WRN-28-10

Reported on 4 benchmarks across 2 tasks · 2 papers · 1 SOTA

Note: results are matched by exact model name. Different papers may use the same name for different model variants.

Adversarial2 results

Adversarial DefenseonCIFAR-10
Accuracy· uses extra data· 2023-09-19
90.03
SOTA
Language Guided Adversarial Purification arXiv:2309.10348
Adversarial DefenseonCIFAR-10
Robust Accuracy· uses extra data· 2023-09-19
71.68
best: 82.96 (ResNet18 (TRADES-ANCRA/PGD-40))
Language Guided Adversarial Purification arXiv:2309.10348

Computer Vision2 results

Image ClassificationonCIFAR-10
Percentage correct· 2021-03-10
97.73
best: 99.5 (ViT-H/14)
MixMo: Mixing Multiple Inputs for Multiple Outputs via Deep Subnetworks arXiv:2103.06132
Image ClassificationonCIFAR-100
Percentage correct· uses extra data· 2021-03-10
85.77
best: 96.08 (EffNet-L2 (SAM))
MixMo: Mixing Multiple Inputs for Multiple Outputs via Deep Subnetworks arXiv:2103.06132