Yihe Dong

Publications

I am a researcher at Princeton University, working on large language models. Previously, I was a research engineer at Google, working on machine learning research with focus on representation learning on structured data, automated feature discovery, and multimodal representation learning. I studied math at Princeton University (undergrad) and the University of Wisconsin - Madison (grad).

Before Google, I worked at Microsoft Research AI on efficient optimal transport, graph representation learning, and vision-language representation models to improve core relevance algorithms for Bing.

Research Interests

My research focuses on two main areas:

• Geometric representation learning: I have developed algorithms for fast optimal transport, graph representation learning, efficient outlier detection in high dimensions, and methods for neural locality sensitive hashing.
• Language and multimodal representation models: I have designed and developed various large language models and multimodal representation models.

In general, I enjoy applying mathematical principles to improve model building and am broadly interested in current research developments in machine learning.