XAI faithfulness evaluation can be flipped by changing perturbation type, partition size, or normalization, allowing a chosen explanation method to win, with a ranking-based mitigation called Mean Resilience Rank.
In: Advances in Neural Information Processing Sys- tems
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.AI 1years
2024 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
From Flexibility to Manipulation: The Slippery Slope of XAI Evaluation
XAI faithfulness evaluation can be flipped by changing perturbation type, partition size, or normalization, allowing a chosen explanation method to win, with a ranking-based mitigation called Mean Resilience Rank.