STEP: Adaptive Spatio-Temporal Expert Prefetching for Low-Latency and Memory-Efficient MoE Inference.
Fangxin Liu, Ning Yang, Zongwu Wang, Chenyang Guan, Haomin Li, Yu Feng, Liqiang Lu, Xiang Li, Siran Yang, Jiamang Wang, Lin Qu, Li Jiang, Haibing Guan
Browse the full ISCA paper archive.