I am a second-year PhD student at the School of Electronic Information and Electrical Engineering (SEIEE), Shanghai Jiao Tong University. I work at the intersection of AI (LLM) for EDA and face anti-spoofing and biometric security. I am fortunate to be advised by Professor Lei He and Associate Professor Ting-Jung Lin.

My recent work focuses on building datasets, benchmarks, and multimodal reasoning systems for AMS circuit understanding and auto-design, while also exploring explainable methods for face anti-spoofing and forgery detection.

If you are interested in discussing research ideas or potential collaborations, feel free to reach out at shiyichen@sjtu.edu.cn.

Research Interests

• AI (LLM) for EDA: applying machine learning, optimization, and multimodal large language models to accelerate analog and mixed-signal circuit analysis, schematic understanding, and circuit auto-design.
• Face Anti-Spoofing and Biometric Security: building explainable and robust multimodal systems for spoofing and forgery detection.

Selected Publications

• FAS-R1: A Unified Multi-Task MLLM for Reasoning Face Anti-Spoofing, arXiv 2026
• UniShield: Unified Face Attack Detection via KG-Informed Multimodal Reasoning, arXiv 2026
• FaceShield: Explainable Face Anti-Spoofing with Multimodal Large Language Models, AAAI 2026
• SHIELD: an evaluation benchmark for face spoofing and forgery detection with multimodal large language models, Visual Intelligence 2025
• AMSnet-KG: A Netlist Dataset for LLM-based AMS Circuit Auto-Design Using Knowledge Graph RAG, ACM TODAES 2025
• AMSbench: A Comprehensive Benchmark for Evaluating MLLM Capabilities in AMS Circuits, arXiv 2025
• AMSNet: Netlist Dataset for AMS Circuits, LAD 2024 Best Paper Award

News

• July 2026: FAS-R1 was released on arXiv.
• May 2026: UniShield was released on arXiv.
• March 2026: FaceShield was published in AAAI 2026.
• October 2025: AMSnet-KG was published in ACM TODAES.
• October 2025: Symbol and Footprint Database for Electronic Components by Agentic Recognition and Generation was accepted to PRCV 2025.
• May 2025: AMSbench was released on arXiv.
• May 2025: AMSnet 2.0 was accepted to LAD 2025.
• June 2025: SHIELD was published in Visual Intelligence.
• May 2024: AMSNet was accepted to LAD 2024 and received the Best Paper Award.

Academic Service

I co-founded the Future EDA and AI Techniques Seminar (FEATs), an online seminar jointly initiated by PhD students from East Institute of Technology, Shanghai Jiao Tong University, and UCLA. FEATs is designed to encourage academic exchange among early-career researchers working at the intersection of AI and EDA, and to provide a collaborative venue for sharing ideas, discussing recent advances, and fostering new connections.

You can find FEATs here.