I am a Ph.D. student in Computer Engineering at the University of California, Santa Barbara, proudly advised by Prof. Zheng Zhang, where I also received my M.S. in Computer Engineering. Before that, I received my B.S. in Electronic Information Engineering from Huazhong University of Science & Technology.
My research is on zeroth-order and memory-efficient optimization — what remains possible when gradients are expensive, unreliable, or simply unavailable. On the machine learning side, I design gradient-free training algorithms that fine-tune large language models at inference-level memory, and work more broadly on optimizers and parameter-efficient methods for large-scale training. On the hardware side, I bring the same machinery to analog/RF circuit design, where the simulator is a black box and every evaluation is expensive; I am currently extending this toward uncertainty-aware design under process variations.
I am also broadly interested in agentic LLMs — in particular how tool-using agents can be brought into EDA workflows — as well as the pre- and post-training of large language models and hardware/software co-design.
You can find more details on Google Scholar and my CV.
Education
- 2025.04 - 2028 (expected), Ph.D. in Computer Engineering, UC Santa Barbara
- 2023.09 - 2025.03, M.S. in Computer Engineering, UC Santa Barbara
- 2019.09 - 2023.06, B.S. in Electronic Information Engineering, Huazhong University of Science & Technology
News
- 2026.08: GRZO: Group-Relative Zeroth-Order Optimization for Large Language Model Fine-Tuning accepted to Findings of the Association for Computational Linguistics: EMNLP 2026.
Experience
- 2025.06 - 2025.09, Software Architect Intern, Cadence Design Systems, Austin, TX
- LLM copilot agent for Voltus, Cadence’s power-integrity signoff solver: natural-language design intent into verified tool commands, GUI actions, and automated root-cause analysis.
- On-premise deployment under enterprise data constraints, via retrieval over EDA documentation, parameter-efficient fine-tuning, and teacher-to-student distillation.
Publications
Circuit Design Optimization
Liyan Tan, Yequan Zhao, Ben F. Jamroz, Ari Feldman, Zheng Zhang
Analog yield optimization · Process variation · Stochastic zeroth-order optimization
ZOAF: Towards Efficient Zeroth-Order Optimization for Analog/RF Circuit Design
Liyan Tan, Yequan Zhao, Jinming Lu, Ben F. Jamroz, Ari Feldman, Zheng Zhang
Analog/RF circuit sizing · Simulation-Efficient Optimization
Machine Learning Optimization
GRZO: Group-Relative Zeroth-Order Optimization for Large Language Model Fine-Tuning
Liyan Tan, Yequan Zhao, Yifan Yang, Ruijie Zhang, Xinling Yu, Zheng Zhang
LLM fine-tuning · Zeroth-order optimization · Variance reduction
IAPO: Input Attribution-Aware Policy Optimization for Tool Use in Small Multimodal Agents
Yifan Yang, Zhen Zhang, Jiayi Tian, Liyan Tan, Zheng Zhang
Multimodal Tool Use · Input Attribution · Policy Optimization
FuRA: Full-Rank Parameter-Efficient Fine-Tuning with Spectral Preconditioning
Yequan Zhao, Ruijie Zhang, Liyan Tan, Niall Moran, Tong Qin, Zheng Zhang
PEFT · Full-rank adaptation · Spectral preconditioning
MUON+: Towards More Effective Muon via One Additional Normalization Step for LLM Pre-training
Ruijie Zhang, Yequan Zhao, Ziyue Liu, Zhengyang Wang, Yupeng Su, Liyan Tan, Zheng Zhang
LLM pre-training · Muon · Normalization