Welcome!

🌙 About me

I’m Yueming Yuan (袁悦茗), a second year CS PhD student at UIUC.

I have a broad interest in accurate and efficient techniques for LLM/VLM training, inference and reasoning.

✨ Research

Previously, I worked on algorithm-system co-design for efficient LLM training/inference, for example:

• (1) Large-scale MoE pretraining, parallelism (~1k GPUs scale) [SC 2025, Best Student Paper Nomination]
• (2) MoE quantization, inference efficiency [MLSys 2025]
• (3) Efficient computation & code-generation for sparse attention. [OOPSLA 2025]