Pith. sign in

REVIEW 26 cited by

Counterfactual Explanations and Algorithmic Recourses for Machine Learning: A Review

Not yet reviewed by Pith; the record is open.

This paper has not been read by Pith yet. Machine review is queued; the pith claim, tier, and objections will appear here once it completes.

SPECIMEN: schema-true, not a live event

T0 review · schema-true

One-sentence machine reading of the paper's core claim.

pith:XXXXXXXX · record.json · timestamp

arxiv 2010.10596 v3 pith:XKXGZ3V6 submitted 2020-10-20 cs.LG cs.AIstat.ML

classification cs.LGcs.AIstat.ML
keywords learningmachinecounterfactualresearchexplainabilityrubricsystemsalgorithms
verification ladder T0 review T1 audit T2 compute T3 formal

Signed reviews

No signed human review yet.

0 comments
read the original abstract

Machine learning plays a role in many deployed decision systems, often in ways that are difficult or impossible to understand by human stakeholders. Explaining, in a human-understandable way, the relationship between the input and output of machine learning models is essential to the development of trustworthy machine learning based systems. A burgeoning body of research seeks to define the goals and methods of explainability in machine learning. In this paper, we seek to review and categorize research on counterfactual explanations, a specific class of explanation that provides a link between what could have happened had input to a model been changed in a particular way. Modern approaches to counterfactual explainability in machine learning draw connections to the established legal doctrine in many countries, making them appealing to fielded systems in high-impact areas such as finance and healthcare. Thus, we design a rubric with desirable properties of counterfactual explanation algorithms and comprehensively evaluate all currently proposed algorithms against that rubric. Our rubric provides easy comparison and comprehension of the advantages and disadvantages of different approaches and serves as an introduction to major research themes in this field. We also identify gaps and discuss promising research directions in the space of counterfactual explainability.

Discussion (0). Continue with ORCID to comment.

Forward citations

Cited by 26 Pith papers

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score. OpenAlex reports about 58 citations worldwide. Full citation record

  1. Back to the Feature: Explaining Video Classifiers with Video Counterfactual Explanations

    cs.CV 2025-11 conditional novelty 7.0 of 10

    BTTF optimizes the initial noise of an image-to-video diffusion model using the target classifier's gradients to produce minimal counterfactual videos that explain video classifiers.

  2. DCFO: Density-Based Counterfactuals for Outliers -- Additional Material

    cs.LG 2025-12 conditional novelty 6.0 of 10

    DCFO partitions the feature space by nearest-neighbour structure to make LOF scores differentiable, then uses gradient-based search to find the closest change that turns an outlier into an inlier.

  3. Enhancing Interpretability and Effectiveness in Recommendation with Numerical Features via Learning to Contrast the Counterfactual samples

    cs.IR 2025-09 conditional novelty 6.0 of 10

    A plug-in contrastive loss that enforces monotonicity between numerical features and recommender outputs, via counterfactual sample synthesis, improves AUC, GAUC, and monotonicity.

  4. VISION: Robust and Interpretable Code Vulnerability Detection Leveraging Counterfactual Augmentation

    cs.AI 2025-08 conditional novelty 6.0 of 10

    LLM-generated counterfactual code pairs with flipped vulnerability labels, used to train a GNN, sharply improve CWE-20 detection and attribution on the released CWE-20-CFA benchmark.

  5. Are machine learning interpretations reliable? A stability study on global interpretations

    stat.ML 2025-05 conditional novelty 6.0 of 10

    Popular machine learning interpretation methods are frequently unstable under small data perturbations, and interpretation stability does not track prediction accuracy.

  6. On Measuring Intrinsic Causal Attributions in Deep Neural Networks

    stat.ML 2025-05 conditional novelty 6.0 of 10

    Intrinsic causal contributions of input features to a neural network's output can be estimated from observational data via causal normalizing flows, and reduce to Sobol indices when inputs are independent.

  7. Why Are You Wrong? Counterfactual Explanations for Language Grounding with 3D Objects

    cs.AI 2025-05 conditional novelty 6.0 of 10

    A method uses word-level sampling and a genetic algorithm to generate counterfactual utterances that flip a language-grounding model's wrong prediction to the correct one.

  8. Towards Robust and Accurate Stability Estimation of Local Surrogate Models in Text-based Explainable AI

    cs.LG 2025-01 conditional novelty 6.0 of 10

    Many similarity measures used to judge adversarial text-XAI attacks are too sensitive, and synonymity-weighted variants change the measured success rates substantially.

  9. Derivative-Free Diffusion Manifold-Constrained Gradient for Unified XAI

    cs.CV 2024-11 conditional novelty 6.0 of 10

    FreeMCG estimates on-manifold classifier gradients from black-box outputs using diffusion particles and an ensemble Kalman filter, and uses the same estimate for feature attribution and counterfactual explanation.

  10. Private Counterfactual Retrieval With Immutable Features

    cs.IT 2024-11 conditional novelty 6.0 of 10

    Proposes two PIR-based protocols for retrieving the exact nearest counterfactual while keeping the user's immutable feature set private, with communication costs 6d+3M and 9(d+M).

  11. AULLM++: Structured-Token-Conditioned Large Language Models for Micro-Expression Action Unit Detection

    cs.CV 2026-03 conditional novelty 5.5 of 10

    AULLM++ fuses multi-granularity visual tokens with FACS-prior AU graph instructions into an LLM prompt and uses counterfactual consistency training to improve micro-expression AU detection and cross-domain Macro-F1.

  12. Negotiating Risk Boundaries in AI for Policing Through Mixed-Stakeholder Deliberation

    cs.AI 2026-08 conditional novelty 5.0 of 10

    A mixed-stakeholder UK workshop rated 13 AI policing use cases, rejecting recidivism risk assessment outright while accepting most others conditionally, and found that a racial-equity focus broadened, not narrowed, th...

  13. Inferring Effects of Major Events through Discontinuity Forecasting of Population Anxiety

    cs.LG 2025-08 conditional novelty 5.0 of 10

    Discontinuity forecasting predicts a county's anxiety jump and slope change after a major event from pre-event trends, reaching out-of-sample correlations of about .76 and .87.

  14. Explainable AI guided unsupervised fault diagnostics for high-voltage circuit breakers

    cs.LG 2025-07 conditional novelty 5.0 of 10

    An unsupervised framework detects, clusters, and explains circuit breaker faults from vibration and acoustic signals using only healthy data for training.

  15. XplainAct: Visualization for Personalized Intervention Insights

    cs.HC 2025-07 conditional novelty 5.0 of 10

    XplainAct combines choropleth maps, LIME/SHAP local explanations, and nearest-neighbor subgrouping to simulate and interpret personalized interventions at the county level.

  16. Integrating Probabilistic Trees and Causal Networks for Clinical and Epidemiological Data

    cs.LG 2025-01 conditional novelty 5.0 of 10

    PCF integrates causal Bayesian networks and probability-tree ensembles to deliver predictions plus intervention and counterfactual analysis on three healthcare datasets.

  17. Graph Counterfactual Explainable AI via Latent Space Traversal

    cs.LG 2025-01 conditional novelty 5.0 of 10

    Counterfactual graph explanations are generated by gradient descent in the latent space of a permutation-equivariant graph VAE, steering the graph's encoding to the opposite class.

  18. Position: Explainability Research Must Prioritize Foundations over Ad-hoc Methods

    cs.LG 2026-06 conditional novelty 4.0 of 10

    Explainable AI research should prioritize definitions, properties, evaluations, and actionability over new ad-hoc methods, on evidence from 617 papers and 34 practitioners.

  19. Interpretable Event Diagnosis in Water Distribution Networks

    cs.AI 2025-05 conditional novelty 4.0 of 10

    Counterfactual fingerprints computed from sensor residuals can distinguish leakages from sensor faults in water distribution networks and provide operators with a contrastive explanation.

  20. Integrating Explainable AI in Medical Devices: Technical, Clinical and Regulatory Insights and Recommendations

    cs.HC 2025-05 conditional novelty 4.0 of 10

    An MHRA-convened expert group and an eight-clinician pilot found that local explainable AI outputs improve trust and accuracy in heart attack risk predictions, while also exposing automation bias risks.

  21. Model-Free Counterfactual Subset Selection at Scale

    cs.LG 2025-02 reject novelty 4.0 of 10

    A one-pass streaming algorithm selects a diverse, relevant subset of real examples as counterfactual explanations, with a claimed 1/5.585 approximation guarantee and O(log k) update time.

  22. Pareto Optimal Algorithmic Recourse in Multi-cost Function

    cs.LG 2025-02 reject novelty 4.0 of 10

    A Bellman-Ford-style dynamic program over an actionability graph returns all Pareto-optimal recourse paths for multiple non-differentiable metric costs, with an epsilon-net sampling scheme proposed for scalability.

  23. Integrating Evidence into the Design of XAI and AI-based Decision Support Systems: A Means-End Framework for End-users in Construction

    cs.HC 2024-12 conditional novelty 4.0 of 10

    A narrative review proposes a means-end framework that connects evidence hierarchies and evidential pluralism to the design of explainable AI decision support systems for construction.

  24. Towards AI-$45^{\circ}$ Law: A Roadmap to Trustworthy AGI

    cs.CY 2024-12 conditional novelty 4.0 of 10

    The paper proposes the AI-45 degree law, a Causal Ladder framework, and five trustworthiness levels as a roadmap toward trustworthy AGI.

  25. Trustworthy AI in Digital Health: A Comprehensive Review of Robustness and Explainability

    cs.AI 2026-08 conditional novelty 3.0 of 10

    A survey of robustness and explainability methods for digital health AI, proposing a taxonomy and illustrating known XAI tools, without new empirical or theoretical results.

  26. CoFE: A Framework Generating Counterfactual ECG for Explainable Cardiac AI-Diagnostics

    cs.AI 2025-08 conditional novelty 3.0 of 10

    CoFE edits ECG signals in a StyleGAN2 latent space so that an AI model changes its prediction, and the resulting feature changes match clinical signs for AF and hyperkalemia.

Pith tools