Zeshen Zhang

About Me

I am a Master's student in Computer Science (M.S.) at the NYU Tandon School of Engineering, expected to graduate in January 2027. I received my B.S. in Computer Science and Technology from Tongji University in 2025. My research interests lie in Machine Learning Systems.

I have been fortunate to work with Prof. Han Zhao (Shanghai Jiao Tong University) and Prof. Yucheng Lu (NYU Shanghai).

I am actively preparing to apply for PhD programs in Fall 2027 and would be happy to connect regarding research collaboration or opportunities.

Publications & Preprints

TACO (Minor Rev.)

From Overloaded to Guaranteed: High-Throughput Multi-SLO Enforcement for LoRA-Assisted On-Premise LLM Deployment
Zeshen Zhang, Han Zhao, Weihao Cui, Quan Chen, Yu Liu, Yongjun Deng, Jing Yang, Jiuchen Shi, Chen Chen, Youmin Chen, Yu Feng, Minyi Guo.
ACM Transactions on Architecture and Code Optimization (TACO)
arXiv

arXiv 2026

FlyWheel: Attention Kernels Built for the TPU
Zishuo Bao, Zeshen Zhang, Linkun Chen, Yucheng Lu.
arXiv preprint, Nov 2026
TPU kernels for softmax attention and Gated DeltaNet on TPU v6e and v7x, reaching 81% of BF16 peak on v6e, up to 1.80× faster than Splash Attention and 3.61× faster GDN prefill.
arXiv | Blog | Code

NeurIPS 2026

Multi-Token Residual Prediction
Yufeng Xu, Zishuo Bao, Qian Wang, Zeshen Zhang, Haoqi Zhang, Bowen Peng, Ang Li, Rahul Chalamala, Yucheng Lu.
Conference on Neural Information Processing Systems (NeurIPS 2026)
arXiv | Code | Models

HPCA 2026

LEGO: Supporting LLM-enhanced Games with One Gaming GPU
Han Zhao^, Weihao Cui^, Zeshen Zhang, Wenhao Zhang, Jiangtong Li, Quan Chen, Pu Pang, Zijun Li, Zhenhua Han, Yuqing Yang, Minyi Guo.
32nd IEEE International Symposium on High-Performance Computer Architecture (HPCA 2026)
Paper

Education

New York University, Tandon School of Engineering
M.S. in Computer Science, expected 2026

Tongji University
B.S. in Computer Science and Technology, 2025