Raymond Yu

About Me

Hey there! I am a BS/MS student studying computer science at the Allen School in the University of Washington, Seattle.

I’m fortunate to be advised by Abhishek Gupta in the WEIRD Lab, where I work on problems in robot learning. My current research focuses on building robotic systems that not only complete tasks, but also reason about their environment and understand its underlying dynamics.

Beyond research, I serve as a teaching assistant for computer vision, helping students develop a strong foundation in perception and its applications in robotics and AI.

Previously, I spent time at Amazon Web Services, and as a student researcher at the Allen Institute for AI on the PRIOR team.

News

• [August. 2025]   Honored to receive the 2025 Microsoft Endowed Scholarship.
• [June. 2025]   Unified World Model won the Best Paper Award at ICML 2025 World Modeling Workshop!
• [April. 2025]   Unified World Model has been accepted to RSS 2025.
• [January. 2025]   DRAWER has been accepted to CVPR 2025!
• [January. 2025]   HAMSTER has been accepted to ICLR 2025.

Publications

Unified World Models

Chuning Zhu, Raymond Yu, Siyuan Feng, Benjamin Burchfiel, Paarth Shah, Abhishek Gupta

RSS 2025

ICML 2025 Workshop on Building Physically Plausible World Models - Best Paper Award

Code Website

---

DRAWER: Digital Reconstruction and Articulation With Environment Realism

Hongchi Xia, Entong Su, Marius Memmel, Arhan Jain, Raymond Yu, Numfor Mbiziwo-Tiapo, Ali Farhadi, Abhishek Gupta, Shenlong Wang, Wei-Chiu Ma

CVPR 2025

Code Website

---

HAMSTER: Hierarchical Action Models For Open-World Robot Manipulation

Yi Li, Yuquan Deng, Jesse Zhang, Joel Jang, Marius Memmel, Raymond Yu, Caelan Garrett, Fabio Ramos, Dieter Fox, Anqi Li, Abhishek Gupta, Ankit Goyal

ICLR 2025

Code Website

---

OPTIMUS: Observing Persistent Transformations in Multi-temporal Unlabeled Satellite-data

Raymond Yu, Paul Han, Josh Myers-Dean, Piper Wolters, Favyen Bastani

WACV 2025

Paper Code

---

Teaching

I teach a little bit when I have time! Here are some of the notes/slides from the lectures:

CSE 455 - Computer Vision

Spring 2025

Structure from Motion

3D reconstruction techniques, bundle adjustment, and building 3D models from 2D image sequences.

Camera Geometry and Calibration

Camera models, intrinsic and extrinsic parameters, and practical camera calibration techniques.

Homography and Geometric Transformations

Understanding projective geometry, homography estimation, and applications in image rectification and panorama stitching.

Feature Descriptors and Matching

Comprehensive overview of SIFT descriptors and robust feature matching techniques.

Feature Detection: Lines and Corners

Feature detection covering Harris corner detection.

Line Detection Algorithms

Deep dive into Hough transforms and RANSAC.

Derivatives and Edge Detection

Fundamental edge detection techniques using mathematical foundations.

Projects

Beyond research, I love tinkering with side projects that combine my interests in robotics, AI, and everyday problems. Here are some of my favorites:

Real2Sim2Star

If robots can cross the sim2real gap, why not idols too?

💡 Turn on your music for the best experience!

I've always wanted to have insane dancing skills. Obviously I didn't end up as Kpop star, so I decided to make myself one (virtually). When Real2Sim2Real was all the rage, I had scanned in countless objects, but I had never tried scanning in humans. In this project, I scanned in a few of my friends, rigged the characters, and made them dance!

Raymond Yu

ryu5 [AT] cs.washington.edu