Adversarial Robustness on CIFAR-10

Metric: Attack: AutoAttack (higher is better)

LeaderboardDataset

Loading chart...

Results

Submit a result

Hide extra data

Sort:

#	Model↕	Attack: AutoAttack▼	Extra Data	Paper	Date↕	Code
1	Stochastic-LWTA/PGD/WideResNet-34-10	82.6	No	Stochastic Local Winner-Takes-All Networks Enabl...	2021-12-05	Code
2	Stochastic-LWTA/PGD/WideResNet-34-5	81.22	No	Stochastic Local Winner-Takes-All Networks Enabl...	2021-12-05	Code
3	Mixed classifier	68.06	Yes	Improving the Accuracy-Robustness Trade-Off of C...	2023-01-29	Code
4	TRADES-ANCRA/ResNet18	59.7	No	Enhancing Robust Representation in Adversarial T...	2023-10-05	Code
5	GLOT-DR	49.94	No	Global-Local Regularization Via Distributional R...	2022-03-01	Code

#1Stochastic-LWTA/PGD/WideResNet-34-10SOTA
82.6
Attack: AutoAttack· 2021-12-05
Stochastic Local Winner-Takes-All Networks Enable Profound Adversarial Robustness Code
#2Stochastic-LWTA/PGD/WideResNet-34-5
81.22
Attack: AutoAttack· 2021-12-05
Stochastic Local Winner-Takes-All Networks Enable Profound Adversarial Robustness Code
#3Mixed classifier
68.06
Attack: AutoAttack· Extra Data· 2023-01-29
Improving the Accuracy-Robustness Trade-Off of Classifiers via Adaptive Smoothing Code
#4TRADES-ANCRA/ResNet18
59.7
Attack: AutoAttack· 2023-10-05
Enhancing Robust Representation in Adversarial Training: Alignment and Exclusion Criteria Code
#5GLOT-DR
49.94
Attack: AutoAttack· 2022-03-01
Global-Local Regularization Via Distributional Robustness Code