Carnegie Mellon University
Stanford University
OpenAI
Tsinghua University
Princeton University
Anirudh GoyalSenior Staff Research Scientist · MetaResearcher↗
Researcher · Academic
Yuanzhi Li
MetaResearch Scientist
Li connects deep-learning theory with ways to build and adapt smaller language models. He co-developed LoRA, which reduces the cost of adapting a model[1], and TinyStories, showing that tiny models can write coherent stories.[2] He also helped create Microsoft’s Phi-4, emphasizing high-quality training data.[3]
Career & education
Current role
Research Scientist
Previously
Work and education history is primarily focused on AI-relevant roles and may not be comprehensive.
Citation Trend
Total citations to scholarly work, including research papers, technical reports, books and theses, as recorded by Google Scholar. Each point is a dated snapshot.
citations74,493As of Oct 22, '26
citations+6,052Since last snapshot, Jun 15, '26
Selected contributions
LoRA: Low-Rank Adaptation of Large Language Models
LoRA adapts large models by training small additions rather than all their parameters.
References
- LoRA: Low-Rank Adaptation of Large Language Modelsarxiv.org
Original coauthored method reduces trainable parameters and memory requirements when adapting models.
- TinyStories: How Small Can Language Models Be and Still Speak Coherent English?arxiv.org
Original two-author paper with Ronen Eldan demonstrates small models producing coherent stories using a synthetic story dataset.
- Phi-4 Technical Reportarxiv.org
Original coauthored product/model report emphasizes high-quality and synthetic training data.
Based on public sources. If accuracy is important, verify details using the links or references.
For corrections, updates, or requests to remove information, contact us.
Reviewed August 8, 2026 · Checked for accuracy and new information.
