Dynamic computation has emerged as a promising avenue to enhance the
inf...
Over the past decade, deep learning models have exhibited considerable
a...
Recently, CLIP-guided image synthesis has shown appealing performance on...
Knowledge distillation is an effective approach to learn compact models
...
Early exiting is an effective paradigm for improving the inference effic...
Training a generative adversarial network (GAN) with limited data has be...
Traditional knowledge distillation transfers "dark knowledge" of a
pre-t...
The backbone of traditional CNN classifier is generally considered as a
...