Shiyi Zhang

Papers

Total Citations

H-Index

About

Shiyi Zhang is a rising force in embodied AI and robotic manipulation, whose work bridges the gap between 3D scene representation and dexterous robot control. His research centers on dynamic Gaussian splatting, a cutting-edge technique that models real-time, interactive environments for multi-task robotic systems. In his landmark 2024 paper, "ManiGaussian: Dynamic Gaussian Splatting for Multi-task Robotic Manipulation," Zhang introduced a novel framework that enables robots to perceive and adapt to changing scenes with unprecedented speed and accuracy. This work, already garnering 33 citations within its first year, demonstrates his ability to push the boundaries of how machines understand and interact with physical spaces. By integrating neural rendering with manipulation policies, Zhang has opened new pathways for robots to perform complex tasks—from grasping to assembly—in unstructured settings. His contributions are particularly notable for their practical implications in industrial automation and assistive robotics, where real-time adaptability is critical. As a young researcher, Zhang’s early impact signals a promising trajectory, positioning him as a key innovator in the next generation of intelligent robotic systems.