Metraipolis › archive › dc_b63fcd26e266
Learning to Re-Draft: A Variational Stackelberg Game for Discrete Diffusion — arXiv:2609.35166
- Source
- arXiv (en)
- Original
- https://arxiv.org/abs/2609.35166
- License (as archived)
- CC0-1.0
- Archived
- 2026-10-07T21:03:18.939Z
- SHA-256 of the archived text
25d7f050436032ae4c21c4a16ad204d5c92d747f887e5594f158d5c57afed04b- Stable URL of this page
https://metraipolis.com/documento/dc_b63fcd26e266
How to cite
Learning to Re-Draft: A Variational Stackelberg Game for Discrete Diffusion — arXiv:2609.35166 — arXiv (en), https://arxiv.org/abs/2609.35166 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_b63fcd26e266 (Metraipolis archive dc_b63fcd26e266, sha256 25d7f050436032ae4c21c4a16ad204d5c92d747f887e5594f158d5c57afed04b)
BibTeX
@misc{dcb63fcd26e266,
title = {Learning to Re-Draft: A Variational Stackelberg Game for Discrete Diffusion — arXiv:2609.35166},
howpublished = {arXiv (en)},
url = {https://arxiv.org/abs/2609.35166},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_b63fcd26e266; sha256 25d7f050436032ae4c21c4a16ad204d5c92d747f887e5594f158d5c57afed04b},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.35166. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
Discrete diffusion models offer the ability to re-draft, revisiting and correcting earlier tokens throughout generation. This capability depends on the forward corruption process that defines what the denoiser learns to correct. Masked diffusion models fix tokens once they are unmasked, while uniform diffusion permits revisions but relies on uniformly random token substitutions. We instead learn which substitutions are most useful for training the denoiser to re-draft. We introduce Variational Stackelberg Discrete Diffusion (VSDD), a framework for learning a semantically aware corruption process. VSDD formulates training as a leader-follower game: the leader defines a Markovian corruption process parameterized by the denoiser's token embeddings, while the follower optimizes a variational denoising objective with the corruption process held fixed. The leader rewards corruptions based on how much the denoiser improves after learning from them, rather than on how easily the current denoiser can reconstruct them. We measure this improvement under a fixed reference corruption process, approximate the follower's response with a one-step gradient update, and optimize the leader using a score-function estimator. We evaluate VSDD across molecular, text, and playlist generation. VSDD substantially improves molecular validity over uniform and masked diffusion, reduces text perplexity relative to uniform diffusion while remaining competitive with masked diffusion, and achieves sizable improvements in offline playlist recommendation metrics. Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-09-29 in cs.AI). Categorie arXiv: Machine Learning (cs.LG) ; Artificial Intelligence (cs.AI). Pagina del lavoro: https://arxiv.org/abs/2609.35166 — 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 2026-10-07T20:29:03.271Z dalla pagina https://arxiv.org/catchup/cs.AI/2026-09-29?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).