mlsanityData to 14 Oct 2026

Flow-based Generative Modeling of Potential Outcomes and Counterfactuals

arXiv 2505.16051 ↗First posted 21 May 2025stat.ML cs.LGcited by 3

Not on the map. Why?

Abstract (arXiv)

Predicting potential and counterfactual outcomes from observational data is central to individualized decision-making, particularly in clinical settings where treatment choices must be tailored to each patient rather than guided solely by population averages. We propose PO-Flow, a continuous normalizing flow (CNF) framework for causal inference that jointly models potential outcome distributions and factual-conditioned counterfactual outcomes. Trained via flow matching, PO-Flow provides a unified approach to individualized potential outcome prediction, conditional average treatment effect estimation, and counterfactual prediction. By encoding an observed factual outcome and decoding under an alternative treatment, PO-Flow provides an encode-decode mechanism for factual-conditioned counterfactual prediction. In addition, PO-Flow supports likelihood-based evaluation of potential outcomes, enabling uncertainty-aware assessment of predictions. A supporting recovery guarantee is established under certain assumptions, and empirical results on benchmark datasets demonstrate strong performance across a range of causal inference tasks within the potential outcomes framework.

Key facts. Cited by 3 papers in the map (data through 14 Oct 2026). Builds on 26 of its 45 references that are in the map. First posted on arXiv 21 May 2025.

Builds on

26 of 45 references.

References · 26

  1. 2020Denoising Diffusion Implicit Modelscited by 9,781
  2. 2020Score-Based Generative Modeling through Stochastic Differential Equationscited by 9,122
  3. 2022Flow Matching for Generative Modelingcited by 5,702
  4. 2022Building Normalizing Flows with Stochastic Interpolantscited by 995
  5. 2017Meta-learners for Estimating Heterogeneous Treatment Effects using Machine Learningcited by 478
  6. 2017Causal Effect Inference with Deep Latent-Variable Modelscited by 410
  7. 2016Learning Representations for Counterfactual Inferencecited by 329
  8. 2020Deep Structural Causal Models for Tractable Counterfactual Inferencecited by 211
  9. 2017Bayesian Inference of Individualized Treatment Effects using Multi-task Gaussian Processescited by 158
  10. 2020Conformal Inference of Counterfactuals and Individual Treatment Effectscited by 151
  11. 2023The probability flow ODE is provably fastcited by 146
  12. 2021Nonparametric Estimation of Heterogeneous Treatment Effects: From Theory to Learning Algorithmscited by 144
14 more
  1. 2016Foundations of Structural Causal Models with Cycles and Latent Variablescited by 134
  2. 2022Causal Transformer for Estimating Counterfactual Outcomescited by 123
  3. 2021On Inductive Biases for Heterogeneous Treatment Effect Estimationcited by 75
  4. 2024Causal machine learning for predicting treatment outcomescited by 74
  5. 2020Generalization Bounds and Representation Learning for Estimation of Potential Outcomes and Causal Effectscited by 72
  6. 2023Counterfactual Identifiability of Bijective Causal Modelscited by 43
  7. 2022Normalizing flow neural networks by JKO schemecited by 39
  8. 2022Identifying Patient-Specific Root Causes with the Heteroscedastic Noise Modelcited by 38
  9. 2018Benchmarking Framework for Performance-Evaluation of Causal Inference Analysiscited by 33
  10. 2022Normalizing Flows for Interventional Density Estimationcited by 21
  11. 2024DiffPO: A causal diffusion model for learning distributions of potential outcomescited by 17
  12. 2024Meta-Learners for Partially-Identified Treatment Effects Across Multiple Environmentscited by 11
  13. 2024Annealing Flow Generative Models Towards Sampling High-Dimensional and Multi-Modal Distributionscited by 11
  14. 2022Comparison of meta-learners for estimating multi-valued treatment heterogeneous effectscited by 10

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2026: 3.

3 papers.

  1. 2026Intervention-Based Time Series Causal Discovery via Simulator-Generated Interventional Distributionscited by 0
  2. 2026Extended Wasserstein-GAN Approach to Causal Distribution Learning: Density-Free Estimation and Minimax Optimalitycited by 0
  3. 2026OrthoGen: A Generative Orthogonal Learner for Time-Varying Treatmentscited by 0

Similar, not cited

Close in wording, no citation either way

  1. 2026Debiased Counterfactual Generation via Flow Matching from Observations
  2. 2025Flow IV: Counterfactual Inference In Nonseparable Outcome Models Using Instrumental Variables
  3. 2023Counterfactual Generative Models for Time-Varying Treatments
  4. 2026DeepImagine: Clinical Trial Outcome Prediction via Stepwise Local Counterfactual Imaginations
  5. 2026Counterfactual Generation via Flow Matching: Coupling-Sensitive End-to-End Rates
  6. 2026RepFlow: Representation Enhanced Flow Matching for Causal Effect Estimation
  7. 2026Causal Diffusion Models for Counterfactual Outcome Distributions in Longitudinal Data
  8. 2026Causal State-Space Model for Causal Inference: Estimating Longitudinal Individual Treatment Effects