Xiaowen Zhang

I am a junior undergraduate student majoring in Electrical and Computer Engineering at Shanghai Jiao Tong University (SJTU), where I work firmly with Prof. David L.S. Hung and Dr. Fengnian Zhao at FDA Lab. I am currently visiting Carnegie Mellon University (CMU) as an exchange student.

My research interests center on Machine Learning Systems, with a particular focus on LLM Inference Efficiency. I am passionate about bridging the gap between algorithms and underlying hardware to build efficient, scalable systems.

Recently, I developed FlowForge, a compile–execute system designed to tackle the computational challenges in physical flow field prediction. By addressing system-level bottlenecks, FlowForge enables efficient and stable training and inference for complex fluid dynamics, significantly outperforming standard deep learning frameworks. (Paper under review)

Education

Carnegie Mellon University (Exchange), 2026.1 - Present
Shanghai Jiao Tong University, 2023.9 - Present

selected publications

Under Review

FlowForge: A Staged Local Rollout Engine for Flow-Field Prediction

Xiaowen Zhang^*, Ziming Zhou^*, Fengnian Zhao, and 1 more author

Under Review, Jan 2026

Bib

@article{FlowForge2026,
  title = {FlowForge: A Staged Local Rollout Engine for Flow-Field Prediction},
  author = {Zhang, Xiaowen and Zhou, Ziming and Zhao, Fengnian and Hung, David L.S.},
  journal = {Under Review},
  year = {2026},
  month = jan,
  url = {https://openreview.net/forum?id=76AuCJQDeU&referrer=%5BAuthor%20Console%5D(%2Fgroup%3Fid%3DICML.cc%2F2026%2FConference%2FAuthors%23your-submissions)},
}