Pith. sign in

REVIEW 2 cited by

Diffusion on the Probability Simplex

Not yet reviewed by Pith; the record is open.

This paper has not been read by Pith yet. Machine review is queued; the pith claim, tier, and objections will appear here once it completes.

SPECIMEN: schema-true, not a live event

T0 review · schema-true

One-sentence machine reading of the paper's core claim.

pith:XXXXXXXX · record.json · timestamp

arxiv 2309.02530 v2 pith:INEJBBGU submitted 2023-09-05 cs.LG stat.ML

classification cs.LGstat.ML
keywords diffusionprobabilitysimplexcontinuousdatadiscretemethodnaturally
verification ladder T0 review T1 audit T2 compute T3 formal
0 comments
read the original abstract

Diffusion models learn to reverse the progressive noising of a data distribution to create a generative model. However, the desired continuous nature of the noising process can be at odds with discrete data. To deal with this tension between continuous and discrete objects, we propose a method of performing diffusion on the probability simplex. Using the probability simplex naturally creates an interpretation where points correspond to categorical probability distributions. Our method uses the softmax function applied to an Ornstein-Unlenbeck Process, a well-known stochastic differential equation. We find that our methodology also naturally extends to include diffusion on the unit cube which has applications for bounded image generation.

Discussion (0). Continue with ORCID to comment.

Forward citations

Cited by 2 Pith papers

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score. Full citation record

  1. TFG-Flow: Training-free Guidance in Multimodal Generative Flow

    cs.LG 2025-01 conditional novelty 7.0 of 10

    TFG-Flow guides multimodal flow models at inference time by weighted Monte Carlo sampling for discrete atom types and gradient ascent for continuous coordinates, improving targeted molecular generation without extra training.

  2. Solving Inverse Problems via Diffusion-Based Priors: An Approximation-Free Ensemble Sampling Approach

    cs.LG 2025-06 conditional novelty 5.0 of 10

    A weighted-particle sampler evolves the posterior through the diffusion model's reverse dynamics, with theoretical error bounds and improved image reconstructions.

Pith tools