Chuyuan Fu
Papers
13
Total Citations
589
H-Index
7
About
Chuyuan Fu is a leading researcher at the intersection of robotics, computer vision, and large-scale machine learning, whose work is pioneering the development of generalist robot policies. Fu’s core research focuses on leveraging massive, internet-scale data to imbue robots with unprecedented generalization and semantic reasoning capabilities. His most impactful contribution is the RT-2 model (267 citations), a seminal work that demonstrates how vision-language-action models can transfer web knowledge directly into end-to-end robotic control, enabling robots to perform emergent tasks they were never explicitly trained on. Fu is also a key architect of the Open X-Embodiment project (over 220 combined citations), a landmark collaboration that created a massive, cross-embodiment dataset and the RT-X model family, establishing a foundational backbone for the field. His work on "Language to Rewards for Robotic Skill Synthesis" further showcases his ability to harness large language models for complex skill acquisition. Beyond these breakthroughs, Fu has demonstrated the real-world viability of deep RL at scale, notably in a system for sorting waste with a fleet of mobile manipulators. His recent contributions to the Gemini Robotics family signal his continued leadership in bringing advanced AI into the physical world.
Research Focus
Key Achievements
Top Papers
- 1RT-2: Vision-Language-Action Models Transfer Web Knowledge to Robotic Control267 citations · 2023
- 2
- 3Open X-Embodiment: Robotic Learning Datasets and RT-X Models101 citations · 2023
- 4Language to Rewards for Robotic Skill Synthesis38 citations · 2023
- 5Jump-Start Reinforcement Learning17 citations · 2022
- 6
- 7
- 8
- 9Evaluating Real-World Robot Manipulation Policies in Simulation4 citations · 2024
- 10Gemini Robotics: Bringing AI into the Physical World4 citations · 2025