2 papers
cs.LG2025
Specter: Fast MoE Inference via Quantization-Enabled Multi-Step Speculative Offloading
Wenfeng Wang, Jiacheng Liu, Xiaofeng Hou +7
Mixture-of-experts (MoE) models often exceed a single accelerator's memory capacity. Offloading experts to host memory creates a severe PCIe I/O bottleneck because data-dependent r…
cs.CL2025
MoE-Prism: Disentangling Monolithic Experts for Elastic MoE Services via Model-System Co-Designs
Xinfeng Xia, Xiaofeng Hou, Jiacheng Liu +5
Mixture-of-Experts (MoE) scales model capacity through sparse activation, and is becoming an important architecture for large language models (LLMs). However, existing MoE serving…