Minzhao Zhu

Papers

Total Citations

H-Index

About

Minzhao Zhu is a leading researcher in embodied AI and robot manipulation, with a focus on enabling agents to operate intelligently in unseen environments. His key contributions lie in object goal navigation (ObjectNav) and the development of generalist robot agents. In his highly cited 2022 work, "Navigating to Objects in Unseen Environments by Distance Prediction" (26 citations), Zhu introduced a novel approach that uses semantically related objects as cues to predict distances to target objects, allowing robots to navigate without pre-built maps—a breakthrough for real-world deployment. More recently, his 2024 paper "GR-2: A Generative Video-Language-Action Model with Web-Scale Knowledge for Robot Manipulation" (7 citations) presents a state-of-the-art generalist robot agent. By pre-training on 38 million Internet video clips and over 50 billion tokens, GR-2 captures world dynamics, enabling versatile and generalizable manipulation. This work bridges web-scale knowledge with physical action, marking a significant step toward robots that can adapt to diverse tasks. Zhu’s research is shaping the future of autonomous systems, making him a notable figure in robotics and AI.