Jingyuan Xu

Staleness-Centric Optimizations for Parallel Diffusion MoE Inference

2024 · arXiv (Cornell University)

Mixture-of-Experts-based (MoE-based) diffusion models demonstrate remarkable scalability in high-fidelity image generation, yet their reliance on expert parallelism introduces critical communication bottlenecks. State-of-the-art methods alleviate such overhead in parallel diffusion inference through computation-communication overlapping, termed displaced …