Fate: Fast Edge Inference of Mixture-of-Experts Models via Cross-Layer Gate
Published in The ACM Web Conference (WWW), 2026
Recommended citation: Fang, Zhiyuan, Xingfan Yu, Yuegui Huang, Zicong Hong, Yufeng Lyu, Wuhui Chen, Yue Yu, and Fan Yu. "Fate: Fast Edge Inference of Mixture-of-Experts Models via Cross-Layer Gate." In Proceedings of the ACM Web Conference 2026, pp. 5537–5546. https://arxiv.org/abs/2502.12224
Fate speeds up Mixture-of-Experts inference on edge and resource-constrained devices with a cross-layer gate, reducing expert-loading overhead.
I contributed as an industrial mentor through the CAAI–MindSpore academic research program, including guidance on unified-memory deployment and expert-prefetcher launch cost on constrained CPUs.