PCoMoE: Shifting MoE Inference from Monolithic Expert Selection to Fine-Grained Path Composition
PCoMoE is presented, a path-compositional execution framework that shifts MoE inference from coarse-grained expert selection to fine-grained path composition and achieves up to a 1.31x end-to-end inference speedup while enhancing model accuracy by 10%.
Zi-Yan Gan, Fangxin Liu, Chenyang Guan et al.
· 0 citations