MESD quantifies disparities in explanation quality across intersectional subgroups by combining label-aware aggregation, empirical-Bayes shrinkage, and CVaR weighting within a multi-objective optimization framework.
Fooling lime and shap: Adversarial attacks on post hoc explanation methods
2 Pith papers cite this work. Polarity classification is still indexing.
2
Pith papers citing it
years
2026 2representative citing papers
ExplainGuard is a zero-trust framework that verifies XAI explanations with three checks, catching several explanation-manipulation attacks on a COMPAS case study.
citing papers explorer
-
MESD: A Risk-Sensitive Metric for Explanation Fairness Across Intersectional Subgroups
MESD quantifies disparities in explanation quality across intersectional subgroups by combining label-aware aggregation, empirical-Bayes shrinkage, and CVaR weighting within a multi-objective optimization framework.
-
ExplainGuard: A Zero Trust Framework for Post-Hoc Explanation Integrity Guarantees in Blackbox XAI Models
ExplainGuard is a zero-trust framework that verifies XAI explanations with three checks, catching several explanation-manipulation attacks on a COMPAS case study.