mlsanityData to 14 Oct 2026

CAFE: Counterfactual Prediction via Fast Posterior Estimation

arXiv 2609.32167 ↗First posted 26 Sep 2026cs.LGcited by 0

Not on the map. Why?

Abstract (arXiv)

Counterfactual prediction estimates an individual's outcome under an alternative intervention given their factual observations. Such outcomes are generally not identifiable from observational data without additional assumptions. Even within the class of fully observed additive noise models (ANMs), different causal graphs can generate the same observational distribution yet imply different individual counterfactual outcomes. Predictions based on a single estimated graph ignore this structural uncertainty. We therefore target a Bayesian counterfactual posterior predictive distribution that combines predictions from plausible SCMs. We introduce CAFE (CounterfActual Prediction via Fast Posterior Estimation), an amortized inference framework that directly approximates the Bayesian counterfactual posterior predictive distribution. We pretrain a transformer-based model on synthetic counterfactual tasks generated from a diverse prior over ANMs. Given an observational dataset, an individual's factual observations, and an intervention, CAFE approximates the corresponding posterior predictive distribution in a single forward pass. Experiments show that CAFE accurately predicts individual counterfactual outcomes in identifiable settings and approximates the posterior predictive distribution when structural uncertainty induced by observationally indistinguishable causal graphs exists. Strong performance in realistic manufacturing and viticulture settings further demonstrates its empirical robustness beyond the assumptions of the training prior.

Key facts. No citing papers in the map yet (data through 14 Oct 2026). Builds on 32 of its 46 references that are in the map. First posted on arXiv 26 Sep 2026.

Builds on

32 of 46 references.

References · 32

  1. 2013Causal Discovery with Continuous Additive Noise Modelscited by 255
  2. 2020Deep Structural Causal Models for Tractable Counterfactual Inferencecited by 211
  3. 2020Algorithmic recourse under imperfect causal knowledge: a probabilistic approachcited by 140
  4. 2025TabPFN-2.5: Advancing the State of the Art in Tabular Foundation Modelscited by 132
  5. 2019Counterfactual Off-Policy Evaluation with Gumbel-Max Structural Causal Modelscited by 117
  6. 2021DiBS: Differentiable Bayesian Structure Learningcited by 106
  7. 2020Causal Autoregressive Flowscited by 99
  8. 2022Deep End-to-end Causal Inferencecited by 95
  9. 2019Learning Counterfactual Representations for Estimating Individual Dose-Response Curvescited by 90
  10. 2025LimiX: Unleashing Structured-Data Modeling Capability for Generalist Intelligencecited by 79
  11. 2020Estimating the Effects of Continuous-valued Interventions using Generative Adversarial Networkscited by 78
  12. 2023Causal normalizing flows: from theory to practicecited by 51
20 more
  1. 2025Do-PFN: In-Context Learning for Causal Effect Estimationcited by 51
  2. 2025CausalPFN: Amortized Causal Effect Estimation via In-Context Learningcited by 49
  3. 2025Foundation Models for Causal Inference via Prior-Data Fitted Networkscited by 32
  4. 2023Modeling Causal Mechanisms with Diffusion Models for Interventional and Counterfactual Queriescited by 24
  5. 2023Self-Compatibility: Evaluating Causal Discovery without Ground Truthcited by 22
  6. 2021Counterfactual Explanations as Interventions in Latent Spacecited by 18
  7. 2023Counterfactual (Non-)identifiability of Learned Structural Causal Modelscited by 17
  8. 2025Estimating Interventional Distributions with Uncertain Causal Graphs through Meta-Learningcited by 13
  9. 2024Counterfactual Generative Modeling with Variational Causal Inferencecited by 11
  10. 2024A Fixed-Point Approach for Causal Generative Modelingcited by 7
  11. 2026Use What You Know: Causal Foundation Models with Partial Graphscited by 7
  12. 2013Using proper divergence functions to evaluate climate modelscited by 5
  13. 2024Hybrid Top-Down Global Causal Discovery with Local Search for Linear and Nonlinear Additive Noise Modelscited by 5
  14. 2024Amortized Inference of Causal Models via Conditional Fixed-Point Iterationscited by 4
  15. 2025CausalMan: A physics-based simulator for large-scale causalitycited by 4
  16. 2026Causal Foundation Models with Continuous Treatmentscited by 4
  17. 2015Causality on Cross-Sectional Data: Stable Specification Search in Constrained Structural Equation Modelingcited by 3
  18. 2026A Causal Foundation Model for Structure and Outcome Predictioncited by 2
  19. 2026Retrospective Counterfactual Prediction by Conditioning on the Factual Outcome: A Cross-World Approachcited by 1
  20. 2026Counterfactual Residual Data Augmentation for Regressioncited by 1

Built on by

Not cited yet.