Shengjun Zhang

VisualGRPO. E-GRPO: High Entropy Steps Drive Effective Reinforcement Learning for Flow Models

GGN. [NeurIPS 2024] Gaussian Graph Network: Learning Efficient and Generalizable Gaussian Representations from Multi-view Images

Scene-Splatter. [CVPR 2025] Scene Splatter: Momentum 3D Scene Generation from Single Image with Video Diffusion Model

GeoAuxNet. [CVPR 2024] GeoAuxNet: Torwards Universal 3D Representation Learning for Multi-sensor Point Clouds