TESSER boosts black-box transferability of ViT-based adversarial attacks by reweighting gradients per token importance and smoothing perturbations spectrally, outperforming ATT on ImageNet benchmarks.
A light recipe to train robust vision transformers
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
-
TESSER: Transfer-Enhancing Adversarial Attacks from Vision Transformers via Spectral and Semantic Regularization
TESSER boosts black-box transferability of ViT-based adversarial attacks by reweighting gradients per token importance and smoothing perturbations spectrally, outperforming ATT on ImageNet benchmarks.