mlsanityData to 7 Oct 2026

Don't Lie to Me: Avoiding Malicious Explanations with STEALTH

arXiv 2301.10407 ↗First posted 25 Jan 2023cs.SE cs.AI cs.CRcited by 2

Not on the map. Why?

Abstract (arXiv)

STEALTH is a method for using some AI-generated model, without suffering from malicious attacks (i.e. lying) or associated unfairness issues. After recursively bi-clustering the data, STEALTH system asks the AI model a limited number of queries about class labels. STEALTH asks so few queries (1 per data cluster) that malicious algorithms (a) cannot detect its operation, nor (b) know when to lie.

Key facts. Cited by 2 papers in the map (data through 7 Oct 2026). Builds on 15 of its 34 references that are in the map. First posted on arXiv 25 Jan 2023.

Builds on

15 of 34 references.

References · 15

  1. 2016"Why Should I Trust You?": Explaining the Predictions of Any Classifiercited by 5,040
  2. 2019Fooling LIME and SHAP: Adversarial Attacks on Post hoc Explanation Methodscited by 440
  3. 2020Explaining by Removing: A Unified Framework for Model Explanationcited by 218
  4. 2021Counterfactual Explanations Can Be Manipulatedcited by 113
  5. 2021Bias in Machine Learning Software: Why? How? What to do?cited by 92
  6. 2020General Pitfalls of Model-Agnostic Interpretation Methods for Machine Learning Modelscited by 74
  7. 2020Can I Trust My Fairness Metric? Assessing Fairness with Unlabeled Data and Bayesian Inferencecited by 31
  8. 2021FairMask: Better Fairness via Model-based Rebalancing of Protected Attributescited by 19
  9. 2020Deceptive AI Explanations: Creation and Detectioncited by 14
  10. 2020Improving LIME Robustness with Smarter Locality Samplingcited by 12
  11. 2021Select Wisely and Explain: Active Learning and Probabilistic Local Post-hoc Explainabilitycited by 12
  12. 2022Unfooling Perturbation-Based Post Hoc Explainerscited by 12
3 more
  1. 2016"Sampling"' as a Baseline Optimizer for Search-based Software Engineeringcited by 8
  2. 2021Better sampling in explanation methods can prevent dieselgate-like deceptioncited by 5
  3. 2019An Exploratory Study on Machine Learning Model Storescited by 1

Built on by

Cited by year

Citing papers per year: 2025: 2.

2 papers.

  1. 2025Fairness Testing through Extreme Value Theorycited by 2
  2. 2025Software Fairness Dilemma: Is Bias Mitigation a Zero-Sum Game?cited by 1