Yanyong Zhang

ReTouch achieves a remarkable 18.4% improvement in success rates for dexterous manipulation by continuously refining tactile predictions in real-time.

ViCo3D achieves a groundbreaking 1.8x improvement in collaborative 3D object detection performance by merging LiDAR data with advanced visual representations from vision foundation models.

Role-aware training can boost video diffusion models' physical consistency by up to 39.4% without sacrificing visual fidelity.

Context-gated latent-action conditioning enables VLA models to achieve unprecedented success rates in robot manipulation tasks without relying on separate action-generation modules.

Humanoid robots can complete laboratory tasks but often fail to meet the precision required for scientific validity, exposing a critical gap in current automation efforts.

GEAR-VLA achieves a remarkable 90.1% success rate in universal grasping tasks, showcasing its ability to generalize across unseen objects and diverse robot embodiments.

Single-view RGB input can revolutionize how robots perceive and manipulate transparent objects, achieving reliable grasping without complex depth sensing.

Recovering static 3D scenes from monocular video with dynamic objects gets a boost: GA-GS leverages diffusion models to inpaint occluded regions, outperforming existing methods, especially in scenarios with large-scale occlusions.

DreamWorld achieves more world-consistent video generation by jointly modeling multiple heterogeneous dimensions of world knowledge, moving beyond surface-level plausibility.