mlsanityData to 7 Oct 2026

Issues with post-hoc counterfactual explanations: a discussion

arXiv 1906.04774 ↗First posted 11 Jun 2019cs.LG cs.AI stat.MLcited by 30

Not on the map. Why?

Abstract (arXiv)

Counterfactual post-hoc interpretability approaches have been proven to be useful tools to generate explanations for the predictions of a trained blackbox classifier. However, the assumptions they make about the data and the classifier make them unreliable in many contexts. In this paper, we discuss three desirable properties and approaches to quantify them: proximity, connectedness and stability. In addition, we illustrate that there is a risk for post-hoc counterfactual approaches to not satisfy these properties.

Key facts. Cited by 30 papers in the map (data through 7 Oct 2026). First posted on arXiv 11 Jun 2019.

Builds on

No reference list yet.

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2021: 5; 2022: 2; 2023: 10; 2024: 7; 2025: 3; 2026: 3.

30 papers.

Sort:

By area:

  1. 2023Robust Counterfactual Explanations for Neural Networks With Probabilistic Guaranteescited by 45
  2. 2021Evaluating Robustness of Counterfactual Explanationscited by 41
  3. 2021Consistent Counterfactuals for Deep Modelscited by 37
  4. 2024Self-eXplainable AI for Medical Image Analysis: A Survey and New Outlookscited by 18
  5. 2021CARE: Coherent Actionable Recourse based on Sound Counterfactual Explanationscited by 17
  6. 2023Generating robust counterfactual explanationscited by 13
  7. 2022Keep Your Friends Close and Your Counterfactuals Closer: Improved Learning From Closest Rather Than Plausible Counterfactual Explanations in an Abstract Settingcited by 11
  8. 2023Achieving Diversity in Counterfactual Explanations: a Review and Discussioncited by 11
  9. 2021Interpretable Machine Learning: Moving From Mythos to Diagnosticscited by 9
  10. 2023Stabilizing Estimates of Shapley Values with Control Variatescited by 9
  11. 2023The Risks of Recourse in Binary Classificationcited by 8
  12. 2024Counterfactual Concept Bottleneck Modelscited by 7
18 more
  1. 2021Understanding Prediction Discrepancies in Machine Learning Classifierscited by 6
  2. 2023Flexible and Robust Counterfactual Explanations with Minimal Satisfiable Perturbationscited by 6
  3. 2023Towards Conceptualization of "Fair Explanation": Disparate Impacts of anti-Asian Hate Speech Explanations on Content Moderatorscited by 6
  4. 2024Why do explanations fail? A typology and discussion on failures in XAIcited by 6
  5. 2025Neural Probabilistic Circuits: Enabling Compositional and Interpretable Predictions through Logical Reasoningcited by 6
  6. 2025Counterfactual Explanations for Model Ensembles Using Entropic Risk Measurescited by 4
  7. 2023Looking deeper into interpretable deep learning in neuroimaging: a comprehensive surveycited by 2
  8. 2024Generally-Occurring Model Change for Robust Counterfactual Explanationscited by 2
  9. 2023Towards User Guided Actionable Recoursecited by 1
  10. 2023T-COL: Generating Counterfactual Explanations for General User Preferences on Variable Machine Learning Systemscited by 1
  11. 2024Weak Robust Compatibility Between Learning Algorithms and Counterfactual Explanation Generation Algorithmscited by 1
  12. 2024In Defence of Post-hoc Explainabilitycited by 1
  13. 2025[Re] Improving Interpretation Faithfulness for Vision Transformerscited by 1
  14. 2022Beyond Model Interpretability: On the Faithfulness and Adversarial Robustness of Contrastive Textual Explanationscited by 0
  15. 2024Enhancing Counterfactual Image Generation Using Mahalanobis Distance with Distribution Preferences in Feature Spacecited by 0
  16. 2026Structure-Aware Robust Counterfactual Explanations via Conditional Gaussian Network Classifierscited by 0
  17. 2026We Need Explanation Cards to Connect Explanation Algorithms to the Real Worldcited by 0
  18. 2026Reliability, Faithfulness, and the Limits of Post-hoc Explanations of Opaque Scientific Modelscited by 0