Hi there! 👋 I graduated from South China University of Technology (SCUT) in June 2026 with a B.Eng. in Control Science and Engineering and a B.Sc. in Mathematics and Applied Mathematics.
I am currently spending my gap year deepening my research experience as a research intern in the Visual Computing Group at Microsoft Research Asia (MSRA), advised by Bei Liu, where I focus on LLM agents. Previously, I was fortunate to be advised by Prof. Zhongxiang Dai at CUHK-Shenzhen, where I worked on multi-agent systems and prompt engineering.
My research interests lie at the intersection of large language models (LLMs) and machine learning, with a focus on LLM agents, embodied agents, multi-agent systems, and bandit algorithms. I am actively seeking PhD opportunities where I can further develop these interests.
Feel free to reach out — I’d be happy to discuss research ideas, explore potential collaborations, or learn about relevant PhD openings. Let’s see what we can create together!
📖 Educations
B.Eng. in Control Science & Engineering.
B.Sc. in Mathematics and Applied Mathematics.
GPA: 3.84/4.0 · Rank: 3/24.
🔥 News
- 2026.06: 🎉🎉 Our paper MASPOB is accepted at ICML 2026 as a Spotlight (Top 2.2% of 23,918 submissions)!
📝 Publications
MASPOB: Bandit-Based Prompt Optimization for Multi-Agent Systems with Graph Neural Networks
Zhi Hong*, Qian Zhang*, Jiahang Sun, Zhiwei Shang, Mingze Kong, Xiangyi Wang, Yao Shu†, Zhongxiang Dai†
ICML 2026 Spotlight, Top 2.2% of 23,918 submissions
Preprints & Under Review
AgentSpec: Understanding Embodied Agent Scaffolds Through Controlled Composition
Jixuan Chen, Jianzhi Shen, Haoqiang Kang, Zhi Hong, Qingyi Jiang, Soham Bose, Yiming Zhang, Leon Leng, Amit Vyas, Lingjun Mao, Siru Ouyang, Kun Zhou, Lianhui Qin
Submitted to EMNLP 2026
FedPOB: Sample-Efficient Federated Prompt Optimization via Bandits
Pingchen Lu*, Zhi Hong*, Zhiwei Shang, Zhiyong Wang, Yikun Ban, Yao Shu, Min Zhang, Shuang Qiu, Zhongxiang Dai†
.png)
Workflow-R1: Group Sub-sequence Policy Optimization for Multi-turn Workflow Construction
Mingze Kong, Zikun Qu, Zhongquan Zhou, Pengyu Liang, Xiang Li, Zhiwei Shang, Zhi Hong, Kaiyu Huang, Zhiyong Wang, Zhongxiang Dai†
- Linear and Neural Dueling Bandits with Delayed Feedback
Xiangyi Wang, Pingchen Lu, Jie Mao, Mingze Kong, Zhi Hong, Zhiyong Wang, Zhongxiang Dai
💻 Research Experience
-
2026.07 – Present, Research Intern – LLM Agents, advised by Bei Liu, Microsoft Research Asia (MSRA), Visual Computing Group.
Conducted research on LLM Agents, including Social Simulation and Automated Research. -
2026.01 – 2026.07, Research Intern – Embodied Agents, advised by Prof. Lianhui Qin, Computer Science and Engineering, UC San Diego.
Conducted research on modular architectures to enhance capabilities of Embodied LLM-Based Agents. -
2024.09 – 2026.07, Research Assistant – Large Language Models, advised by Prof. Zhongxiang Dai, School of Data Science, CUHK-Shenzhen.
Conducted research on Multi-Agent Systems and Prompt Engineering for LLM-based systems.
🎖 Honors and Awards
- 2024 First Prize – China Mathematics Competition (CMC, Category A, Province Level)
- 2023 – 2025 National Encouragement Scholarship – Ministry of Education, China
- 2023 – 2025 Merit Student – South China University of Technology
💬 Talks
- 2026.07: Invited talk at Shanghai Jiao Tong University on Multi-Agent Systems (MAS).
🛠 Skills & Interests
- Languages: English (TOEFL 98), Chinese (Native).
Programming: Python, LaTeX, C++, MATLAB. - Interests: Traveling, Basketball, Swimming.