Zhixu Du

PhD candidate in Electrical and Computer Engineering at Duke University. I work on efficient machine learning systems (Mixture-of-Experts serving and load balancing, state space models), federated learning, and auditing of autonomous multi-agent AI systems.

GitHub Semantic Scholar

Publications

Under review
What Does a Token Cost? A Mixture-of-Agents Measurement of Sufficient Per-Token Compute
Zhixu Du, Weijia Han, Hai (Helen) Li, Yiran Chen
arXiv PDF Project page
MLSys 2024
SiDA: Sparsity-Inspired Data-Aware Serving for Efficient and Scalable Large Mixture-of-Experts Models
Zhixu Du, Shiyu Li, Yuhao Wu, Xiangyu Jiang, Jingwei Sun, Qilin Zheng, Yongkai Wu, Ang Li, Hai (Helen) Li, Yiran Chen
arXiv PDF OpenReview Code Project page
DistributedML @ CoNEXT 2022
Rethinking Normalization Methods in Federated Learning
Zhixu Du, Jingwei Sun, Ang Li, Pin-Yu Chen, Jianyi Zhang, Hai (Helen) Li, Yiran Chen
arXiv PDF Publisher Code Project page
NeurIPS 2025 Workshop (ML for Systems)
MoE-GPS: Guidlines for Prediction Strategy for Dynamic Expert Duplication in MoE Load Balancing
Haiyue Ma, Zhixu Du, Yiran Chen
arXiv PDF OpenReview Project page
arXiv preprint 2026
AUDITA: certified auditing and causal attribution of adverse outcomes in autonomous multi-agent systems
Zhixu Du, Yiran Chen
arXiv PDF Code Project page
NeurIPS 2026
Swimba: Switch Mamba Model Scales State Space Models
Zhixu Du, Krishna Teja Chitty-Venkata, Murali Emani, Venkatram Vishwanath, Hai (Helen) Li, Yiran Chen
arXiv PDF Project page

Last updated 2026-10-22.