A consistency-times-responsiveness robustness metric, based on rank-biased overlap of segment rankings, ranks GradCAM++ as most noise-robust and EigenCAM and AblationCAM as least.
Revisiting the evaluation of class activation mapping for explainability: A novel metric and experi- mental analysis,
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
citation-role summary
background 1
citation-polarity summary
fields
cs.CV 1years
2025 1verdicts
CONDITIONAL 1roles
background 1polarities
unclear 1representative citing papers
citing papers explorer
-
Assessing the Noise Robustness of Class Activation Maps: A Framework for Reliable Model Interpretability
A consistency-times-responsiveness robustness metric, based on rank-biased overlap of segment rankings, ranks GradCAM++ as most noise-robust and EigenCAM and AblationCAM as least.