Preprint
Nov 2026
PagedMoE: Cross-Precision Expert Caching for Token-Level Dynamic Precision in SSD-Resident MoE Inference
PagedMoE, which stores each expert once using a prefix-compatible tree codebook and bitplane layout, and directly executes cached 1-8-bit prefixes without materializing full-precision experts.
Yu-Tong Lin, Yingdi Shan, Shao-Yuan Chen et al.
· 0 citations