α-TCAV replaces TCAV's hard indicator with a tunable smooth function to create a unified probabilistic framework with lower variance and guidance for parameter choice or Bayes-optimal scoring.
arXiv preprint arXiv:2103.08307 , year=
2 Pith papers cite this work. Polarity classification is still indexing.
2
Pith papers citing it
representative citing papers
SAAD adaptively weights adversarial training samples by their transferability to the teacher, yielding higher AutoAttack robustness than prior distillation methods on CIFAR and Tiny-ImageNet without extra compute.
citing papers explorer
-
$\alpha$-TCAV: A Unified Framework for Testing with Concept Activation Vectors
α-TCAV replaces TCAV's hard indicator with a tunable smooth function to create a unified probabilistic framework with lower variance and guidance for parameter choice or Bayes-optimal scoring.
-
Sample-wise Adaptive Weighting for Transfer Consistency in Adversarial Distillation
SAAD adaptively weights adversarial training samples by their transferability to the teacher, yielding higher AutoAttack robustness than prior distillation methods on CIFAR and Tiny-ImageNet without extra compute.