Research map / Image generation / Autoregressive and latent generation
TabMT: Generating tabular data with masked transformers
Manbir S Gulati, Paul F Roysdon
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 changed | Where | Measured | Effect |
|---|---|---|---|
| Use masked generation rather than the strongest listed prior generators | TabMT · 15 tabular datasets | MLE | Mixed |
| Tune per-field temperature to increase privacy | TabMT · AB | DCR: 0.050 -> 0.249 | Improved |
| Tune per-field temperature to increase privacy | TabMT · AB | MLE: 0.550 -> 0.533 | No clear effect |
| Train with 25% of values missing | TabMT · AD | MLE: -0.001 | No clear effect |
| Train with 25% of values missing | TabMT · KI | MLE: -0.008 | Hurt |
| Increase model size from TabMT-S to TabMT-L | TabMT · CIDDS-001 | Diversity(%): 89.81 -> 99.43 | Improved |
| Increase model size from TabMT-S to TabMT-L | TabMT · CIDDS-001 | Precision(%): 82.58 -> 88.10 | Improved |
| Increase model size from TabMT-S to TabMT-L | TabMT · CIDDS-001 | Recall(%): 91.88 -> 91.12 | No clear effect |
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.
- TabDDPM — TabDDPM: Modelling Tabular Data with Diffusion Models (ICML 2023) · better on 8 of 12 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
- Masked Language Modeling Becomes Conditional Density Estimation for Tabular Data Synthesis (AAAI 2025 · similarity 0.89)
- HARMONIC: Harnessing LLMs for Tabular Data Synthesis and Privacy Protection (NeurIPS 2024 · similarity 0.89)
- CuTS: Customizable Tabular Synthetic Data Generation (ICML 2024 · similarity 0.89)
- TabNAT: A Continuous-Discrete Joint Generative Framework for Tabular Data (ICML 2025 · similarity 0.89)
- TabSDS: a Lightweight, Fully Non-Parametric, and Model Free Approach for Generating Synthetic Tabular Data (ICML 2025 · similarity 0.88)
- Breaking the Quality–Privacy Tradeoff in Tabular Data Generation via In-Context Learning (NeurIPS 2026 · similarity 0.88)
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.
