Jiaming Liang

Real-robot trials are costly and slow, but DataLadder enables scalable evaluations and data generation through a seamless interplay between robots, simulations, and human feedback.

Bridging the gap between human manipulation and robotic control, JoyAI-RA unlocks enhanced cross-embodiment behavior learning through multi-source pretraining.

A novel proximal sampler achieves unbiased constrained sampling with minimal oracle access, outperforming existing methods that rely on projections or barrier functions.