1 paper · 1 filter
Zhaoyi Yan, Binghui Chen, Yunfan Liu +1
Knowledge distillation (KD) aims to transfer knowledge from a large-scale teacher model to a lightweight one, significantly reducing computational and storage requirements. However…