Skip to content

Hexu Zhao

Office: 424, 60 5th Ave, New York, NY 10011

Work Email: hz3496 [at] nyu [dot] edu

CV image/svg+xml CV CV Google Scholar image/svg+xml Google Scholar Google Scholar dblp image/svg+xml dblp dblp

Bio

I am a Ph.D. student of Computer Science at NYU Courant, advised by Prof. Jinyang Li and Prof. Aurojit Panda. My research focuses on Machine Learning System, especially the emerging niche of System for Gaussian Splatting. From a systems perspective, I leverage both distributed computation and kernel optimization techniques. From a workload perspective, I optimize for both static and dynamic scene reconstruction. Previously, I obtained my bachelor’s degree in 2023 from Honored Yao Class, Tsinghua University. In high school, I taught myself to win Zibo City’s first National Informatics Olympiad gold medal (NOI2018).

"I’ve always tried to find and take on the small things in this world that I’m meant, more than anyone else, to do."

News

[01/2025] I will join NVIDIA Spatial Intelligence Lab as a Research Scientist Intern in 2025 Summer. See you in Santa Clara!

[03/2024] I will join Microsoft DeepSpeed Team as a Research Scientist Intern in 2024 Summer. See you in Seattle!

Education

New York University Courant InstituteImage title

Sept. 2023 -- Present

Ph.D. in Computer Science, advised by Prof. Jinyang Li and Prof. Aurojit Panda

Tsinghua UniversityImage title

Sept. 2019 -- June 2023

B.Eng. in Computer Science (Honored Yao Class)

Experience

NVIDIA Spatial Intelligence LabImage title

May. 2025 – Now

Research Scientist Intern

Microsoft DeepspeedImage title

May. 2024 – Aug. 2024

Research Scientist Intern

Publications

CLM: Removing the GPU Memory Barrier for 3D Gaussian Splatting

*Hexu Zhao, *Xiwen Min, Xiaoteng Liu, Moonjun Gong, Yiming Li, Ang Li, Saining Xie, Jinyang Li, Aurojit Panda

ASPLOS 2026

On Scaling Up 3D Gaussian Splatting Training

Hexu Zhao, *Haoyang Weng, *Daohan Lu, Ang Li, Jinyang Li, Aurojit Panda, Saining Xie

ICLR2025 Oral   arXiv image/svg+xml arXiv arXiv arXiv   Code   Project Page

On Optimizing the Communication of Model Parallelism

*Yonghao Zhuang, *Hexu Zhao, Lianmin Zheng, Zhuohan Li, Eric P. Xing, Qirong Ho, Joseph E. Gonzalez, Ion Stoica, Hao Zhang

MLSys2023   arXiv image/svg+xml arXiv arXiv arXiv  

Fully Hyperbolic Neural Networks

Weize Chen, Xu Han, Yankai Lin, Hexu Zhao, Zhiyuan Liu, Peng Li, Maosong Sun, Jie Zhou

ACL2022   arXiv image/svg+xml arXiv arXiv arXiv  

Development of a Doctor-in-the-loop Interpretation Framework for Insulin Titration in Diabetes: a Proof-of-concept Study

*Haowei He, *Zhen Ying, Biao Li, Yujuan Fan, Ping Wang, Jiaping Lu, Liming Wu, Hexu Zhao, Xiaoying Li, Yang Yuan, Ying Chen

In submission to The Lancet Diabetes & Endocrinology Journal   Google Scholar image/svg+xml Google Scholar Google Scholar