AI RESEARCH

EvoESAP: Non-Uniform Expert Pruning for Sparse MoE

arXiv CS.LG

ArXi:2603.06003v1 Announce Type: new Sparse Mixture-of-Experts (SMoE) language models achieve strong capability at low per-token compute, yet deployment remains memory- and throughput-bound because the full expert pool must be d and served. Post-