AI Paper Atlas

Research map / Image generation / Fast diffusion sampling

Unified Discrete Diffusion for Categorical Data

Lingxiao Zhao, Xueying Ding, Lijun Yu, Leman Akoglu

JMLR 2025 · Image generation › Fast diffusion sampling · approach: Discrete diffusion solvers · 26 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
combine VLB and CE losses instead of CE only (ablation)USD3 (discrete-time) · Lakh Pianoroll Piano3-gram Hellinger distance: 0.574 -> 0.552Improved
combine VLB and CE losses instead of VLB only (ablation)USD3 (discrete-time) · Lakh Pianoroll Piano-P3-gram train-to-test ratio: 2.863 -> 3.083Improved
combine VLB and CE losses instead of CE only (ablation)USD3 (continuous-time) · Lakh Pianoroll Piano-P3-gram train-to-test ratio: 2.637 -> 3.083Improved
use continuous-time instead of discrete-time trainingUSD3 · Lakh Pianoroll Piano3-gram Hellinger distance: 0.552 -> 0.543Improved
use CE only loss instead of VLB (ablation)USD3 (continuous-time) · Lakh Pianoroll Piano3-gram Prop. of Out.: 0.270 -> 0.346Hurt
use USD3 loss instead of D3PM lossdiscrete-time diffusion · VQCIFAR10FID: 18.08 -> 13.25Improved
add CE-only USD3 loss instead of combined VLB+CE (ablation)USD3 (continuous-time) · VQCIFAR10FID: 13.09 -> 11.97Improved
use continuous-time instead of discrete-time trainingUSD3* · VQCIFAR10FID: 12.07 -> 14.03Hurt

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.

Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy