Yingwei Noah Song

Hi there, I am a first year PhD student in student in Electrical and Computer Engineering at the University of Arizona, advised by Prof. Huanrui Yang. My research is currently in an exploratory phase, but I am broadly interested in 3D reconstruction, diffusion-based generative models, video synthesis, and efficient learning methods. I am especially curious about how to build semantically meaningful and computationally efficient generative systems that can handle complex spatial and temporal data.

Before coming to UArizona, I completed my M.S. in Computer Science at Brown University in 2025, where I worked with Prof. James Tompkin on neural rendering and representation learning using NeRF. Before that, I received my B.S. from the University of Wisconsin–Madison in 2023, double majoring in Computer Science and Data Science, with a minor in Math.

In the summer of 2024, I was a research intern at the Harvard Visual Computing Lab, where I collaborated with Prof. Hanspeter Pfister and Dr. Wanhua Li on 3D Gaussian Splatting with language-based supervision.

In 2025, I interned at Hedra AI, working with Hongwei Yi on diffusion-based video generation and efficient generative modeling. I gained HPC experience with k8s, Airflow, and developed a custom Slurm-based orchestration system to streamline large-scale preprocessing. I drove the creation of a pipeline that filtered raw videos into training-ready datasets and embeddings, producing over 10 million clips—a milestone I am especially proud of.

Beyond research, I am especially fond of Siamese cats and rabbits.

A quote that resonates with me is from William T. Freeman: “Slow down to speed up.”