1 paper · 1 filter
Yifan Wu, Yiqi Wang, Xichen Ye +5
Knowledge Distillation (KD) is widely used to obtain compact models for efficient inference in resource-constrained environments. Yet the computational overhead of the distillation…