Jun Cen

Jun Cen is a researcher in embodied AI and computer vision, currently affiliated with JD Explore Academy. He is best known for his work on open-world semantic segmentation, 3D object detection, and vision-language-action models, with publications appearing at major conferences including ICCV, CVPR, NeurIPS, and ECCV. His recent research includes the WorldVLA project, which explores autoregressive action world models for robotic manipulation.