AI Paper Atlas

Research map / Image generation / Autoregressive and latent generation

Q-Tab: Quantized Tabular Data Generator

Julian Wustl, Philipp Haid, Yarema Okhrin, Claudius Schnörr

ICML 2026 · Image generation › Autoregressive and latent generation · approach: Generative latent space tooling

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 residual LFQ with residual-only bit-flip corruption instead of flat LFQ with global corruption (ablation)Q-Tab tokenizer · Shoppers/Beijing/Default/Adult meanMLE gap (%): 11.24 -> 7.80Improved
add bit-flip corruption to flat LFQ (ablation)LFQ tokenizer · mean over 4 datasetsduplicate rate (%): 30.54 -> 9.60Improved
add global bit-flip corruption to flat LFQ (ablation)LFQ tokenizer · mean over 4 datasetsMLE gap (%): 7.99 -> 11.24Hurt
use residual corruption versus none (no corruption LFQ) (ablation)Q-Tab · mean over 4 datasetsMLE gap (%): 7.99 -> 7.80No clear effect
sample categorical tokens first (Cat->Num) instead of numerical first (ablation)Q-Tab sampler · Shoppers, BeijingDCR: 0.6176 -> 0.534Improved
sample categorical tokens first instead of uniform random order (ablation)Q-Tab sampler · Shoppers, Beijingquality: 0.979 -> 0.982Improved
change sampling order (ablation)Q-Tab sampler · Shoppers, BeijingMLE (AUC/RMSE): AUC 0.919 / 0.9176 / 0.919No clear effect
increase LFQ bit size B beyond 3-4 (ablation)Q-Tab · sensitivity analysisutility vs duplicatesMixed
increase bit-flip corruption epsilon above 0.05 (ablation)Q-Tab · sensitivity analysisAUCROCHurt
add moderate MLM label smoothing on residual tokens (ablation)Q-Tab · sensitivity analysisweighted-F1Improved
use Q-Tab instead of TabDiff for MLEQ-Tab · Average gap to real dataaverage gap (%): 6.0 -> 4.6Improved
use Q-Tab instead of TabDiff for distributional fidelityQ-Tab · Average DF gapDF gap (%): 2.0 -> 2.2No clear effect

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 2 of 3 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