{"ID":23507353,"CreatedAt":"2026-09-18T02:21:44.056544415Z","UpdatedAt":"2026-09-20T18:11:56.143995915Z","DeletedAt":null,"paper_url":"https://arxiv.org/abs/2609.20581","arxiv_id":"2609.20581","title":"Limits of Confidence in Diffusion","abstract":"Discrete diffusion, including remasking and uniform-state samplers, generate a sequence by writing multiple token positions per step, drawing each from a per-position distribution and choosing which positions to write from those same distributions. For domains of general interest (pixels, phonemes, or words) there are inherent dependencies between tokens. We show that a step matches the training distribution only when the positions it writes are conditionally independent given the tokens already fixed, that no product of per-position distributions can match a dependent group, and that per-position distributions do not determine whether a group is dependent: two joint distributions can have identical per-position marginals while differing in which combinations of values occur. On ScanAndAdd, a synthetic task whose joint distribution is available in closed form, we verify that every group of two or more undetermined positions a confidence ranking writes is dependent, and measure the generated distribution to be $29\\times$ the sampling-noise floor total variation while per-sample metrics are $1.0$.","short_abstract":"Discrete diffusion, including remasking and uniform-state samplers, generate a sequence by writing multiple token positions per step, drawing each from a per-position distribution and choosing which positions to write from those same distributions. For domains of general interest (pixels, phonemes, or words) there are...","url_abs":"https://arxiv.org/abs/2609.20581","url_pdf":"https://arxiv.org/pdf/2609.20581v1","authors":"[\"Russ Webb\",\"Amitis Shidani\",\"Alice Bizeul\",\"Dan Busbridge\"]","published":"2026-09-17T15:36:53Z","proceeding":"cs.AI","tasks":"[\"cs.AI\"]","methods":"[\"Diffusion Model\"]","has_code":false}