2026 Yuxiong He: Computer Science Researcher – H-Index, Publications & Awards

Discipline name	D-Index	World Ranking	Current World Ranking	National Ranking	Current National Ranking	Publications	Citations
Computer Science	41	8929	8678	3800	3650	150	6066

Overview

Yuxiong He is a researcher affiliated with Microsoft in the United States, specializing in the field of Computer Science with a focus on Artificial Intelligence. Their scholarly contribution includes over 120 publications, predominantly in areas related to advanced neural network applications and natural language processing techniques.

The primary research subfields covered by Yuxiong He include:

Artificial Intelligence
Computer Vision and Pattern Recognition
Electrical and Electronic Engineering
Molecular Biology
Hardware and Architecture

The main topics addressed in their work span:

Topic Modeling
Advanced Neural Network Applications
Natural Language Processing Techniques
Multimodal Machine Learning Applications
Stochastic Gradient Optimization Techniques
Domain Adaptation and Few-Shot Learning
Ferroelectric and Negative Capacitance Devices

Yuxiong He has published extensively in several venues, with the highest number of publications appearing in arXiv (Cornell University). Other notable publication venues include:

arXiv (Cornell University)
Proceedings of the AAAI Conference on Artificial Intelligence
Nature Methods
IEEE Transactions on Pattern Analysis and Machine Intelligence
Proceedings of the Fifteenth ACM International Conference on Web Search and Data Mining

Representative recent papers authored by Yuxiong He include:

Using DeepSpeed and Megatron to Train Megatron-Turing NLG 530B, A Large-Scale Generative Language Model (2022, arXiv)
OpenFold: retraining AlphaFold2 yields new insights into its learning mechanisms and capacity for generalization (2024, Nature Methods)
ZeroQuant: Efficient and Affordable Post-Training Quantization for Large-Scale Transformers (2022, arXiv)
ZeRO-Offload: Democratizing Billion-Scale Model Training (2021, arXiv)
ZeRO-Infinity: Breaking the GPU Memory Wall for Extreme Scale Deep Learning (2021, arXiv)

Frequent collaborators who have coauthored multiple works with Yuxiong He include:

Minjia Zhang
Zhewei Yao
Samyam Rajbhandari
Xiaoxia Wu
Ammar Ahmad Awan

Best Publications

ZeRO: Memory optimizations Toward Training Trillion Parameter Models

Samyam Rajbhandari;Jeff Rasley;Olatunji Ruwase;Yuxiong He

721
Citations
DeepSpeed: System Optimizations Enable Training Deep Learning Models with Over 100 Billion Parameters

Jeff Rasley;Samyam Rajbhandari;Olatunji Ruwase;Yuxiong He

646
Citations
DeepSpeed- Inference: Enabling Efficient Inference of Transformer Models at Unprecedented Scale

Unknown

267
Citations
ZeRO-infinity: breaking the GPU memory wall for extreme scale deep learning

Samyam Rajbhandari;Olatunji Ruwase;Jeff Rasley;Shaden Smith

246
Citations
Graph query processing using plurality of engines

Sameh Elnikety;Yuxiong He;Sherif Sakr

218
Citations
ZeroQuant: Efficient and Affordable Post-Training Quantization for Large-Scale Transformers

Unknown

206
Citations
Adaptive work-stealing with parallelism feedback

Kunal Agrawal;Charles E. Leiserson;Yuxiong He;Wen Jing Hsu

154
Citations
Provably-Efficient Job Scheduling for Energy and Fairness in Geographically Distributed Data Centers

Shaolei Ren;Yuxiong He;Fei Xu

150
Citations
The Cilkview scalability analyzer

Yuxiong He;Charles E. Leiserson;William M. Leiserson

144
Citations
Learning Intrinsic Sparse Structures within Long Short-Term Memory

Wei Wen;Yuxiong He;Samyam Rajbhandari;Minjia Zhang

124
Citations
Few-to-Many: Incremental Parallelism for Reducing Tail Latency in Interactive Services

E. Haque;Yong hun Eom;Yuxiong He;Sameh Elnikety

124
Citations
Predictive parallelization: taming tail latencies in web search

Myeongjae Jeon;Saehoon Kim;Seung-won Hwang;Yuxiong He

116
Citations
Swayam: distributed autoscaling to meet SLAs of machine learning inference services with resource efficiency

Arpan Gujarati;Sameh Elnikety;Yuxiong He;Kathryn S. McKinley

110
Citations
Performance Modeling and Scalability Optimization of Distributed Deep Learning Systems

Feng Yan;Olatunji Ruwase;Yuxiong He;Trishul Chilimbi

106
Citations
ZeRO-Offload: Democratizing Billion-Scale Model Training

Jie Ren;Samyam Rajbhandari;Reza Yazdani Aminabadi;Olatunji Ruwase

105
Citations
Adaptive Scheduling with Parallelism Feedback

Kunal Agrawal;Y. He;W.-J. Hsu;C.E. Leiserson

100
Citations
Zeta: scheduling interactive services with partial execution

Yuxiong He;Sameh Elnikety;James Larus;Chenyu Yan

92
Citations
DeepCPU: serving RNN-based deep learning models 10x faster

Minjia Zhang;Samyam Rajbhandari;Wenhan Wang;Yuxiong He

80
Citations
Mercury: A memory-constrained spatio-temporal real-time search on microblogs

Amr Magdy;Mohamed F. Mokbel;Sameh Elnikety;Suman Nath

76
Citations
G-SPARQL: a hybrid engine for querying large attributed graphs

Sherif Sakr;Sameh Elnikety;Yuxiong He

75
Citations
ZeRO: Memory Optimization Towards Training A Trillion Parameter Models.

Samyam Rajbhandari;Jeff Rasley;Olatunji Ruwase;Yuxiong He

69
Citations
Adaptive work stealing with parallelism feedback

Kunal Agrawal;Yuxiong He;Charles E. Leiserson

62
Citations

Frequent Co-Authors

Sameh Elnikety Microsoft (United States)

Shaolei Ren University of California, Riverside

Charles E. Leiserson MIT

Kathryn S. McKinley Google (United States)

Mohamed F. Mokbel University of Minnesota

Suman Nath Microsoft (United States)

Jianfeng Gao Microsoft (United States)

Scott Rixner Rice University

Alan L. Cox Rice University

David G. Andersen Carnegie Mellon University

External Links

Personal website of Yuxiong He

If you think any of the details on this page are incorrect, let us know.

Related Online Degrees & Career Pathways

Exploring computer science opens doors to a wide array of online degree options and diverse career pathways. Many students interested in technology and engineering will find related programs that combine strong technical skills with real-world application.

For those drawn to environmental impact, an online environmental engineering degree blends engineering principles with ecological awareness, preparing graduates to tackle sustainability challenges.

Students considering advanced studies in engineering fields may benefit from enrolling in the cheapest online master's mechanical engineering programs, which provide cost-effective education in design and manufacturing technologies.

If you have a passion for the foundations of technology and research, the best online physics degree programs offer flexibility while covering critical topics in quantum mechanics, electronics, and more.

As data analysis becomes central to modern business, students can also consider specialized data science degrees that prepare graduates for highly sought-after careers in analytics, artificial intelligence, and business intelligence.