How Robust are Discriminatively Trained Zero-Shot Learning Models?

Re-assign community