Yuchen Yang

@yuchen-yang · 1 works

Recent posts by Yuchen Yang

• PagedWeight: Efficient MoE LLM Serving with Dynamic Quality-Aware Weight Quantization - Mixture-of-Experts (MoE) is a popular class of large language models (LLMs), offering high efficiency and accuracy. However, in KV-cache-intensive serving scenarios, MoEs often exhibit a tension… · Paper · arxiv.org

Topics Yuchen Yang posts about

• MoE on Instagram - 1 post · Best MoE creators