TransBoost: Improving the Best ImageNet Performance using Deep Transduction

Open paper
Benchmark Model Rank Results
image-classification-on-cifar-10TransBoost-ResNet50#76Percentage correct: 97.61
image-classification-on-dtdTransBoost-ResNet50#9Accuracy: 76.49
image-classification-on-flowers-102TransBoost-ResNet50#32Accuracy: 97.85%
image-classification-on-food-101-1TransBoost-ResNet50#4Accuracy (%): 84.30
image-classification-on-imagenetTransBoost-ViT-S#391Top 1 Accuracy: 83.67%Number of params: 22.05M
image-classification-on-imagenetTransBoost-ConvNext-T#518Top 1 Accuracy: 82.46%Number of params: 28.59M
image-classification-on-imagenetTransBoost-Swin-T#553Top 1 Accuracy: 82.16%Number of params: 71.71M
image-classification-on-imagenetTransBoost-ResNet50-StrikesBack#631Top 1 Accuracy: 81.15%Number of params: 25.56M
image-classification-on-imagenetTransBoost-ResNet152#659Top 1 Accuracy: 80.64%Number of params: 60.19M
image-classification-on-imagenetTransBoost-ResNet101#694Top 1 Accuracy: 79.86%Number of params: 44.55M
image-classification-on-imagenetTransBoost-ResNet50#746Top 1 Accuracy: 79.03%
image-classification-on-imagenetTransBoost-EfficientNetB0#775Top 1 Accuracy: 78.60%Number of params: 5.29M
image-classification-on-imagenetTransBoost-MobileNetV3-L#841Top 1 Accuracy: 76.81%Number of params: 5.48M
image-classification-on-imagenetTransBoost-ResNet34#848Top 1 Accuracy: 76.70%Number of params: 21.8M
image-classification-on-imagenetTransBoost-ResNet18#925Top 1 Accuracy: 73.36%Number of params: 11.69M
image-classification-on-stanford-carsTransBoost-ResNet50#11Accuracy: 90.80%

Rank counts only results with a code link.