Xiaofeng Wang

Scalable Training for Vector-Quantized Networks with 100% Codebook Utilization

ReconDreamer-RL: Enhancing Reinforcement Learning via Diffusion-based Scene Reconstruction

Motion-R1: Enhancing Motion Generation with Decomposed Chain-of-Thought and RL Binding

GigaVideo-1: Advancing Video Generation via Automatic Feedback with 4 GPU-Hours Fine-Tuning

RoboTransfer: Geometry-Consistent Video Diffusion for Robotic Visual Policy Transfer

EgoDemoGen: Novel Egocentric Demonstration Generation Enables Viewpoint-Robust Manipulation

WonderFree: Enhancing 3D World Generation via Video Diffusion Prior with Multi-view Consistency

MimicDreamer: Aligning Human and Robot Demonstrations for Scalable VLA Training

EMMA: Generalizing Real-World Robot Manipulation via Generative Visual Transfer