Fig. 2
From: A practical generalization metric for deep networks benchmarking
TradeOff points of two kinds models, CLIP and EfficientNet (denoted as “\(\star\)”). The solid vertical lines indicate the selection of trade-off points on each marginals. (a–c) CLIP on ImageNet, (d–f) EfficientNet on ImageNet, (g–i) CLIP on CIFAR-100, (j–l) EfficientNet on CIFAR-100. Neither CLIP nor EfficientNet consistently dominates across all conditions. While EfficientNet achieves better robustness and lower error rates on ImageNet, CLIP performs slightly better in zero-shot scenarios on CIFAR-100.
