Zhongming Yu

San Diego, California

I am a third-year Ph.D. advised by Prof. Jishen Zhao in Computer Science and Engineering Department of UC San Diego, starting in 2022. I am now aiming to build efficient systems for machine learning. Before coming to UCSD, I received my B.E. degree from the Department of Electronic Engineering at Tsinghua University. I have a broad interest in combining machine learning and computer systems, with a special focus on the interaction between ML and System. I have previously focused on more acceleration for sparse computing (e.g. GNN, Pointcloud acceleration) and devoted myself to some open-source projects like dgSPARSE and CogDL. Now I am working on agent framework that could help design better system, especially moving to MLsys self-evolution. Some topics that I have been working on are listed below:

LLM Agent System for MLSys
Machine Learning Compiler
High-performance GPU Kernels
Autonomous Software Engineering (SWEBench)

Feel free to contact me if you share any common interests.

news

Feb 6, 2025	We just released our software issue localization framework OrcaLoca, check it out at Github.
Dec 10, 2024	We published our LLM4RTL project on arxiv! Check it out at MAGE.

selected publications

mlsys

Understanding gnn computational graph: A coordinated computation, io, and memory perspective

Hengrui Zhang*, Zhongming Yu*, Guohao Dai, and 4 more authors

Proceedings of Machine Learning and Systems, 2022

* stands for equal contribution

Bib HTML Code

@article{zhang2022understanding,
  title = {Understanding gnn computational graph: A coordinated computation, io, and memory perspective},
  author = {Zhang*, Hengrui and Yu*, Zhongming and Dai, Guohao and Huang, Guyue and Ding, Yufei and Xie, Yuan and Wang, Yu},
  journal = {Proceedings of Machine Learning and Systems},
  note = {* stands for equal contribution},
  volume = {4},
  pages = {467--484},
  year = {2022},
}

mlsys

HyperGef: A Framework Enabling Efficient Fusion for Hypergraph Neural Network on GPUs

Zhongming Yu, Guohao Dai, Shang Yang, and 6 more authors

Proceedings of Machine Learning and Systems, 2023

Bib HTML Code

@article{yu2023hypergef,
  title = {HyperGef: A Framework Enabling Efficient Fusion for Hypergraph Neural Network on GPUs},
  author = {Yu, Zhongming and Dai, Guohao and Yang, Shang and Zhang, Genghan and Zhang, Hengrui and Zhu, Feiwen and Yang, June and Zhao, Jishen and Wang, Yu},
  journal = {Proceedings of Machine Learning and Systems},
  volume = {5},
  year = {2023},
}

mlsys

Exploiting Hardware Utilization and Adaptive Dataflow for Efficient Sparse Convolution in 3D Point Clouds

Ke Hong*, Zhongming Yu*, Guohao Dai, and 4 more authors

Proceedings of Machine Learning and Systems, 2023

* stands for equal contribution

Bib HTML Code

@article{hong2023exploiting,
  title = {Exploiting Hardware Utilization and Adaptive Dataflow for Efficient Sparse Convolution in 3D Point Clouds},
  author = {Hong*, Ke and Yu*, Zhongming and Dai, Guohao and Yang, Xinhao and Lian, Yaoxiu and Xu, Ningyi and Wang, Yu},
  journal = {Proceedings of Machine Learning and Systems},
  volume = {5},
  note = {* stands for equal contribution},
  year = {2023},
}