Andrew Peng

I am an Honours BSc student in Computer Science with minors in Mathematics and Statistics at the University of Toronto.

I am currently a research intern with the Vector Institute and U of T, advised by Prof. Babak Taati.

profile photo

Research

I work on machine learning systems for vision and remote sensing, focusing on generative model mechanics, 3D vision, and distillation optimization. My current projects involve diffusion transformers, 3D Gaussian Splatting, synthetic hyperspectral data.

DiTGuiding Diffusion Transformers and Improving Distillation
Andrew Peng, with Prof. Babak Taati
Vector Institute & University of Toronto, 2026—present

Analyzing internal dynamics in Diffusion Transformers to develop interventions that guide generation.

λSynthesizing Hyperspectral Data using Generative Models to Train Spectral Unmixing Methods for Low-Cost Crop Residue Cover Mapping
Ege Artan, Shuo Chen, Andrew Peng, Sammuel Aldrich Karya, Kyaw Thiha
2026 ISPRS Congress

Decoupled generative DDIM/TCVAE pipeline to generate synthetic data for hyperspectral crop residue mapping.

32173D Gaussian Splatting for Human Reposing
Andrew Peng, Team Lead
Vector Institute & University of Toronto, 2026—present

Developing a 3D Gaussian Splatting pipeline for human reposing.

Experience

Research Intern, Vector Institute & University of Toronto · 2026—present
Deep Learning Team Lead, University of Toronto Aerospace Team · 2026—present
Lead Developer, LabPath Hackathon · 2025
Lead Researcher, Youreka Canada Toronto Chapter · 2025

Toronto, Canada · University of Toronto

Andrew
Peng.

Computer science student and researcher building generative models for how we see, understand, and shape the world.

01 / ABOUT

I’m an Honours BSc candidate in Computer Science with a minor in Mathematics at the University of Toronto. My interests sit at the intersection of machine learning, computer vision, and scientific research—especially generative models and 3D representations.

02 / SELECTED RESEARCH

Making generative models
more useful in the real world.

Current work in remote sensing, scientific machine learning, and generative models.

02
3217

NASA GLEE Team 3217 · 2022—2025

Engineering for the lunar surface

Led a 10-person team building and testing an Arduino lunar satellite, including C software for sensor data transmission and systems work around power, automation, and deployment constraints.

CEmbedded systemsLeadership

03 / EXPERIENCE

A few places I’ve learned.

2026—present

Research Intern · Vector Institute & U of T

Recipient of a $12,000 U of T Department of Computer Science Research Grant.

2026—present

Deep Learning Team Lead · U of T Aerospace Team

Leading a team exploring generative approaches to remote sensing.

2025

Lead Developer · LabPath Hackathon

Developed a thyroid-cancer risk index with a Python Random Forest model; work published in URNCST.

2025

Lead Researcher · Youreka Canada

Led a four-person team studying associations between air pollution and Lyme disease.

04 / HIGHLIGHTS

100%Graduating average
TDSB Top Scholar
3rdAMC 12A, 2024
127.5 / 150
1stYoureka Toronto
Research Symposium
ARCTRoyal Conservatory
of Music Diploma