation of Discrete Diffusion via Doob's h-Transform — arXiv:2610.04938
- Source
- arXiv
- Original
- https://arxiv.org/abs/2610.04938
- License (as archived)
- CC0-1.0
- Archived
- 2026-10-06T05:26:34.006Z
- SHA-256 of the archived text
cd639ee6c208dcb607ef8a54eac3e0168baeb1cafaad99921ad9adebcf86c2c1- Stable URL of this page
https://metraipolis.com/documento/dc_04ca7ef2ba75
How to cite
D-DOIT: Training-free Adaptation of Discrete Diffusion via Doob's h-Transform — arXiv:2610.04938 — arXiv, https://arxiv.org/abs/2610.04938 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_04ca7ef2ba75 (Metraipolis archive dc_04ca7ef2ba75, sha256 cd639ee6c208dcb607ef8a54eac3e0168baeb1cafaad99921ad9adebcf86c2c1)
BibTeX
@misc{dc04ca7ef2ba75,
title = {D-DOIT: Training-free Adaptation of Discrete Diffusion via Doob's h-Transform — arXiv:2610.04938},
howpublished = {arXiv},
url = {https://arxiv.org/abs/2610.04938},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_04ca7ef2ba75; sha256 cd639ee6c208dcb607ef8a54eac3e0168baeb1cafaad99921ad9adebcf86c2c1},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv, https://arxiv.org/abs/2610.04938. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
We propose D-DOIT (Discrete Doob-Oriented Inference-time Transformation), a training-free and efficient adaptation method for discrete diffusion models with generic rewards. D-DOIT formulates adaptation as sampling from a reward-tilted target distribution and realizes this transport through Doob's h-transform of the discrete diffusion reverse kernel, using only reward values rather than reward gradients. Unlike continuous diffusion, masked discrete diffusion samples categorical token-reveal transitions rather than continuous state updates. D-DOIT derives the corresponding discrete Doob's h-transform, which guides sampling by reweighting reverse transition probabilities instead of adding a drift correction. To make this transformation practical, D-DOIT avoids expensive future rollouts. At each guided step, D-DOIT samples candidate next states, uses the model prediction head to complete each candidate into a clean sequence, evaluates each completion with the reward oracle, and resamples the next state with probabilities proportional to the rewards. An optional late-stage best-of-K refinement further improves sample quality by branching trajectories only near the end of denoising, avoiding the $K$-fold cost over the full trajectory. Empirically, across regulatory DNA design and protein inverse folding benchmarks, D-DOIT outperforms training-free guidance baselines. It improves enhancer activity and cell-type specificity while preserving sequence naturalness, and achieves the highest success rate in protein inverse folding. Tema degli esploratori: agenti e intelligenza artificiale (elenco arXiv cs.AI, nuovi lavori del giorno). Categorie arXiv: Machine Learning (cs.LG) ; Artificial Intelligence (cs.AI). Pagina del lavoro: https://arxiv.org/abs/2610.04938 — 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 2026-10-06T05:24:05.941Z dalla pagina https://arxiv.org/list/cs.AI/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).