mlsanityData to 7 Oct 2026

The Dangers of Post-hoc Interpretability: Unjustified Counterfactual Explanations

arXiv 1907.09294 ↗First posted 22 Jul 2019cs.LG cs.AI stat.MLcited by 97

Not on the map. Why?

Abstract (arXiv)

Post-hoc interpretability approaches have been proven to be powerful tools to generate explanations for the predictions made by a trained black-box model. However, they create the risk of having explanations that are a result of some artifacts learned by the model instead of actual knowledge from the data. This paper focuses on the case of counterfactual explanations and asks whether the generated instances can be justified, i.e. continuously connected to some ground-truth data. We evaluate the risk of generating unjustified counterfactual examples by investigating the local neighborhoods of instances whose predictions are to be explained and show that this risk is quite high for several datasets. Furthermore, we show that most state of the art approaches do not differentiate justified from unjustified counterfactual examples, leading to less useful explanations.

Key facts. Cited by 97 papers in the map (data through 7 Oct 2026). First posted on arXiv 22 Jul 2019.

Builds on

No reference list yet.

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2021: 22; 2022: 23; 2023: 14; 2024: 17; 2025: 16; 2026: 5.

97 papers.

Sort:

By area:

  1. 2021Interpretable Machine Learning: Fundamental Principles and 10 Grand Challengescited by 364
  2. 2022From Anecdotal Evidence to Quantitative Evaluation Methods: A Systematic Review on Evaluating Explainable AIcited by 299
  3. 2022Text and Patterns: For Effective Chain of Thought, It Takes Two to Tangocited by 128
  4. 2021CARLA: A Python Library to Benchmark Algorithmic Recourse and Counterfactual Explanation Algorithmscited by 85
  5. 2021If Only We Had Better Counterfactual Explanations: Five Key Deficits to Rectify in the Evaluation of Counterfactual XAI Techniquescited by 82
  6. 2021Counterfactuals and Causability in Explainable Artificial Intelligence: Theory, Algorithms, and Applicationscited by 59
  7. 2022Visual Recognition with Deep Nearest Centroidscited by 59
  8. 2023Learning Bottleneck Concepts in Image Classificationcited by 49
  9. 2021Generating Interpretable Counterfactual Explanations By Implicit Minimisation of Epistemic and Aleatoric Uncertaintiescited by 41
  10. 2023This Looks Like Those: Illuminating Prototypical Concepts Using Multiple Visualizationscited by 36
  11. 2021Deep learning for temporal data representation in electronic health records: A systematic review of challenges and methodologiescited by 34
  12. 2022On the Robustness of Sparse Counterfactual Explanations to Adverse Perturbationscited by 32
38 more
  1. 2024Facing the Elephant in the Room: Visual Prompt Tuning or Full Finetuning?cited by 31
  2. 2022Interpretable Geometric Deep Learning via Learnable Randomness Injectioncited by 27
  3. 2022Scalable Interpretability via Polynomialscited by 25
  4. 2024Interpretable Image Classification with Adaptive Prototype-based Vision Transformerscited by 25
  5. 2021Knowledge-Grounded Self-Rationalization via Extractive and Natural Language Explanationscited by 24
  6. 2021Counterfactual Shapley Additive Explanationscited by 22
  7. 2022Equivariant and Invariant Grounding for Video Question Answeringcited by 22
  8. 2021Characterizing the risk of fairwashingcited by 21
  9. 2022Attribution-based Explanations that Provide Recourse Cannot be Robustcited by 20
  10. 2022The privacy issue of counterfactual explanations: explanation linkage attackscited by 20
  11. 2022Improvement-Focused Causal Recourse (ICR)cited by 20
  12. 2022VCNet: A self-explaining model for realistic counterfactual generationcited by 20
  13. 2023Even if Explanations: Prior Work, Desiderata & Benchmarks for Semi-Factual XAIcited by 20
  14. 2021A Few Good Counterfactuals: Generating Interpretable, Plausible and Diverse Counterfactual Explanationscited by 19
  15. 2024MICA: Towards Explainable Skin Lesion Diagnosis via Multi-Level Image-Concept Alignmentcited by 18
  16. 2025A Comprehensive Survey on Self-Interpretable Neural Networkscited by 18
  17. 2021CARE: Coherent Actionable Recourse based on Sound Counterfactual Explanationscited by 17
  18. 2021Uncertainty Estimation and Out-of-Distribution Detection for Counterfactual Explanations: Pitfalls and Solutionscited by 16
  19. 2024M2PT: Multimodal Prompt Tuning for Zero-shot Instruction Learningcited by 16
  20. 2024Neural Concept Bindercited by 14
  21. 2022Counterfactual Explanations for Misclassified Images: How Human and Machine Explanations Differcited by 12
  22. 2024Neural Clustering based Visual Representation Learningcited by 12
  23. 2023Achieving Diversity in Counterfactual Explanations: a Review and Discussioncited by 11
  24. 2025Re-Imagining Multimodal Instruction Tuning: A Representation Viewcited by 9
  25. 2025textttDroid: A Resource Suite for AI-Generated Code Detectioncited by 9
  26. 2021Solving the Class Imbalance Problem Using a Counterfactual Method for Data Augmentationcited by 8
  27. 2023TIGTEC : Token Importance Guided TExt Counterfactualscited by 8
  28. 2023The Utility of "Even if..." Semifactual Explanation to Optimise Positive Outcomescited by 8
  29. 2021On the Importance of Domain-specific Explanations in AI-based Cybersecurity Systems (Technical Report)cited by 7
  30. 2022Benchmarking Instance-Centric Counterfactual Algorithms for XAI: From White Box to Black Boxcited by 7
  31. 2021Individual Explanations in Machine Learning Models: A Survey for Practitionerscited by 6
  32. 2023The Contextual Lasso: Sparse Linear Models via Deep Neural Networkscited by 6
  33. 2024Understanding Disparities in Post Hoc Machine Learning Explanationcited by 6
  34. 2025GlyTwin: Digital Twin for Glucose Control in Type 1 Diabetes Through Optimal Behavioral Modifications Using Patient-Centric Counterfactualscited by 5
  35. 2021Modeling Users and Online Communities for Abuse Detection: A Position on Ethics and Explainabilitycited by 4
  36. 2022Sensing accident-prone features in urban scenes for proactive driving and accident preventioncited by 4
  37. 2022Learning to Select Prototypical Parts for Interpretable Sequential Data Modelingcited by 4
  38. 2021Twin Systems for DeepCBR: A Menagerie of Deep Learning and Case-Based Reasoning Pairings for Explanation and Data Augmentationcited by 3

50 most cited of 97 All 97 (JSON)