mlsanityData to 14 Oct 2026

S-LIME: Stabilized-LIME for Model Explanation

arXiv 2106.07875 ↗First posted 15 Jun 2021stat.ML cs.LGcited by 52

Not on the map. Why?

Abstract (arXiv)

An increasing number of machine learning models have been deployed in domains with high stakes such as finance and healthcare. Despite their superior performances, many models are black boxes in nature which are hard to explain. There are growing efforts for researchers to develop methods to interpret these black-box models. Post hoc explanations based on perturbations, such as LIME, are widely used approaches to interpret a machine learning model after it has been built. This class of methods has been shown to exhibit large instability, posing serious challenges to the effectiveness of the method itself and harming user trust. In this paper, we propose S-LIME, which utilizes a hypothesis testing framework based on central limit theorem for determining the number of perturbation points needed to guarantee stability of the resulting explanation. Experiments on both simulated and real world data sets are provided to demonstrate the effectiveness of our method.

Key facts. Cited by 52 papers in the map (data through 14 Oct 2026). Builds on 26 of its 60 references that are in the map. First posted on arXiv 15 Jun 2021.

Builds on

26 of 60 references.

References · 26

  1. 2014Adam: A Method for Stochastic Optimizationcited by 37,240
  2. 2015Distilling the Knowledge in a Neural Networkcited by 10,324
  3. 2017A Unified Approach to Interpreting Model Predictionscited by 6,555
  4. 2016"Why Should I Trust You?": Explaining the Predictions of Any Classifiercited by 5,070
  5. 2013Deep Inside Convolutional Networks: Visualising Image Classification Models and Saliency Mapscited by 2,371
  6. 2017Understanding Black-box Predictions via Influence Functionscited by 1,741
  7. 2017Towards A Rigorous Science of Interpretable Machine Learningcited by 1,416
  8. 2016Recurrent Neural Networks for Multivariate Time Series with Missing Valuescited by 554
  9. 2017Interpretation of Neural Networks is Fragilecited by 444
  10. 2019Fooling LIME and SHAP: Adversarial Attacks on Post hoc Explanation Methodscited by 443
  11. 2016European Union regulations on algorithmic decision-making and a "right to explanation"cited by 344
  12. 2020Towards Understanding Ensemble, Knowledge Distillation and Self-Distillation in Deep Learningcited by 297
14 more
  1. 2018Scalable and accurate deep learning for electronic health recordscited by 280
  2. 2018Techniques for Interpretable Machine Learningcited by 238
  3. 2019Fairwashing: the risk of rationalizationcited by 81
  4. 2019DLIME: A Deterministic Local Interpretable Model-Agnostic Explanations Approach for Computer-Aided Diagnosis Systemscited by 59
  5. 2015Interpretable Classification Models for Recidivism Predictioncited by 44
  6. 2015False Discoveries Occur Early on the Lasso Pathcited by 31
  7. 2019Unbiased Measurement of Feature Importance in Tree-Based Methodscited by 30
  8. 2020OptiLIME: Optimized LIME Explanations for Diagnostic Computer Algorithmscited by 30
  9. 2020Why distillation helps: a statistical perspectivecited by 21
  10. 2017An Improved Multi-Output Gaussian Process RNN with Real-Time Validation for Early Sepsis Detectioncited by 16
  11. 2019Early detection of sepsis utilizing deep learning on electronic health record event sequencescited by 15
  12. 2020Improving LIME Robustness with Smarter Locality Samplingcited by 12
  13. 2018Approximation Trees: Statistical Stability in Model Distillationcited by 9
  14. 2017When Is the First Spurious Variable Selected by Sequential Regression Procedures?cited by 3

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2021: 4; 2022: 7; 2023: 10; 2024: 10; 2025: 12; 2026: 9.

52 papers.

Sort:

By area:

  1. 2023Adversarial attacks and defenses in explainable artificial intelligence: A surveycited by 48
  2. 2023GLIME: General, Stable and Local LIME Explanationcited by 21
  3. 2023Calibrated Explanations: with Uncertainty Information and Counterfactualscited by 16
  4. 2021Partial Order in Chaos: Consensus on Feature Attributions in the Rashomon Setcited by 12
  5. 2022Motif-guided Time Series Counterfactual Explanationscited by 12
  6. 2024M-CELS: Counterfactual Explanation for Multivariate Time Series Data Guided by Learned Saliency Mapscited by 10
  7. 2021ControlBurn: Feature Selection by Sparse Forestscited by 9
  8. 2022Interpretation of Black Box NLP Models: A Surveycited by 9
  9. 2023Stabilizing Estimates of Shapley Values with Control Variatescited by 9
  10. 2023Leveraging Model-based Trees as Interpretable Surrogate Models for Model Distillationcited by 8
  11. 2023Explaining black boxes with a SMILE: Statistical Model-agnostic Interpretability with Local Explanationscited by 8
  12. 2024Beyond Pixels: Enhancing LIME with Hierarchical Features and Segmentation Foundation Modelscited by 8
38 more
  1. 2024Feature Responsiveness Scores: Model-Agnostic Explanations for Recoursecited by 8
  2. 2025Which LIME should I trust? Concepts, Challenges, and Solutionscited by 8
  3. 2022Rank-LIME: Local Model-Agnostic Feature Attribution for Learning to Rankcited by 7
  4. 2023Shaping Up SHAP: Enhancing Stability through Layer-Wise Neighbor Selectioncited by 7
  5. 2024Statistical Significance of Feature Importance Rankingscited by 7
  6. 2021Uncertainty Quantification of Surrogate Explanations: an Ordinal Consensus Approachcited by 6
  7. 2024Understanding Disparities in Post Hoc Machine Learning Explanationcited by 6
  8. 2024Why do explanations fail? A typology and discussion on failures in XAIcited by 6
  9. 2022Marginal Effects for Non-Linear Prediction Functionscited by 5
  10. 2024Interpretable Machine Learning for Survival Analysiscited by 4
  11. 2025Explaining Large Language Models with gSMILEcited by 4
  12. 2025Explainable Knowledge Graph Retrieval-Augmented Generation (KG-RAG) with KG-SMILEcited by 4
  13. 2022A Generic Approach for Reproducible Model Distillationcited by 3
  14. 2024Interpretability in Symbolic Regression: a benchmark of Explanatory Methods using the Feynman data setcited by 3
  15. 2024Towards consistency of rule-based explainer and black box model -- fusion of rule induction and XAI-based feature importancecited by 3
  16. 2024Mapping the Mind of an Instruction-based Image Editing using SMILEcited by 3
  17. 2025Evaluating the stability of model explanations in instance-dependent cost-sensitive credit scoringcited by 3
  18. 2021When Stability meets Sufficiency: Informative Explanations that do not Overwhelmcited by 2
  19. 2023Unsupervised discovery of Interpretable Visual Conceptscited by 2
  20. 2025Attribution Explanations for Deep Neural Networks: A Theoretical Perspectivecited by 2
  21. 2022On the Bias-Variance Characteristics of LIME and SHAP in High Sparsity Movie Recommendation Explanation Taskscited by 1
  22. 2022Explaining Machine Learning DGA Detectors from DNS Traffic Datacited by 1
  23. 2023CLIMAX: An exploration of Classifier-Based Contrastive Explanationscited by 1
  24. 2025Coherent Local Explanations for Mathematical Optimizationcited by 1
  25. 2026Regional Explanations: Bridging Local and Global Variable Importancecited by 1
  26. 2026ConceptSMILE: Auditing the Trustworthiness of Concept-Based Explainable AIcited by 1
  27. 2023EBLIME: Enhanced Bayesian Local Interpretable Model-agnostic Explanationscited by 0
  28. 2025EVolutionary Independent DEtermiNistiC Explanationcited by 0
  29. 2025Green LIME: Improving AI Explainability through Design of Experimentscited by 0
  30. 2025MindfulLIME: A Stable Solution for Explanations of Machine Learning Models with Enhanced Localization Precision -- A Medical Image Case Studycited by 0
  31. 2025Are We Merely Justifying Results ex Post Facto? Quantifying Explanatory Inversion in Post-Hoc Model Explanationscited by 0
  32. 2025ITL-LIME: Instance-Based Transfer Learning for Enhancing Local Explanations in Low-Resource Data Settingscited by 0
  33. 2025A Model of Causal Explanation on Neural Networks for Tabular Datacited by 0
  34. 2026LIME-LLM: Probing Models with Fluent Counterfactuals, Not Broken Textcited by 0
  35. 2026Enhancing LIME using Neural Decision Treescited by 0
  36. 2026Feature Attribution Stability Suite: How Stable Are Post-Hoc Attributions?cited by 0
  37. 2026BREAD: Baseline-Referenced Explanations for Anomaly Diagnosiscited by 0
  38. 2026When Superpixels Fail on Documents: A Study of Segmentation for LIME Explanationscited by 0

50 most cited of 52 All 52 (JSON)