About

I am a first-year PhD student in the Department of Electrical and Computer Engineering at the University of Minnesota, advised by Prof. Mingyi Hong. I received my B.S. in Computer Science from East China University of Science and Technology in 2025.

My research interests focus on LLM Agents, especially post-training (RLVR/Agentic RL) and Multi-agent systems for General/Coding Agents.

You can find my CV here: Zijian Zhang’s Curriculum Vitae.

Education

• University of Minnesota, MN, USA (2025.09 – 2030.06)
  Full-funded PhD Student in Department of Electrical and Computer Engineering
  Advised by Prof. Mingyi Hong | GPA: 4/4

• East China University of Science and Technology, Shanghai, China (2021.09 – 2025.06)
  B.S. in Computer Science | GPA: 88/100

Publications

1. StitchCUDA: An Automated Multi-Agent End-to-End GPU Programming Framework with Rubric-based Agentic Reinforcement Learning
   Shiyang Li*, Zijian Zhang* (Co-First Author), Winson Chen, Yuebo Luo, Mingyi Hong, Caiwen Ding
   (Under review) ICML, 2026

2. CudaForge: An Agent Framework with Hardware Feedback for CUDA Kernel Optimization
   Zijian Zhang, Rong Wang, Shiyang Li, Yuebo Luo, Mingyi Hong, Caiwen Ding
   (Under review) ICML, 2026

3. InfantAgent-Next: A Multimodal Generalist Agent for Automated Computer Interaction
   Bin Lei, Weitai Kang, Zijian Zhang, Winson Chen, Xi Xie, Shan Zuo, Mimi Xie, Ali Payani, Mingyi Hong, Yan Yan, Caiwen Ding
   NeurIPS, 2025

4. GRAPE: Generalizing Robot Policy via Preference Alignment
   Zijian Zhang, Kaiyuan Zheng, Zhaorun Chen, Joel Jang, Yi Li, Chaoqi Wang, Mingyu Ding, Dieter Fox, Huaxiu Yao
   ICRA 2026; ICLR Workshop, 2025

5. AnyPrefer: An Automatic Framework for Preference Data Synthesis
   Yiyang Zhou, …, Zijian Zhang, …, Huaxiu Yao
   ICLR, 2025

Research Experience

• PhD Student, Mingyi Hong’s Lab @ University of Minnesota (2025.08 – present)
  Working on multi-agent systems for CUDA code generation (StitchCUDA, CudaForge) and multimodal agents (InfantAgent-NEXT).

• Research Assistant, Huaxiu Yao’s Lab @ University of North Carolina at Chapel Hill (2024.05 – 2025.01)
  Introduced AnyPrefer and GRAPE for VLA model post-training via preference optimization.

• Research Assistant, Machine Learning Group @ Microsoft Research Asia (2024.10 – 2025.04)
  Worked on LLM training optimization, math reasoning enhancement, and training dynamics analysis.

• Research Assistant, InternLM2 Team @ Shanghai AI Lab (2023.11 – 2024.05)
  Proposed efficient data selection methods for InternLM2 SFT and contributed to InternLM/InternLM2 development.

Contact

Email / GitHub / Google Scholar