World's Best Scientists 2026 revealed!
Award Badge
Rising Stars
2025

D-Index & Metrics

Rising Stars

D-Index
40
Citations
9229
World Ranking
642
National Ranking
95

Engineering and Technology

D-Index
36
Citations
8584
World Ranking
8569
National Ranking
2366

Research.com Recognitions

  • 2025 - Research.com Rising Stars Award

Overview

Yuanzhi Li is affiliated with Carnegie Mellon University in the United States and has a research focus primarily within the field of Computer Science, with a particular emphasis on Artificial Intelligence. Their work spans several subfields including Computer Vision and Pattern Recognition, Statistics and Probability, Statistical and Nonlinear Physics, and Information Systems.

Their research topics cover a range of areas in machine learning and AI techniques, featuring:

  • Topic Modeling
  • Natural Language Processing Techniques
  • Stochastic Gradient Optimization Techniques
  • Domain Adaptation and Few-Shot Learning
  • Generative Adversarial Networks and Image Synthesis
  • Neural Networks and Applications
  • Machine Learning and Algorithms

Li has contributed to a variety of recent research papers including:

  • Textbooks Are All You Need II: phi-1.5 technical report, 2023, arXiv (Cornell University)
  • Phi-3 Technical Report: A Highly Capable Language Model Locally on Your Phone, 2024, arXiv (Cornell University)
  • Towards Understanding Ensemble, Knowledge Distillation and Self-Distillation in Deep Learning, 2020, arXiv (Cornell University)
  • Learning and generalization in overparameterized neural networks, going beyond two layers, 2025, arXiv (Cornell University)
  • LoRA: Low-Rank Adaptation of Large Language Models, 2021, arXiv (Cornell University)

The venues where Yuanzhi Li publishes reflect a strong presence in open-access platforms, predominantly the arXiv repository affiliated with Cornell University, where they have over sixty publications. Other publication venues include the SSRN Electronic Journal, Electronic Journal of Statistics, Proceedings of the 44th International Conference on Software Engineering, and Mathematical Programming.

Collaborative efforts feature several frequent coauthors, including:

  • Zeyuan Allen-Zhu
  • Dhruv Malik
  • Quanquan Gu
  • Samy Jelassi
  • Sébastien Bubeck

Yuanzhi Li's research output highlights extensive engagement with topics in deep learning and neural network architectures, participating in advancing understanding and technical developments in ensemble methods, knowledge distillation, and adaptation of large language models.

Best Publications

  • LoRA: Low-Rank Adaptation of Large Language Models.

    Edward J. Hu;Yelong Shen;Phillip Wallis;Zeyuan Allen-Zhu

  • A Convergence Theory for Deep Learning via Over-Parameterization

    Zeyuan Allen-Zhu;Yuanzhi Li;Zhao Song

  • Learning and Generalization in Overparameterized Neural Networks, Going Beyond Two Layers

    Zeyuan Allen-Zhu;Yuanzhi Li;Yingyu Liang

  • Learning Overparameterized Neural Networks via Stochastic Gradient Descent on Structured Data

    Yuanzhi Li;Yingyu Liang

  • Convergence Analysis of Two-layer Neural Networks with ReLU Activation

    Yuanzhi Li;Yang Yuan

  • A Latent Variable Model Approach to PMI-based Word Embeddings

    Sanjeev Arora;Yuanzhi Li;Yingyu Liang;Tengyu Ma

  • Linear Algebraic Structure of Word Senses, with Applications to Polysemy

    Sanjeev Arora;Yuanzhi Li;Yingyu Liang;Tengyu Ma

  • An Alternative View: When Does SGD Escape Local Minima?

    Robert Kleinberg;Yuanzhi Li;Yang Yuan

  • Algorithmic Regularization in Over-parameterized Matrix Sensing and Neural Networks with Quadratic Activations

    Yuanzhi Li;Tengyu Ma;Hongyang Zhang

  • Towards Understanding Ensemble, Knowledge Distillation and Self-Distillation in Deep Learning.

    Zeyuan Allen-Zhu;Yuanzhi Li

  • Towards Explaining the Regularization Effect of Initial Large Learning Rate in Training Neural Networks

    Yuanzhi Li;Colin Wei;Tengyu Ma

  • Algorithmic Framework for Model-based Deep Reinforcement Learning with Theoretical Guarantees.

    Yuping Luo;Huazhe Xu;Yuanzhi Li;Yuandong Tian

  • On the Convergence Rate of Training Recurrent Neural Networks

    Zeyuan Allen-Zhu;Yuanzhi Li;Zhao Song

  • What Can ResNet Learn Efficiently, Going Beyond Kernels?

    Zeyuan Allen-Zhu;Yuanzhi Li

  • NEON2: Finding Local Minima via First-Order Oracles

    Zeyuan Allen-Zhu;Yuanzhi Li

  • LazySVD: Even Faster SVD Decomposition Yet Without Agonizing Pain

    Zeyuan Allen-Zhu;Yuanzhi Li

  • Feature Purification: How Adversarial Training Performs Robust Deep Learning

    Zeyuan Allen-Zhu;Yuanzhi Li

  • Much Faster Algorithms for Matrix Scaling

    Zeyuan Allen-Zhu;Yuanzhi Li;Rafael Oliveira;Avi Wigderson

  • First Efficient Convergence for Streaming k-PCA: A Global, Gap-Free, and Near-Optimal Rate

    Zeyuan Allen-Zhu;Yuanzhi Li

  • Physics of Language Models: Part 3.1, Knowledge Storage and Extraction

    Unknown

  • Backward Feature Correction: How Deep Learning Performs Deep Learning

    Zeyuan Allen-Zhu;Yuanzhi Li

  • Even Faster SVD Decomposition Yet Without Agonizing Pain

    Zeyuan Allen Zhu;Yuanzhi Li

Frequent Co-Authors

Zeyuan Allen-Zhu
Zeyuan Allen-Zhu Meta Platforms, Inc.
Sébastien Bubeck
Sébastien Bubeck Microsoft (United States)
Yingyu Liang
Yingyu Liang University of Wisconsin–Madison
Tengyu Ma
Tengyu Ma Stanford University
Yin Tat Lee
Yin Tat Lee Microsoft (United States)
Sanjeev Arora
Sanjeev Arora Princeton University
Elad Hazan
Elad Hazan Princeton University
Aaron Sidford
Aaron Sidford Stanford University
Avi Wigderson
Avi Wigderson Institute for Advanced Study
Zhao Song
Zhao Song Adobe Systems (United States)

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

Report an issue

We appreciate your kind effort to assist us to improve this page, it would be helpful providing us with as much detail as possible in the text box below:

Related Online Degrees & Career Pathways

Expanding your education in Engineering and Technology can open doors to new leadership roles and diverse career paths. Many professionals are enhancing their technical expertise with advanced business and management credentials gained through flexible online programs.

If you’re looking to develop your skills in leadership, an online master's in organizational leadership can help you master the human and strategic aspects of leading technical teams. For those interested in the intersection of business and engineering, MBA programs in operations management online focus on streamlining processes and optimizing performance — invaluable for careers in production, supply chain, and project management.

The admissions process is now more accessible than ever, with many online mba no gre programs available, removing barriers for busy professionals. Affordability is also a priority for many students, and there are wide-ranging options for advancing your career, including online mba under $35k designed to offer quality education at a manageable cost.

Consider these flexible online degrees to complement your technical background and pursue leadership within engineering, technology, or operations management sectors in the USA.

Best Scientists Citing Yuanzhi Li

Trending Scientists

Recently Published Articles