Beichen Huang

About Me

Hello and welcome! I’m Beichen Huang, a first-year MSCS student at UIUC, supervised by Prof. Minjia Zhang at SSAIL. I am working on LLM inference, and my current research interests are:

Efficient test time scaling
Model compression and quantization

I am currently looking for an intern position for summer 2026.

News

[Jan. 2026] A preprint about the efficient parallel thinking is ready at PDF.
[Mar. 2025] Serve as AE reviewer for MLSys 2025.
[Feb. 2025] Our work MiLo is accepted to MLSys 2025. See you in Santa Clara!
[Nov. 2024] A preprint about an efficient 3-bit quantization system for MoE is ready.
[Sept. 2024] Our paper about Fractional Programming for Clustering is accepted to NeurIPS 2024.
[April. 2024] Our paper about Aerial_IRS is accepted to ICASSP 2024.

Publications

MiLo: Efficient Quantized MoE Inference with Mixture of Low-Rank Compensators

Beichen Huang*, Yueming Yuan*, Zelei Shao*, Minjia Zhang (*Equal Contributors)

The Eighth Annual Conference on Machine Learning and Systems (MLSys), 2025.

PDF

Multidimensional Fractional Programming for Normalized Cuts

Yannan Chen*, Beichen Huang*, Licheng Zhao, Kaiming Shen (*Equal Contributors)

the Thirty-Eighth Annual Conference on Neural Information Processing Systems (NeurIPS), 2024.

PDF

Aerial-IRS-Assisted Load Balancing In Downlink Networks

Shuyi Ren, Beichen Huang, Xiaoyang Li, Kaiming Shen

IEEE International Conference on Acoustics, Speech, and Signal Processing (ICASSP), 2024.

PDF

Services

Reviewer: ACL, MLSys, TPAMI