Preprint
Aug 2026
Launch-Bound and Substitutable: Why Three Inference Optimizations Fail to Pay Off in Mixture-of-Experts Models
A fourth result ties the three together: leaving the routers in FP16 lowers drift by 20% while raising loss, so routing fidelity and output quality are separable objectives.
Gokulakannan Sakthivel, Jerry Wu, A. Rajendra et al.
· 0 citations