Jeffrey Zhao

Google DeepMind (United Kingdom)

Papers

2

Total Citations

36

H-Index

2

About

Jeffrey Zhao is a leading researcher at the intersection of robotics, computer vision, and multimodal machine learning, with a core focus on enabling long-horizon reasoning for embodied agents. His most impactful work centers on the development of **RoboVQA**, a pioneering framework that tackles the critical challenge of teaching robots to understand and execute complex, multi-step tasks in the real world. Zhao’s major contribution lies in rethinking data collection for robotics: he proposed a scalable, bottom-up scheme that captures intrinsically diverse, realistic interactions, achieving a **2.2x higher throughput** than traditional, top-down methods. This approach allows robots to learn high-level reasoning over both medium and long horizons, moving beyond simple pick-and-place to sophisticated, context-aware manipulation. With his 2024 paper already garnering **34 citations** in a short time, Zhao’s work is rapidly shaping the future of generalist robotics. His research is particularly notable for bridging the gap between large-scale vision-language models and physical action, providing a practical pathway for robots to reason, plan, and act in unstructured environments—a foundational step toward truly autonomous, helpful machines.

Research Focus

Key Achievements

2
H-Index
2
Papers
36
Total Citations
18
Avg Citations/Paper
🏆 Most Cited Paper
RoboVQA: Multimodal Long-Horizon Reasoning for Robotics
34 citations · 2024
📈 Most Prolific Year: 2024 (1 Papers)
🤝 Key Collaborators: 22
🏛 Institutions: Google DeepMind (United Kingdom)

Top Papers

  1. 1
  2. 2

Key Collaborators

Contact & Links

Available for collaboration
Content generated · 12 days ago