Brian Yang

Applied Scientist · Wayve Labs

I am an Applied Scientist at Wayve Labs on the Language and Reasoning team. Previously, I completed my PhD in Robotics at Carnegie Mellon University, advised by Katerina Fragkiadaki and Jeff Schneider, and worked at Meta AI. Before that I did my undergrad at UC Berkeley, where I was fortunate to work with Dinesh Jayaraman, Roberto Calandra, Sergey Levine, and Kris Pister.

My research is about the data bottleneck in robot learning: the rare and safety-critical behaviors that are scarce in real-world logs. During my PhD I worked on this for autonomous driving, using language to specify the behavior we want and generative models to synthesize it, and treating the two as a single data engine for training driving policies. Earlier, during my undergrad, I worked on a variety of real-world robot learning problems ranging from low-cost manipulation to microrobot locomotion.

Email/CV/PhD thesis/GitHub/LinkedIn

Publications · 11 papers, 1 thesis

★ 2027

EurekaDrive: Vision-Language Guided Program Search for Language-Conditioned Diffusion Driving Policies

Mineui Hong, Harsh Shah, Brian Yang, Eric Shau, Alexandre Capone, Katerina Fragkiadaki, Jeff Schneider ICRA 2027 submission

EurekaDrive: reward code and a diffusion policy iteratively refined across three rounds of feedback from a critic vision-language model

2026

GANDALF: Grounding Autonomous Driving Agents via Language Feedback

Brian Yang, Jeff Schneider, Katerina Fragkiadaki ECCV 2026 submission

GANDALF: reward functions guide a scene-conditioned flow prior to produce counterfactual driving demonstrations

2026

A Language-Guided Generative Data Engine for Autonomous Driving

Brian Yang PhD thesis, The Robotics Institute, Carnegie Mellon University · CMU-RI-TR-26-16

PDF

Cover of the thesis: A Language-Guided Generative Data Engine for Autonomous Driving

★ 2024

Diffusion-ES: Gradient-free Planning with Diffusion for Autonomous Driving and Zero-Shot Instruction Following

Brian Yang, Huangyuan Su, Nikolaos Gkanatsios, Tsung-Wei Ke, Ayush Jain, Jeff Schneider, Katerina Fragkiadaki CVPR 2024

arXiv PDF Website

2024

Learning Driving Policies with Offline Counterfactual Reactive Simulation

Adam Villaflor, Brian Yang, Katerina Fragkiadaki, John Dolan, Jeff Schneider ICML 2024 submission

2024

Tractable Joint Prediction and Planning Over Discrete Behavior Modes for Urban Driving

Adam Villaflor, Brian Yang, Huangyuan Su, Katerina Fragkiadaki, John Dolan, Jeff Schneider ICRA 2024

arXiv PDF Code

2022

Fully Autonomous Real-World Reinforcement Learning with Applications to Mobile Manipulation

Charles Sun, Jędrzej Orbik, Coline Devin, Brian Yang, Abhishek Gupta, Glen Berseth, Sergey Levine CoRL 2022

arXiv PDF Blog Code

2020

MAVRIC: Morphology-Agnostic Visual Robotic Control

Brian Yang*, Dinesh Jayaraman*, Glen Berseth, Alexei Efros, Sergey Levine RA-L + ICRA 2020

arXiv PDF Project page

2020

DIGIT: A Novel Design for a Low-Cost Compact High-Resolution Tactile Sensor with Application to In-Hand Manipulation

Mike Lambeta*, Po-Wei Chou*, Stephen Tian, Brian Yang, Benjamin Maloon, Victoria Rose Most, Dave Stroud, Raymond Santos, Ahmad Byagowi, Gregg Kammerer, Dinesh Jayaraman, Roberto Calandra RA-L + ICRA 2020

arXiv PDF Project page