mlsanityData to 7 Oct 2026

DiffPuter: Empowering Diffusion Models for Missing Data Imputation

arXiv 2405.20690 ↗First posted 31 May 2024cs.LGcited by 17

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

Abstract (arXiv)

Generative models play an important role in missing data imputation in that they aim to learn the joint distribution of full data. However, applying advanced deep generative models (such as Diffusion models) to missing data imputation is challenging due to 1) the inherent incompleteness of the training data and 2) the difficulty in performing conditional inference from unconditional generative models. To deal with these challenges, this paper introduces DiffPuter, a tailored diffusion model combined with the Expectation-Maximization (EM) algorithm for missing data imputation. DiffPuter iteratively trains a diffusion model to learn the joint distribution of missing and observed data and performs an accurate conditional sampling to update the missing values using a tailored reversed sampling strategy. Our theoretical analysis shows that DiffPuter's training step corresponds to the maximum likelihood estimation of data density (M-step), and its sampling step represents the Expected A Posteriori estimation of missing values (E-step). Extensive experiments across ten diverse datasets and comparisons with 17 different imputation methods demonstrate DiffPuter's superior performance. Notably, DiffPuter achieves an average improvement of 6.94% in MAE and 4.78% in RMSE compared to the most competitive existing method.

Key facts. Cited by 17 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 26 of its 42 references that are in the map. First posted on arXiv 31 May 2024.

Builds on

26 of 42 references.

Closest ancestors · 4

Same direction, newest first

  1. 2023Mixed-Type Tabular Data Synthesis with Score-based Diffusion in Latent Space14 Oct
  2. 2023MissDiff: Training Diffusion Models on Tabular Data with Missing Values2 Jul
  3. 2022Diffusion models for missing value imputation in tabular data31 Oct
  4. 2022TabDDPM: Modelling Tabular Data with Diffusion Models30 Sep

Other references · 22

  1. 2020Denoising Diffusion Probabilistic Modelscited by 20,851
  2. 2019PyTorch: An Imperative Style, High-Performance Deep Learning Librarycited by 20,160
  3. 2020Score-Based Generative Modeling through Stochastic Differential Equationscited by 8,964
  4. 2022Elucidating the Design Space of Diffusion-Based Generative Modelscited by 2,994
  5. 2015Variational Inference with Normalizing Flowscited by 2,161
  6. 2022RePaint: Inpainting using Denoising Diffusion Probabilistic Modelscited by 1,419
  7. 2021Maximum Likelihood Training of Score-Based Diffusion Modelscited by 670
  8. 2018GAIN: Missing Data Imputation using Generative Adversarial Netscited by 349
  9. 2018Handling Incomplete Heterogeneous Data using VAEscited by 148
  10. 2020Handling Missing Data with Graph Representation Learningcited by 132
  11. 2014Matrix Completion and Low-Rank SVD via Fast Alternating Least Squarescited by 104
  12. 2020Missing Data Imputation using Optimal Transportcited by 99
10 more
  1. 2018EDDI: Efficient Dynamic Discovery of High-Value Information with Partial VAEcited by 85
  2. 2022HyperImpute: Generalized Iterative Imputation with Automatic Model Selectioncited by 81
  3. 2020VAEM: a Deep Generative Model for Heterogeneous Mixed Type Datacited by 66
  4. 2023Generating and Imputing Tabular Data via Diffusion and Flow-based Gradient-Boosted Treescited by 60
  5. 2021MIRACLE: Causally-Aware Imputation via Learning Missing Data Mechanismscited by 50
  6. 2020MCFlow: Monte Carlo Flow Models for Data Imputationcited by 37
  7. 2023ReMasker: Imputing Tabular Data with Masked Autoencodingcited by 35
  8. 2023Transformed Distribution Matching for Missing Value Imputationcited by 31
  9. 2022Missing Data Imputation and Acquisition with Deep Hierarchical Models and Hamiltonian Monte Carlocited by 19
  10. 2022Data Imputation with Iterative Graph Reconstructioncited by 15

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2024: 1; 2025: 8; 2026: 8.

17 papers.

Sort:

By area:

  1. 2025Beyond Log-Concavity and Score Regularity: Improved Convergence Bounds for Score-Based Generative Models in W2-distancecited by 27
  2. 2024Diffusion-nested Auto-Regressive Synthesis of Heterogeneous Tabular Datacited by 4
  3. 2025Incomplete Data, Complete Dynamics: A Diffusion Approachcited by 2
  4. 2025One Pipeline, Many Transformers: Pattern-Specific Imputation Specialists for Tabular Missing Datacited by 2
  5. 2025Interpretable Generative and Discriminative Learning for Multimodal and Incomplete Clinical Datacited by 2
  6. 2025DeepIFSAC: Deep Imputation of Missing Values Using Feature and Sample Attention within Contrastive Frameworkcited by 1
  7. 2026CoreFlow: Low-Rank Matrix Generative Modelscited by 1
  8. 2025Imputation-free Learning of Tabular Data with Missing Values using Incremental Feature Partitions in Transformercited by 0
  9. 2025IMAGIC-500: IMputation benchmark on A Generative Imaginary Country (500k samples)cited by 0
  10. 2025An Interdisciplinary and Cross-Task Review on Missing Data Imputationcited by 0
  11. 2026Nearly Optimal Bayesian Inference for Structural Missingnesscited by 0
  12. 2026Order-Agnostic Autoregressive Modelling with Missing Datacited by 0
5 more
  1. 2026Missing Pattern Recognized Diffusion Imputation Model for Missing Not At Randomcited by 0
  2. 2026Latent Diffusion for Missing Datacited by 0
  3. 2026Learning What Not to Impute: An Uncertainty-Aware Diffusion Framework for Meaningful Missingnesscited by 0
  4. 2026TabSODA: Tabular Diffusion based Imputation with Skip Pattern Detection and Ordinal Awarenesscited by 0
  5. 2026Imputation Meets Clustering: Exploiting Latent Subgroup Structure for Missing Data Recoverycited by 0

Similar, not cited

Close in wording, no citation either way

  1. 2025Diffusion Transformers for Imputation: Statistical Efficiency and Uncertainty Quantification
  2. 2024Diffusion Models for Tabular Data Imputation and Synthetic Data Generation
  3. 2026Impute-EM: Native Mixed-State Diffusion Models for Heterogeneous Data Imputation
  4. 2024DiffImpute: Tabular Data Imputation With Denoising Diffusion Probabilistic Model
  5. 2025MissDDIM: Deterministic and Efficient Conditional Diffusion for Tabular Data Imputation
  6. 2024Self-Supervision Improves Diffusion Models for Tabular Data Imputation
  7. 2025MissHDD: Hybrid Deterministic Diffusion for Hetrogeneous Incomplete Data Imputation
  8. 2025RefiDiff: Progressive Refinement Diffusion for Efficient Missing Data Imputation