Ziyang Yu | Emory University

About Me

Welcome! I’m Ziyang Yu, a PhD student at Emory University advised by Dr. Liang Zhao. Before this, I was a senior machine learning engineer at Alibaba Group. I obtained my M.Eng degree in Electrical and Computer Engineering from University of Waterloo.

My research encompasses Large Language Models, and Graph Neural Networks. I hold a strong academic background and have contributed to multiple research projects, constantly pushing the boundaries of what’s possible in the machine learning field.

I am open to potential positions in the China/Canada/United States, including internships and full-time roles. If you’re interested in collaborating or discussing opportunities, please feel free to reach out.

Research Interests

Graph Neural Networks: Distributed training of GNN.
Large Language Models: Knowledge Distillation, Agentic System, Code Generation.

Projects

ResearchAtlas: A coverage-oriented, graph-based academic paper discovery non-commercial website I designed and built to improve literature review beyond traditional Top-K ranking search. You are welcome to visit the project website for more details.

LLMCostCut: A framework for selective LLM invocation and online distillation to reduce the cost of using large language models while maintaining accuracy. It routes queries through a small student model and falls back to the teacher LLM only when the student is uncertain, achieving around 10× cost reduction and nearly 1000× speedup with accuracy up to 95%. You can find the code and documentation on GitHub.

Research

Distilling LLM Reasoning into Graph of Concept Predictors

Ziyang Yu, Liang Zhao

Under Review

PDF

Experience Retrieval-Augmentation with Electronic Health Records Enables Accurate Discharge QA

Justice Ou, Tinglin Huang, Yilun Zhao, Ziyang Yu, Peiqing Lu

KDD 2025 Workshop on Prompt Optimization

PDF Code

Efficient Imputation for Patch-based Tabular Missing Data via Cluster-regularized Optimal Transport

Yuyu Liu, Jiannan Yang, Ziyang Yu, Weishen Pan, Fei Wang, Tengfei Ma

Under Review

PDF

Staleness-Alleviated Distributed GNN Training via Online Dynamic-Embedding Prediction

Guangji Bai*, Ziyang Yu*, Zheng Chai, Yue Cheng, Liang Zhao

SIAM International Conference on Data Mining (SDM), 2025.

PDF Code

Beyond Efficiency: A Systematic Survey of Resource-Efficient Large Language Models

Guangji Bai, Zheng Chai, Chen Ling, Shiyu Wang, Jiaying Lu, Nan Zhang, Tingwei Shi, Ziyang Yu, Mengdan Zhu, Yifei Zhang, Carl Yang, Yue Cheng, Liang Zhao

Under Review of ACM Computing Surveys

PDF Project Page

Education Experience

	Emory University, 2025.08-Present Ph.D. in Computer Science
	University of Waterloo, 2023.09-2025.01 M.Eng. in Electrical and Computer Engineering
	Southern University of Science and Technology, 2019.08-2023.06 B.S. in Mathematics and Applied Mathematics (Fields Honors Class)

Industrial Experience

	Alibaba Group, 2025.02-2025.08 Senior Machine Learning Engineer
	Baidu, 2024.01-2024.03 Machine Learning Engineer Intern
	Biogeometry, 2023.04-2023.07 Backend Engineer Intern

Services

Conference Reviewers/Program Committee

SIAM International Conference on Data Mining (SDM25)