Instant-Fold: In-Context Imitation Learning for Deformable Object Manipulation
One-shot garment folding from a human demonstration trained entirely in simulation.
Yilong Wang
I am a second-year PhD student in the Robot Learning Lab at Imperial College London, advised by Dr. Edward Johns. I began my PhD in February 2025 and am currently a Student Researcher with the Robotics Team at Google DeepMind in London.
Prior to my PhD, I completed an MRes in Artificial Intelligence and Machine Learning in the same lab, and an MEng in Electronic and Information Engineering in the same university, working with Prof. Yiannis Demiris in the Personal Robotics Lab.
My research interests lie in the intersection of computer vision, machine learning and robotics, with a focus on developing scalable imitation learning methods with minimal human efforts.
Publications
SynthICL: Scalable In-context Imitation Learning with Synthetic Data
RGB-only in-context imitation learning trained entirely on scalable synthetic data.
Observer-Actor: Active Vision Imitation Learning with Sparse-View Gaussian Splatting
Active viewpoint optimization within Gaussian Splats for imitation learning.
One-Shot Dual-Arm Imitation Learning
One-shot imitation via visual servoing and trajectory replay for precise tasks.