Compositional robustness testing for vision models
-
Updated
May 28, 2026 - Python
Compositional robustness testing for vision models
Black-Box Neural Network Hardening Through Surrogate Model Defense Integration
Evasion attacks (FGSM, PGD, C&W, APGD, Square, AutoAttack-style ensemble), adversarial training (PGD-AT, TRADES, FGSM-RS) and randomized smoothing certification on the full CIFAR-10 test set, checked against torchattacks 3.5.1 and mapped to MITRE ATLAS v2026.10
Controlled PyTorch comparison of standard and adversarial training on GTSRB, with PGD, AutoAttack, reproducible evaluation, and measured accuracy–robustness trade-offs.
To associate your repository with the autoattack topic, visit your repo's landing page and select "manage topics."