mlsanityData to 7 Oct 2026

TabDiff: a Mixed-type Diffusion Model for Tabular Data Generation

arXiv 2410.20626 ↗First posted 27 Oct 2024cs.LGcited by 60

On the map: Diffusion models → Flow matching, time series and science

Abstract (arXiv)

Synthesizing high-quality tabular data is an important topic in many data science tasks, ranging from dataset augmentation to privacy protection. However, developing expressive generative models for tabular data is challenging due to its inherent heterogeneous data types, complex inter-correlations, and intricate column-wise distributions. In this paper, we introduce TabDiff, a joint diffusion framework that models all mixed-type distributions of tabular data in one model. Our key innovation is the development of a joint continuous-time diffusion process for numerical and categorical data, where we propose feature-wise learnable diffusion processes to counter the high disparity of different feature distributions. TabDiff is parameterized by a transformer handling different input types, and the entire framework can be efficiently optimized in an end-to-end fashion. We further introduce a mixed-type stochastic sampler to automatically correct the accumulated decoding error during sampling, and propose classifier-free guidance for conditional missing column value imputation. Comprehensive experiments on seven datasets demonstrate that TabDiff achieves superior average performance over existing competitive baselines across all eight metrics, with up to 22.5% improvement over the state-of-the-art model on pair-wise column correlation estimations. Code is available at https://github.com/MinkaiXu/TabDiff.

Key facts. Cited by 60 papers in the map (data through 7 Oct 2026). Placed in Flow matching, time series and science, one of 14 directions in Diffusion models. Builds on 27 of its 34 references that are in the map. First posted on arXiv 27 Oct 2024.

Builds on

27 of 34 references.

Closest ancestors · 7

Same direction, newest first

  1. 2023Continuous Diffusion for Mixed-Type Tabular Data16 Dec
  2. 2023Mixed-Type Tabular Data Synthesis with Score-based Diffusion in Latent Space14 Oct
  3. 2023CoDi: Co-evolving Contrastive Diffusion Models for Mixed-type Tabular Synthesis25 Apr
  4. 2022Diffusion models for missing value imputation in tabular data31 Oct
  5. 2022STaSy: Score-based Tabular data Synthesis8 Oct
  6. 2022TabDDPM: Modelling Tabular Data with Diffusion Models30 Sep
  7. 2022DiGress: Discrete Denoising diffusion for graph generation29 Sep

Other references · 20

  1. 2020Denoising Diffusion Probabilistic Modelscited by 20,851
  2. 2021High-Resolution Image Synthesis with Latent Diffusion Modelscited by 18,694
  3. 2020Score-Based Generative Modeling through Stochastic Differential Equationscited by 8,964
  4. 2021Diffusion Models Beat GANs on Image Synthesiscited by 8,185
  5. 2015Deep Unsupervised Learning using Nonequilibrium Thermodynamicscited by 7,005
  6. 2022Classifier-Free Diffusion Guidancecited by 5,717
  7. 2016XGBoost: A Scalable Tree Boosting Systemcited by 5,120
  8. 2019Generative Modeling by Estimating Gradients of the Data Distributioncited by 4,067
  9. 2022Elucidating the Design Space of Diffusion-Based Generative Modelscited by 2,994
  10. 2021Structured Denoising Diffusion Models in Discrete State-Spacescited by 1,665
  11. 2021Variational Diffusion Modelscited by 1,145
  12. 2024Simple and Effective Masked Diffusion Language Modelscited by 800
8 more
  1. 2019Modeling Tabular data using Conditional GANcited by 695
  2. 2024Simplified and Generalized Masked Diffusion for Discrete Datacited by 606
  3. 2022Language Models are Realistic Tabular Data Generatorscited by 290
  4. 2021How Faithful is your Synthetic Data? Sample-level Metrics for Evaluating and Auditing Generative Modelscited by 193
  5. 2018Data Synthesis based on Generative Adversarial Networkscited by 184
  6. 2020Handling Missing Data with Graph Representation Learningcited by 132
  7. 2023Restart Sampling for Improving Generative Processescited by 81
  8. 2022SOS: Score-based Oversampling for Tabular Datacited by 29

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2025: 18; 2026: 42.

60 papers.

Sort:

By area:

  1. 2025ResearchCodeBench: Benchmarking LLMs on Implementing Novel Machine Learning Research Codecited by 34
  2. 2025The DCR Delusion: Measuring the Privacy Risk of Synthetic Datacited by 11
  3. 2025Learning Flexible Forward Trajectories for Masked Molecular Diffusioncited by 6
  4. 2025Position: Foundation Models for Tabular Data within Systemic Contexts Need Groundingcited by 5
  5. 2025Tabby: A Language Model Architecture for Tabular and Structured Data Synthesiscited by 4
  6. 2026Cascaded Flow Matching for Heterogeneous Tabular Data with Mixed-Type Featurescited by 4
  7. 2025Graph-Conditional Flow Matching for Relational Data Generationcited by 3
  8. 2025RelDiff: Relational Data Generative Modeling with Graph-Based Diffusion Modelscited by 3
  9. 2025CausalDiffTab: Mixed-Type Causal-Aware Diffusion for Tabular Data Generationcited by 3
  10. 2025Masked Diffusion Models are Secretly Learned-Order Autoregressive Modelscited by 3
  11. 2026ExplainerPFN: Towards tabular foundation models for model-free zero-shot feature importance estimationscited by 3
  12. 2025TabTreeFormer: Tabular Data Generation Using Hybrid Tree-Transformercited by 2
38 more
  1. 2025Generative Models for Synthetic Data: Transforming Data Mining in the GenAI Eracited by 2
  2. 2026Improving TabPFN's Synthetic Data Generation by Integrating Causal Structurecited by 2
  3. 2026TabQueryBench: A Query-Centric Benchmark for Synthetic Tabular Datacited by 2
  4. 2026Measuring the Dependency Gap: Diagnosing Inter-Column Fidelity in Tabular Generative Modelscited by 2
  5. 2025A Closer Look on Memorization in Tabular Diffusion Model: A Data-Centric Perspectivecited by 1
  6. 2025SurvDiff: A Diffusion Model for Generating Synthetic Data in Survival Analysiscited by 1
  7. 2025Causal-Aware Tabular GANs with Reinforcement Learningcited by 1
  8. 2026Rethinking Anonymity Claims in Synthetic Data Generation: A Model-Centric Privacy Attack Perspectivecited by 1
  9. 2026Team, Then Trim: An Assembly-Line LLM Framework for High-Quality Tabular Data Generationcited by 1
  10. 2026TabDLM: Free-Form Tabular Data Generation via Joint Numerical-Language Diffusioncited by 1
  11. 2026A Sobering Look at Tabular Data Generation via Probabilistic Circuitscited by 1
  12. 2026TabSCM: A practical Framework for Generating Realistic Tabular Datacited by 1
  13. 2026Breaking the Quality-Privacy Tradeoff in Tabular Data Generation via In-Context Learningcited by 1
  14. 2026Inference-Time Refinement Closes the Synthetic-Real Gap in Tabular Diffusioncited by 1
  15. 2026AugMask: Training Diffusion Models on Incomplete Tabular Data via Stochastic Augmentation and Maskingcited by 1
  16. 2026GEAR: Generative Expansion and Real Anchoring for Two-Stage Distillation of Tabular Foundation Modelscited by 1
  17. 2025Critical Challenges and Guidelines in Evaluating Synthetic Tabular Data: A Systematic Reviewcited by 0
  18. 2025Disjoint Generation of Synthetic Datacited by 0
  19. 2025Entering the Era of Discrete Diffusion Models: A Benchmark for Schr"odinger Bridges and Entropic Optimal Transportcited by 0
  20. 2025Instruction Tuning of Large Language Models for Tabular Data Generation-in One Daycited by 0
  21. 2026Exploring Transformer Placement in Variational Autoencoders for Tabular Data Generationcited by 0
  22. 2026CTTVAE: Latent Space Structuring for Conditional Tabular Data Generation on Imbalanced Datasetscited by 0
  23. 2026Harpoon: Generalised Manifold Guidance for Conditional Tabular Diffusioncited by 0
  24. 2026Measuring Privacy Risks and Tradeoffs in Financial Synthetic Data Generationcited by 0
  25. 2026From Noise to Order: Learning to Rank via Denoising Diffusioncited by 0
  26. 2026Autoregressive Synthesis of Sparse and Semi-Structured Mixed-Type Datacited by 0
  27. 2026CDMT-EHR: A Continuous-Time Diffusion Framework for Generating Mixed-Type Time-Series Electronic Health Recordscited by 0
  28. 2026Synthesizing real-world distributions from high-dimensional Gaussian Noise with Fully Connected Neural Networkcited by 0
  29. 2026Self-Improving Tabular Language Models via Iterative Reward-Guided Post-Trainingcited by 0
  30. 2026On Privacy in Data-Space Tabular Diffusion Models: Influential Factors, Attacker Knowledge, and Metricscited by 0
  31. 2026Active Tabular Augmentation via Policy-Guided Diffusion Inpaintingcited by 0
  32. 2026FERMI: Exploiting Relations for Membership Inference Against Tabular Diffusion Modelscited by 0
  33. 2026Mixed neural posterior estimation for simulators with discrete and continuous parameterscited by 0
  34. 2026TabKDE: Simple and Scalable Tabular Data Generation with Kernel Density Estimatescited by 0
  35. 2026SDPM: Survival Diffusion Probabilistic Model for Continuous-Time Survival Analysiscited by 0
  36. 2026Geometry-Aware Tabular Diffusioncited by 0
  37. 2026TabSODA: Tabular Diffusion based Imputation with Skip Pattern Detection and Ordinal Awarenesscited by 0
  38. 2026Accelerating Reproducible Research in Synthetic EHR Generationcited by 0

50 most cited of 60 All 60 (JSON)

Similar, not cited

Close in wording, no citation either way

  1. 2025TabRep: Training Tabular Diffusion Models with a Simple and Effective Continuous Representation
  2. 2023AutoDiff: combining Auto-encoder and Diffusion model for tabular data synthesizing
  3. 2023MissDiff: Training Diffusion Models on Tabular Data with Missing Values
  4. 2023Generating and Imputing Tabular Data via Diffusion and Flow-based Gradient-Boosted Trees
  5. 2024Self-Supervision Improves Diffusion Models for Tabular Data Imputation
  6. 2025Diffusion and Flow Matching Models for Tabular Data: A Survey
  7. 2026Generating Benchmark Health Data Using a Tabular Diffusion Transformer
  8. 2024Balanced Mixed-Type Tabular Data Synthesis with Diffusion Models