Mingzhen Li

About

I am currently a postdoctoral researcher at the Chinese University of Hong Kong (CUHK), supervised by Prof. Bei Yu. My research focuses on electronic design automation (EDA) for analog circuits, with particular emphasis on representing analog design knowledge for AI —making domain expertise accessible to and actionable by AI systems.

Research Overview

Analog circuit design relies heavily on expert knowledge to navigate a vast and complex design space. By making such knowledge explicit and traceable, reasoning can guide design decisions and reduce unnecessary simulation-driven exploration. We therefore study how to represent this knowledge for AI:

1. Symbolic analysis: generates symbolic expressions as explicit evidence for AI reasoning.
2. Topology abstraction: extracts analog-specific signal paths for graph learning.
3. Knowledge refinement: cleans, structures, and aligns design assets across multiple views for IP reuse.

Projects

Publications

Journal Papers

• [J4] Yapeng Li, Peng Xu, Mingzhen Li, Tsung-Yi Ho, Bei Yu, and Tinghuan Chen, “Analog Circuit Representation Learning Informed by Kirchhoff’s Current Law,” IEEE Transactions on Computer-Aided Design of Integrated Circuits and Systems, accepted July 2026. Available here.
• [J3] Mingzhen Li, Bo Li, Bei Yu, and Guoyong Shi, “Compensation-Aware Topology Synthesis of Multistage Op-Amp,” ACM Transactions on Design Automation of Electronic Systems, published online August 4, 2026. Available here.
• [J2] Mingzhen Li and Guoyong Shi, “A Signal-Path Recognition Approach to Multistage Op-Amp Pole-Zero Extraction With Applications,” IEEE Transactions on Computer-Aided Design of Integrated Circuits and Systems, vol. 45, no. 7, pp. 3045–3058, 2026. Available here.
• [J1] Mingzhen Li, Xisheng Zhang, and Guoyong Shi, “Op-Amp sizing via behavioral constraint generation and Gm/ID sampling,” Integration, the VLSI Journal, vol. 105, Art. no. 102503, 2025. Available here.

Conference Papers

• [C4] Mingzhen Li, Bo Li, Tingjie Yang, and Guoyong Shi, “PZ-Agent: A Symbolic-Engine Enhanced LLM Agent for Op-Amp Topology Inquisition,” in 2025 International Symposium of Electronics Design Automation (ISEDA), pp. 730–735, 2025. Available here.
• [C3] Xisheng Zhang, Mingzhen Li, Qixu Xie, and Guoyong Shi, “A Subcircuit Matching Approach to Structural Analog Circuit Model Generation and Sizing,” in 2024 2nd International Symposium of Electronics Design Automation (ISEDA), pp. 131–136, 2024. Available here.
• [C2] Mingzhen Li, Limin Hao, Zhihang Wu, Wei Wu, Leibin Ni, and Guoyong Shi, “An Event-Driven Method for Fast Delay Distribution Estimation of Array Multipliers,” in 2023 International Symposium of Electronics Design Automation (ISEDA), pp. 156–161, 2023. Available here.
• [C1] Limin Hao, Mingzhen Li, Zhihang Wu, Wei Wu, Leibin Ni, and Guoyong Shi, “Test Vector Generation for Array-form Multipliers Using a Genetic Algorithm,” in 2023 International Symposium of Electronics Design Automation (ISEDA), pp. 110–115, 2023. Available here.

Education

• Ph.D., Shanghai Jiao Tong University (SJTU), China, 2021–2026
  • Advisor: Prof. Guoyong Shi
• M.S. in Electrical Engineering, Arizona State University (ASU), United States, 2014–2016
• B.Eng. in Automation, Nanjing University of Aeronautics and Astronautics (NUAA), China, 2010–2014

Academic Service

• Reviewer, IEEE Transactions on Computer-Aided Design of Integrated Circuits and Systems

Teaching Experience

• Teaching Assistant, Mixed-Signal Circuit Design and Automation Methods, SJTU (Fall 2023, Fall 2024, Fall 2025)
• Teaching Assistant, Introduction to Design Automation, SJTU (Fall 2022, Spring 2023, Spring 2024)

Related Work Experience

• Research Intern, Technical Committee, Primarius Technologies, 2026
• Digital Front-end Design, Huawei HiSilicon, 2018–2021
  • Patent: “Secure Transmission Method and Apparatus” (CN114465775A; assigned to Huawei)