Yuanbin Wang

I am an undergraduate student majoring in Electronic Science and Technology at Shanghai Jiao Tong University.I am currently working with Prof. Li Wei at the National University of Singapore.

My current research interest is AI for Electronic Design Automation (EDA), especially AI agents that can interact with EDA tools for circuit design, simulation, verification, and design-space optimization.

With a background in digital and analog electronics, embedded systems, machine learning, and industrial AI, I am interested in the intersection of intelligent agents and hardware design.

Email  /  Google Scholar  /  Github

Research

I am interested in AI for electronic design automation, hardware-aware foundation models, tool-using agents, circuit design and verification, design-space exploration, and machine learning for electronic manufacturing. My current focus is to explore how AI agents can understand design specifications, use EDA tools, analyze simulation feedback, and iteratively improve hardware designs.

We propose a hierarchical video representation based on spatio-temporally consistent proxy nodes, enabling robust modeling of dynamic scenes and controllable video editing under motion, occlusion, and viewpoint changes.

We propose a variational perturbation method for personalized federated learning under statistical heterogeneity. The method compares prior and posterior data distributions to improve the robustness of model updates.

We propose an LLM-agent-driven scheduling system for asynchronous federated learning. GFLAgent performs client selection and tier assignment using real-time and historical signals to improve time-to-accuracy and training efficiency.

Last updated: 2026