Yu Zhang

Ph.D. in Computer Science and Engineering

The Chinese University of Hong Kong
Email:yuzhang712@link.cuhk.edu.hk

Biography

I graduated with a Ph.D. in Computer Science and Engineering from The Chinese University of Hong Kong (CUHK) in July 2026, advised by Prof. Bei Yu. I am currently a Member of Technical Staff at Agentrys, where I focus on design verification and agentic EDA. My research interests include efficient and reliable LLM training and inference, agentic EDA, and hardware–software co-design. Outside of work, I enjoy swimming and hiking.

Research Directions

LLM Agent and Reasoning: [C12, C13, C14]
LLM Efficiency: [C9, C10, C11]
EDA: [C4, C5, C6, C7, C8, J2]

Publications

Conference Papers

[C14] Jenny Zhang, NVIDIA Cosmos Team, “Cosmos 3: Omnimodal World Models for Physical AI”, NVIDIA Technical Report, Technical Report.
[C13] Yu Zhang, Peter Belcak, Shizhe Diao, Yonggan Fu, Shaona Ghosh, Morteza Mardani, David Austin, Eileen Long, Bei Yu, Pavlo Molchanov, “PROBE: PROcess-Based BEnchmark for Hallucination Detection”, Findings of the Association for Computational Linguistics (ACL).
[C12] Peter Belcak, Yu Zhang, Shizhe Diao, David Austin, Yonggan Fu, Ryan Angilly, Yingyan Celine Lin, Eileen Long, Pavlo Molchanov, “HallucinationHunter: Fine-Grained Factual Grounding of Generated Text”, ICLR 2026 Workshop Reliable Autonomy.
[C11] Yu Zhang, Hui-Ling Zhen, Mingxuan Yuan, Bei Yu, “MOSS: Efficient and Accurate FP8 LLM Training with Microscaling and Automatic Scaling”, International Conference on Learning Representations (ICLR).
[C10] Yu Zhang, Mingzi Wang, Lancheng Zou, Wulong Liu, Hui-Ling Zhen, Mingxuan Yuan, Bei Yu, “MixPE: Quantization and Hardware Co-design for Efficient LLM Inference”, arXiv.
[C9] Yu Zhang, Hui-Ling Zhen, Zehua Pei, Yingzhao Lian, Lihao Yin, Mingxuan Yuan, Bei Yu, “DiLA: Enhancing LLM Tool Learning with Differential Logic Layer”, ACM Conference on Knowledge Discovery and Data Mining (KDD).
[C8] Yu Zhang, Hui-Ling Zhen, Mingxuan Yuan, Bei Yu, “DiffSAT: Differential MaxSAT Layer for SAT Solving”, IEEE/ACM International Conference on Computer-Aided Design (ICCAD).
[C7] Yuan Pu, Fangzhou Liu, Yu Zhang, Zhuolun He, Kai-Yuan Chao, Yibo Lin, Bei Yu, “Lesyn: Placement-aware Logic Resynthesis for Non-Integer Multiple-Cell-Height Designs”, ACM/IEEE Design Automation Conference (DAC).
[C6] Yu Zhang, Yuan Pu, Fangzhou Liu, Peiyu Liao, Kai-Yuan Chao, Keren Zhu, Yibo Lin and Bei Yu, “Multi-Electrostatics Based Placement for Non-Integer Multiple-Height Cells”, International Symposium on Physical Design (ISPD).
[C5] Yu Zhang, Wei Lin, Sisi Chen, Qingyu Song, Jiaxun Lu, Yunfeng Shao, Bei Yu and Hong Xu, “Fed2Com: Towards Efficient Compression in Federated Learning”, International Conference on Computing, Networking and Communications (ICNC).
[C4] Yu Zhang, Yifan Chen, Zhonglin Xie, Hong Xu, Zaiwen Wen, Yibo Lin and Bei Yu, “LRSDP: Low-rank SDP for Triple Patterning Lithography Layout Decomposition”, Design Automation Conference (DAC).
[C3] Xudong Wang, Guoming Tang, Yi Wang, Srinivasan Keshav, Yu Zhang, “Evsense: A robust and scalable approach to non-intrusive ev charging detection”, ACM International Conference on Future Energy Systems (e-Energy).
[C2] Yu Zhang, Guoming Tang, Qianyi Huang, Yi Wang, Hong Xu, Xudong Wang, “More Behind Your Electricity Bill: a Dual-DNN Approach to Non-Intrusive Load Monitoring”, IEEE International Conference on Cyber, Physical and Social Computing (CPSCom).
[C1] Yu Zhang, Guoming Tang, Qianyi Huang, Kui Wu, Yangjing Wu, Yi Wang, “Investigating Low-Battery Anxiety of Mobile Users”, IEEE International Conference on Cyber, Physical and Social Computing (CPSCom).

Journal Papers

[J2] Jiahao Xu, Yuan Pu, Yu Zhang, Fangzhou Liu, Wenjian Yu, Kai-Yuan Chao, Bei Yu, “Easyn: Resynthesis and Legalization Joint NIMCH Design Placement”, IEEE Transactions on Computer-Aided Design of Integrated Circuits and Systems (TCAD).
[J1] Yu Zhang, Guoming Tang, Qianyi Huang, Yi Wang, Kui Wu, Keping Yu, Xun Shao, “FedNILM: Applying Federated Learning to NILM Applications at the Edge”, IEEE Transactions on Green Communications and Networking (TGCN).

Patent

“一种基于模型松弛的大模型新微调方法”, 92056131CN01.

Experiences

Member of Technical Staff, Agentrys Inc., Santa Clara, U.S., Oct 2026 - Present
Topic: Agentic EDA, LLM for Verification
Research Intern, NVIDIA Research, Santa Clara, U.S., Sep 2025 - Feb 2026
Topic: Efficient Large Language Model, LLM Hallucination Detection
Research Intern, Huawei Noah's Ark Lab, HKSAR, Aug 2023 - Aug 2024
Topic: Hardware/software co-design for efficient LLM inference; Differential solver layer for LLM
Research Assistant, The Chinese University of Hong Kong, HKSAR, Nov 2021 - Jul 2022
Topic: Communication-efficient and convergence fast federated learning
Research Intern, Microsoft Research Asia, Beijing, P.R. China, Jan 2021 - Aug 2021
Topic: AISimulator

Education

Ph.D. Computer Science and Engineering, The Chinese University of Hong Kong, Aug 2022 - Jul 2026
B.S. Statistics, Central South University, Sept 2015 - Jun 2019