Scaling Pre-trained Language Models to Deeper via Parameter-efficient Architecture

MoE
Re-assign community