Hanlin Wu

Robots can master delicate tool manipulation directly from unpaired human videos, achieving a 73% performance leap over SOTA by anchoring control in object-centric 3D pose priors rather than raw end-to-end pixels.

A new framework, Open-MOPD, boosts capability integration in multi-teacher distillation from 35.6% to 83.4% by addressing critical optimization imbalances.

Weak models can teach strong ones how to act better, boosting performance without the heavy lifting of direct RL training.

Ditch slow, iterative diffusion for remote sensing super-resolution: FlowGS uses flow matching and Gaussian splatting to generate high-resolution images with competitive quality and much faster inference.