Yuanzhi Li

Assistant Professor, CMU, Carnegie Mellon University

Joined

February 2019

Names

Yuanzhi Li (Preferred)

Emails

****@stanford.edu (Confirmed)

****@andrew.cmu.edu (Confirmed)

****@gmail.com (Confirmed)

****@mbzuai.ac.ae (Confirmed)

Personal Links

DBLP

Career & Education History

Assistant Professor

CMU, Carnegie Mellon University (andrew.cmu.edu)

2019 – Present

Advisors, Relations & Conflicts

No relations added

Expertise

No areas of expertise listed

Publications

SCALAR: Self-Supervised Composition and Learning of Skills with LLM Planning and RL
Renos Zabounidis, Yue Wu, Simon Stepputtis, Tom Mitchell, Yuanzhi Li, Katia P. Sycara
- LAW
- Readers: Everyone
Mobile-MMLU: A Mobile Intelligence Language Understanding Benchmark
Sondos Mahmoud Bsharat, Mukul Ranjan, Aidar Myrzakhan, Jiacheng Liu, Bowei Guo, Shengkun Tang, Zhuang Liu, Yuanzhi Li, Zhiqiang Shen
- Accepted by DMLR
- Readers: Everyone
Physics of Language Models: Part 1, Learning Hierarchical Language Structures
Zeyuan Allen-Zhu, Yuanzhi Li
- Accepted by TMLR
- Readers: Everyone
Phi-3 technical report: A highly capable language model locally on your phone
Marah I Abdin, Jyoti Aneja, Hany Hassan Awadalla, Ahmed Hassan Awadallah, Ammar Ahmad Awan, Weizhu Chen, Zeqi Lin, Yuanzhi Li, David Majercak, Donghan Yu, Lu Yuan, Dan Iter, Wen Wen
- OpenReview Archive Direct Upload
- Readers: Everyone
Understanding the Evolution of the Neural Tangent Kernel at the Edge of Stability
Kaiqi Jiang, Jeremy Cohen, Yuanzhi Li
- NeurIPS 2025 poster
- Readers: Everyone
Beyond Parameter Count: Implicit Bias in Soft Mixture of Experts
Youngseog Chung, Dhruv Malik, Jeff Schneider, Yuanzhi Li, Aarti Singh
- Accepted by TMLR
- Readers: Everyone
On the Clean Generalization and Robust Overfitting in Adversarial Training from Two Theoretical Views: Representation Complexity and Training Dynamics
Binghui Li, Yuanzhi Li
- ICML 2025 poster
- Readers: Everyone
Adversarial Training Can Provably Improve Robustness: Theoretical Analysis of Feature Learning Process Under Structured Data
Binghui Li, Yuanzhi Li
- M3L Poster
- Readers: Everyone
Mixture of Parrots: Mixtures of experts improve memorization more than reasoning
Samy Jelassi, Clara Mohri, David Brandfonbrener, Alex Gu, Nikhil Vyas, Nikhil Anand, David Alvarez-Melis, Yuanzhi Li, Sham M. Kakade, Eran Malach
- M3L Oral
- Readers: Everyone
Physics of Language Models: Part 3.2, Knowledge Manipulation
Zeyuan Allen-Zhu, Yuanzhi Li
- ICLR 2025 Poster
- Readers: Everyone

View all 143 publications

Co-Authors

View all 180 co-authors