Zhuofu Chen

Hi there! I am a first-year CS Ph.D. student at Princeton advised by Prof. Ravi Netravali. I obtained my Bachelor's degree from Tongji University, enrolled in Computer Science Elite Class 2025.

Previously, I was fortunate to work with Prof. Zhihao Jia at CMU Catalyst as a research intern. I also had a wonderful internship at SJTU IPADS advised by Prof. Xingda Wei and Prof. Rong Chen. My academic journey began at Tongji University, under the mentorship of Prof. Zhijun Ding.

Feel free to drop me an e-mail if you would like to connect!

~ Email | CV | Github | Scholar ~

Interests

My general purpose is to redesign next-generation datacenter/cloud operating systems to bridge the gap between evolving hardware and emerging needs of software, with a current focus on serving machine learning workloads and supporting compound AI systems.

Publications

(* indicates equal contribution)

Kimi K2: Open Agentic Intelligence
Kimi Team (was part of the project while interning at Kimi in spring 2025)
[paper] [code]
Large model architecture; Training.

AdaServe: SLO-Customized LLM Serving with Fine-Grained Speculative Decoding
Zikun Li*, Zhuofu Chen*, Remi Delacourt, Gabriele Oliaro, Zeyu Wang, Qinghan Chen, Shuhuai Lin, April Yang, Zhihao Zhang, Zhuoming Chen, Sean Lai, Xupeng Miao, Zhihao Jia
under review. [paper] [code]
Large language model serving; SLO customization; Speculative decoding.

Characterizing Network Requirements for GPU API Remoting in AI Applications
Tianxia Wang*, Zhuofu Chen*, Xingda Wei, Jinyu Gu, Rong Chen, Haibo Chen
under review. [paper] [code]
GPU disaggregation; Transparent API remoting; Proxy Optimization.

TidalDecode: Fast and Accurate LLM Decoding with Position Persistent Sparse Attention
Lijie Yang*, Zhihao Zhang*, Zhuofu Chen, Zikun Li, Zhihao Jia
International Conference on Learning Representations (ICLR), 2025. [paper] [code]
Sparse attention; Efficient decoding.

Updated at Aug. 2025

This web is a modification to Rishab Khincha's