AI Paper Atlas

Research map / Image generation / Fast diffusion sampling

AugMask: Training Diffusion Models on Incomplete Tabular Data via Stochastic Augmentation and Masking

Jungkyu Kim, Taeyoung Park, Kibok Lee

ICML 2026 · Image generation › Fast diffusion sampling · approach: Diffusion theory and design space · 0 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
add conditional stochastic augmentation (AugFull) to missing-unaware generatorsTabSyn · MCAR 90 scenariosSummary rank: 13.7 -> 6.3Improved
add AugFull to TVAETVAE · MCAR 90 scenariosSummary rank: 17.7 -> 11.2Improved
add AugFull to ARFARF · MCAR 90 scenariosSummary rank: 18.5 -> 12.4Improved
add observed-only loss masking on top of augmentation (AugMask vs AugFull) (ablation)TabDiff · MCAR 90 scenariosSummary rank: 5.2 -> 3.2Improved
add observed-only loss masking on top of augmentation (ablation)CDTD · MCAR 90 scenariosSummary rank: 5.8 -> 3.8Improved
add observed-only loss masking on top of augmentation (ablation)TabDDPM · MCAR 90 scenariosSummary rank: 7.5 -> 7.0Improved
use stochastic instead of deterministic conditional augmentation (ablation)AugMask · Adult and othersQualityImproved
vary augmenter context bias and stochasticity (misspecification) (ablation)AugMask vs AugFull · 5x5 sweep of eta and lambdaDelta QualityImproved
cycle among cached augmentations instead of one fixed draw (ablation)LGBM-S augmentation · Adult beta-Recallbeta-Recall: 0.3305 -> 0.1553Hurt
cycle among cached augmentations (ablation)LGBM-S augmentation · Adult MLE utilityMLE: 0.8877 -> 0.8924No clear effect
mix a fixed anchor (80%) with cycling (ablation)LGBM-S augmentation · Adult beta-Recallbeta-Recall: 0.1679 -> 0.2724Mixed
use AugMask for CDTD under heavy missingnessCDTD · Adult pmiss=0.7UMAP IoU: 0.1574 -> 0.9285Improved

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