Cross-Modal Alternating Learning with Task-Aware Representations for Continual Learning

Open paper
Benchmark Model Rank Results
continual-learning-on-cifar100-10-tasksALTA-ViTB/16#1Average Accuracy: 92.85
continual-learning-on-cifar100-10-tasksALTA-RN50x4#2Average Accuracy: 84.91
continual-learning-on-cifar100-10-tasksALTA-RN101#4Average Accuracy: 84.77
continual-learning-on-cifar100-10-tasksALTA-RN50#5Average Accuracy: 83.87
continual-learning-on-tiny-imagenet-10tasksALTA-ViTB/16#1Average Accuracy: 89.80
continual-learning-on-tiny-imagenet-10tasksALTA-RN50x4#2Average Accuracy: 84.73
continual-learning-on-tiny-imagenet-10tasksALTA-RN101#3Average Accuracy: 83.35
continual-learning-on-tiny-imagenet-10tasksALTA-RN50#4Average Accuracy: 81.07

Rank counts only results with a code link.