Research map / Image generation / Autoregressive and latent generation
TabDDPM: Modelling Tabular Data with Diffusion Models
Akim Kotelnikov, Dmitry Baranchuk, Ivan Rubachev, Artem Babenko
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 |
|---|---|---|---|
| Generate synthetic training data with TabDDPM rather than SMOTE | CatBoost evaluation on tabular datasets · Abalone, Adult, Buddy, California Housing, Churn Modeling | ML efficiency (R2 or F1): Abalone: 0.549 -> 0.550; Adult: 0.791 -> 0.795; Buddy: 0.891 -> 0.906; California Housing: 0.840 -> 0.836; Churn Modelin | Mixed |
| Generate synthetic data with TabDDPM rather than SMOTE | Synthetic-data privacy on Abalone · Abalone | privacy attack ROCAUC: 0.967 -> 0.505 | Improved |
| Generate synthetic data with TabDDPM rather than SMOTE | Synthetic-data privacy on Abalone · Abalone | mean DCR: 0.018 -> 0.061 | Improved |
| Use TabDDPM rather than SMOTE to model numerical feature distributions | Tabular synthetic data · Abalone | Wasserstein distance: 0.002 -> 0.005 | Hurt |
| Use CatBoost-guided hyperparameter tuning rather than MLP-guided tuning | TabDDPM · Selected tabular datasets | ML utility with MLP evaluation: The table reports dataset-specific values; for example, Abalone: 0.559 -> 0.569 and California Housing: 0.803 -> 0.809 | Mixed |
Later papers that compared with this method
6 later accepted papers report TabDDPM as a baseline in their results tables; 6 of them report a better number on most of the results they share with it.
- Cascaded Flow Matching for Heterogeneous Tabular Data with Mixed-Type Features (ICML 2026 · better on 1 of 1 shared result)
- Q-Tab: Quantized Tabular Data Generator (ICML 2026 · better on 1 of 1 shared result)
- Continuous Diffusion for Mixed-Type Tabular Data (ICLR 2025 · better on 2 of 2 shared results)
- Understanding and Mitigating Memorization in Diffusion Models for Tabular Data (ICML 2025 · better on 1 of 1 shared result)
- CTSyn: A Foundation Model for Cross Tabular Data Generation (ICLR 2025 · better on 1 of 1 shared result)
- TabMT: Generating tabular data with masked transformers (NeurIPS 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
- TabDiff: a Mixed-type Diffusion Model for Tabular Data Generation (ICLR 2025 · similarity 0.90)
- Continuous Diffusion for Mixed-Type Tabular Data (ICLR 2025 · similarity 0.90)
- D2R2: Diffusion-based Representation with Random Distance Matching for Tabular Few-shot Learning (NeurIPS 2024 · similarity 0.89)
- Understanding and Mitigating Memorization in Diffusion Models for Tabular Data (ICML 2025 · similarity 0.88)
- TabGLM: Tabular Graph Language Model for Learning Transferable Representations Through Multi-Modal Consistency Minimization (AAAI 2025 · similarity 0.87)
- TabNAT: A Continuous-Discrete Joint Generative Framework for Tabular Data (ICML 2025 · similarity 0.87)
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.
