**Title:** NVIDIA Researcher Yuke Zhu Discusses Advancements in Humanoid Robotics and Foundation Models
**Summary:** Yuke Zhu presents key insights from an ICRA'26 keynote, emphasizing the rapid progress in building human-like machines, the scalability of human data for training robot foundation models, and the role of world models in enabling policy generalization. The talk highlights NVIDIA's work on projects like GR00T, EgoScale, SONIC, and WAM, including a demo of a model learning to assemble the YCB airplane with zero teleoperation data. The discussion underscores ongoing advancements in embodied AI and robotics.
Evidence and limits
Published automatically after robotics and source-evidence checks; no manual editorial approval is recorded. Source assertions are not independently verified. Missing information remains not reported.
- Environment:
- Not reported
- Control:
- Not reported
- Data origin:
- Not reported
Source excerpts and review record
No manual editorial approval recorded.
Original source quotation: “Sharing a copy of my ICRA'26 keynote talk from three months ago. 3 key messages:”
Source E1
Original source quotation: “Building a human-like machine, a dream that has fascinated humankind for centuries, is rapidly becoming a reality.”
Source E2
Original source quotation: “Human data is, by far, the most scalable source of training data to fuel robot foundation models.”
Source E3
Original source quotation: “World models are a "data sponge" that can absorb massive amounts of multimodal data for policy generalization.”
Source E4
Original source quotation: “The talk brought together our work over the past year: GR00T, EgoScale, SONIC, WAM, and more.”
Source E5
Original source quotation: “Our (now old!) model learning to assemble the YCB airplane with zero teleop data.”
Source E6
Sharing a copy of my ICRA'26 keynote talk from three months ago. 3 key messages:
1. Building a human-like machine, a dream that has fascinated humankind for centuries, is rapidly becoming a reality.
2. Human data is, by far, the most scalable source of training data to fuel robot foundation models.
3. World models are a "data sponge" that can absorb massive amounts of multimodal data for policy generalization.
Video: https://www.youtube.com/watch?v=cNU_8t7i8Mk
Slides: https://yukezhu.me/talks/building_generalist_humanoid_robots.pdf
The talk brought together our work over the past year: GR00T, EgoScale, SONIC, WAM, and more. Here’s the same demo I showed on stage: Our (now old!) model learning to assemble the YCB airplane with zero teleop data.
We've made incredible progress since then. What's coming next? 🚀
Source：Yuke Zhu · x.com