Is Large-Scale Pretraining the Secret to Good Domain Generalization?

Open in new window