mlsanityData to 7 Oct 2026

Measuring axiomatic soundness of counterfactual image models

arXiv 2303.01274 ↗First posted 2 Mar 2023cs.CV cs.LGcited by 34

Not on the map. Why?

Abstract (arXiv)

We present a general framework for evaluating image counterfactuals. The power and flexibility of deep generative models make them valuable tools for learning mechanisms in structural causal models. However, their flexibility makes counterfactual identifiability impossible in the general case. Motivated by these issues, we revisit Pearl's axiomatic definition of counterfactuals to determine the necessary constraints of any counterfactual inference model: composition, reversibility, and effectiveness. We frame counterfactuals as functions of an input variable, its parents, and counterfactual parents and use the axiomatic constraints to restrict the set of functions that could represent the counterfactual, thus deriving distance metrics between the approximate and ideal functions. We demonstrate how these metrics can be used to compare and choose between different approximate counterfactual inference models and to provide insight into a model's shortcomings and trade-offs.

Key facts. Cited by 34 papers in the map (data through 7 Oct 2026). Builds on 41 of its 68 references that are in the map. First posted on arXiv 2 Mar 2023.

Builds on

41 of 68 references.

References · 41

  1. 2020Denoising Diffusion Probabilistic Modelscited by 20,851
  2. 2019PyTorch: An Imperative Style, High-Performance Deep Learning Librarycited by 20,160
  3. 2016Image-to-Image Translation with Conditional Adversarial Networkscited by 4,510
  4. 2013Auto-Encoding Variational Bayescited by 3,980
  5. 2017Progressive Growing of GANs for Improved Quality, Stability, and Variationcited by 2,967
  6. 2020Shortcut Learning in Deep Neural Networkscited by 1,928
  7. 2019Normalizing Flows for Probabilistic Modeling and Inferencecited by 1,315
  8. 2017CyCADA: Cycle-Consistent Adversarial Domain Adaptationcited by 1,079
  9. 2017StarGAN: Unified Generative Adversarial Networks for Multi-Domain Image-to-Image Translationcited by 1,036
  10. 2016Improving Variational Inference with Inverse Autoregressive Flowcited by 636
  11. 2018Disentangling by Factorisingcited by 614
  12. 2017Unsupervised Image-to-Image Translation Networkscited by 602
29 more
  1. 2016f-GAN: Training Generative Neural Samplers using Variational Divergence Minimizationcited by 504
  2. 2019Variational Autoencoders and Nonlinear ICA: A Unifying Frameworkcited by 494
  3. 2017PixelCNN++: Improving the PixelCNN with Discretized Logistic Mixture Likelihood and Other Modificationscited by 415
  4. 2017Causal Effect Inference with Deep Latent-Variable Modelscited by 408
  5. 2016Ladder Variational Autoencoderscited by 315
  6. 2021Self-Supervised Learning with Data Augmentations Provably Isolates Content from Stylecited by 308
  7. 2019Adapting Neural Networks for the Estimation of Treatment Effectscited by 285
  8. 2020Very Deep VAEs Generalize Autoregressive Models and Can Outperform Them on Imagescited by 261
  9. 2018Learning Not to Learn: Training Deep Neural Networks with Biased Datacited by 255
  10. 2017Variational Inference of Disentangled Latent Concepts from Unlabeled Observationscited by 240
  11. 2020Weakly-Supervised Disentanglement Without Compromisescited by 237
  12. 2019Interpretable Counterfactual Explanations Guided by Prototypescited by 223
  13. 2020Deep Structural Causal Models for Tractable Counterfactual Inferencecited by 210
  14. 2021Semantic Segmentation with Generative Models: Semi-Supervised Learning and Strong Out-of-Domain Generalizationcited by 141
  15. 2021Counterfactual Generative Networkscited by 111
  16. 2019BIVA: A Very Deep Hierarchy of Latent Variables for Generative Modelingcited by 109
  17. 2017Learning Independent Causal Mechanismscited by 104
  18. 2017Multi-Level Variational Autoencoder: Learning Disentangled Representations from Grouped Observationscited by 100
  19. 2020The Hessian Penalty: A Weak Prior for Unsupervised Disentanglementcited by 90
  20. 2022Diffusion Causal Models for Counterfactual Estimationcited by 87
  21. 2019Weakly Supervised Disentanglement with Guaranteescited by 82
  22. 2020Identifying Causal-Effect Inference Failure with Uncertainty-Aware Modelscited by 57
  23. 2020Evaluating and Mitigating Bias in Image Classifiers: A Causal Perspective Using Counterfactualscited by 56
  24. 2020A Sober Look at the Unsupervised Learning of Disentangled Representations and their Evaluationcited by 52
  25. 2018Counterfactuals uncover the modular structure of deep generative modelscited by 51
  26. 2018Fairness Through Causal Awareness: Learning Latent-Variable Models for Biased Datacited by 39
  27. 2020Why resampling outperforms reweighting for correcting sampling bias with stochastic gradientscited by 34
  28. 2021A Critical Look at the Consistency of Causal Estimation With Deep Latent Variable Modelscited by 29
  29. 2020A theory of independent mechanisms for extrapolation in generative modelscited by 17

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2023: 5; 2024: 12; 2025: 9; 2026: 8.

34 papers.

Sort:

By area:

  1. 2023High Fidelity Image Counterfactuals with Probabilistic Causal Modelscited by 53
  2. 2024On the Challenges and Opportunities in Generative AIcited by 31
  3. 2024An X-Ray Is Worth 15 Features: Sparse Autoencoders for Interpretable Radiology Report Generationcited by 23
  4. 2024Benchmarking Counterfactual Image Generationcited by 22
  5. 2023From Identifiable Causal Representations to Controllable Counterfactual Generation: A Survey on Causal Generative Modelingcited by 21
  6. 2024Mitigating attribute amplification in counterfactual image generationcited by 12
  7. 2024Robust image representations with counterfactual contrastive learningcited by 12
  8. 2024Counterfactual contrastive learning: robust representations via causal image synthesiscited by 11
  9. 2024Counterfactual Generative Modeling with Variational Causal Inferencecited by 10
  10. 2024Modular Learning of Deep Causal Generative Models for High-dimensional Causal Inferencecited by 9
  11. 2025Diffusion Counterfactual Generation with Semantic Abductioncited by 8
  12. 2024Semi-Supervised Learning for Deep Causal Generative Modelscited by 6
22 more
  1. 2025Causally Steered Diffusion for Automated Video Counterfactual Generationcited by 5
  2. 2025Counterfactual Identifiability via Dynamic Optimal Transportcited by 5
  3. 2023Towards Characterizing Domain Counterfactuals For Invertible Latent Causal Modelscited by 4
  4. 2024Conditional Generative Models are Sufficient to Sample from Any Causal Effect Estimandcited by 4
  5. 2025Causal-Adapter: Taming Text-to-Image Diffusion for Faithful Counterfactual Generationcited by 4
  6. 2024MACAW: A Causal Generative Model for Medical Imagingcited by 3
  7. 2025Flow Stochastic Segmentation Networkscited by 3
  8. 2023Punctate White Matter Lesion Segmentation in Preterm Infants Powered by Counterfactually Generative Learningcited by 2
  9. 2025Evaluation of 3D Counterfactual Brain MRI Generationcited by 2
  10. 2025Segmentor-Guided Counterfactual Fine-Tuning for Locally Coherent and Targeted Image Synthesiscited by 2
  11. 2025AI Alignment in Medical Imaging: Unveiling Hidden Biases Through Counterfactual Analysiscited by 1
  12. 2025Synthesising Counterfactual Explanations via Label-Conditional Gaussian Mixture Variational Autoencoderscited by 1
  13. 2026CEPAE: Conditional Entropy-Penalized Autoencoders for Time Series Counterfactualscited by 1
  14. 2026Pixel-level Counterfactual Contrastive Learning for Medical Image Segmentationcited by 1
  15. 2026When the Edit Changes the Patient: Measuring Identity Preservation in Counterfactual Retinal Imagescited by 1
  16. 2026Generating Chest X-Ray Counterfactuals by Specialising Foundation Image Modelscited by 1
  17. 2023VALUED -- Vision and Logical Understanding Evaluation Datasetcited by 0
  18. 2024Natural Counterfactuals With Necessary Backtrackingcited by 0
  19. 2026Positional Segmentor-Guided Counterfactual Fine-Tuning for Spatially Localized Image Synthesiscited by 0
  20. 2026Counterfactual Stress Testing for Image Classification Modelscited by 0
  21. 2026Causal Variational Deep Embedding: A Family of Interventional Generators for Confounded Imagescited by 0
  22. 2026To do(x) or not to do(x): Medical Image Counterfactuals for Dataset Augmentationcited by 0