CasMoE: A Cascaded Framework for Efficient MoE Inference on Resource-constrained Devices.
Chengcheng Wang, Haowen He, Liang Zhao, Xiaoheng Deng, Lixin Duan, Shaohua Wan
Browse the full AAAI paper archive.
Chengcheng Wang, Haowen He, Liang Zhao, Xiaoheng Deng, Lixin Duan, Shaohua Wan
Browse the full AAAI paper archive.