Adversarial Robustness on CIFAR-10

7개 결과 · ⬇ CSV · JSON

Accuracy

Rank	Model	Accuracy	Robust Accuracy	Attack: AutoAttack	Extra Training Data	Paper	Code	Year
1	A Geometric Probe of the Accuracy-Robust 자동 추출	95.63	–	–		A Geometric Probe of the Accuracy-Robustness Trade-off: Sharp Boundaries in Symmetry-Breaking Dimensional Expansion		2026
2	Mixed classifier	95.23	68.06	68.06	✓	Improving the Accuracy-Robustness Trade-Off of Classifiers via Adaptive Smoothing	codelion/adaptive-classifier · bai-yt/adaptivesmoothing	2023
3	Neural Architecture Search with Mixed Bi 자동 추출	95.16	–	–		Neural Architecture Search with Mixed Bio-inspired Learning Rules		2025
4	Stochastic-LWTA/PGD/WideResNet-34-10	92.26	84.3	82.6		Stochastic Local Winner-Takes-All Networks Enable Profound Adversarial Robustness	konpanousis/adversarial-lwta-autoattack	2021
5	Stochastic-LWTA/PGD/WideResNet-34-5	91.88	83.4	81.22		Stochastic Local Winner-Takes-All Networks Enable Profound Adversarial Robustness	konpanousis/adversarial-lwta-autoattack	2021
6	GLOT-DR	84.13	–	49.94		Global-Local Regularization Via Distributional Robustness	viethoang1512/glot	2022
7	TRADES-ANCRA/ResNet18	81.70	–	59.70		Enhancing Robust Representation in Adversarial Training: Alignment and Exclusion Criteria	changzhang777/ancra	2023