Robustness Challenges in Model Distillation and Pruning for Natural Language Understanding

Open in new window