Zhenyang Chen retweeted
The future of robotics is agentic.
I distilled GPT 6 Astra into a tiny 4B parameter model and it learned to drive a robot arm!
Before: It couldn't finish a single task, and at one point decided the bowl right in front of it didn't exist.
After: 0 -> 22.5% on scenes it was never trained on 🧵
Zhenyang Chen retweeted
Over the past week, I’ve been overwhelmed by the love and support of friends, industry colleagues, and people I've never met. Thank you to everyone who has shown kindness to my family and me. It means the world to me.
I came to the United States 53 years ago, and becoming an American citizen is one of the greatest gifts and proudest moments of my life. NVIDIA has been my life’s work. Building it into the company it is today alongside my colleagues has been a privilege and an honor. Nowhere but in America could a company like NVIDIA be possible. I love our country, and I love NVIDIA.
I did not live up to the standards expected of everyone at NVIDIA, and I have made serious mistakes in my personal life. I created problems for the Board that I deeply regret. I own my mistakes completely, and I am committed to cooperating with the Board’s independent investigation.
Since my departure, I’ve seen an outpouring of calls from NVIDIA’s employees, investors, business partners, and customers for me to return as CEO. Many continue to reach out to me directly as the company navigates complex technology, market, competitive, and geopolitical dynamics. NVIDIA is a once-in-a-generation company. If the Board of Directors decides that my return is in the best interest of our shareholders, our employees, the company, and America, I would be honored to answer these calls and return as CEO.
Thank you again to everyone who has shared their good wishes. I am profoundly humbled by your support. To my family: I love you, and I will spend the rest of my life being worthy of yours.
Jensen
Zhenyang Chen retweeted
Two years ago, we released Genesis, a universal physics engine for robotics.
Today, I’m excited to announce 🍎 Eden — a robot learning framework built from the ground up for agents, unleashing full potential of Genesis.
One task definition. From simulation to real robots. Sim-to-sim and sim-to-real just got easier.
Coding agents build tasks 1.6–4× faster than in Isaac Lab.
🧵 1/11
That’s a wrap! CoRL 2026 was such a blast! Even the FAA wanted to extend it another day.
Loved the energy across the technical program, exhibitions, and social events. Signing off as General Chair. Grateful to everyone who made it happen.
See you in Paris @corl_conf next year!
🤖 Made with AI
Zhenyang Chen retweeted
I had a great time seeing everyone at CoRL this week.
My talk was called "Learning in the factories", and I talked about:
- Walden's longest-running deployment to date,
- the value of coupling deployment data with large-scale pretraining, and
- a few things I've learned as I've broadened my focus from AI/robotics to include large-scale manufacturing.
I'm excited to keep sharing more of Walden's story as we go. Things are moving very fast!
-
youtube.com/live/uPXoJm-L3YY…
Zhenyang Chen retweeted
Okay, here's one thing that's been bothering me after attending #CoRL2026. 🧵
To me, whether to work in a given area is such a personal choice, made by individual scientists depending on their research taste and idiosyncrasies.
The best part about science is that we don't have to agree!! I can think that Area or Problem X is not worth working on, and anyone can (and probably will) disagree with me and end up proving me wrong with evidence.
To me, science involves trying to achieve consensus on (ideally falsifiable) hypotheses *after* conducting research. Not before. Debates are very useful and necessary for this.
But perhaps we should think more carefully about claims like "Area X is not worth working on". I heard many of these being freely dispensed at this past week.
I feel that these subjective claims are nearly impossible to verify. More importantly, I worry that such sweeping claims may cause many junior researchers working in that very area to feel considerable distress.
Note that this is very different from lobbying for everyone to work on our pet problems by highlighting their motivations and potential impact. Such lobbying has long been a part of science, and will likely always be.
Maybe we can keep science weird and highly entropic! The very diversity of our idiosyncratic research agendas is what helps increase the odds that at least some of us will eventually identify a path to our ultimate destination — knowledge!
Meet GR00T 2, our most advanced, open System One model for robotics research. It learns skills from human demonstrations and predicts how its actions will change the physical world.
Those skills can carry across different robot bodies, reducing the need to teach each robot every task from scratch.
Watch GR00T 2 put that research into motion. 🧵
Zhenyang Chen retweeted
We proudly gave @OpenAI our NVIDIA DGX-1 supercomputer to work on #AI's toughest problems. nvda.ws/2aRtOhE
Meet WARP 10:30a at P3–039 Lone Star today and learn about how a good retargeter shape policy behavior #CoRL2026
Human data is the cheapest source of whole-body mobile manipulation demos. Turning it into robot data is the hard part.🤖
WARP retargets OFFLINE human motion into replayable robot actions in closed form — precise, consistent, faithful to whole-body intent.
🌍 warp-retargeting.github.io/
Meet GR00T 2, our new generation of world-action models (WAMs): it imagines how the physical world will unfold and generates the actions that get it there. One model runs on parallel grippers, dexterous hands, and humanoids.
research.nvidia.com/labs/gea…
Introducing GR00T 2: our most advanced, open System One model for robotics. It runs entirely on your robot’s onboard compute, fully untethered, and you own every weight of it. GR00T 2 features:
1. Long-horizon tasks: dexterous manipulation and whole-body control chained over 20+ steps and 5+ minutes, such as performing chemistry lab experiments or assembling model airplanes end to end.
2. X-Adapt: adapting from easy data sources (human wearables and simulation) and deploying zero-shot to diverse real robots.
3. Unified Cognition: native integrations with frontier agents for advanced physical reasoning, robust error recovery, and safety monitoring.
It's an open-source buffet! We are releasing not only the model weights and X-Adapt recipes (coming December), but also our agentic framework, simulation engine, annotation engine, hardware design, and even training data (100K hours, that's ~11 yrs of experience), all available today!
Everything that moves will be autonomous. Anyone can become a roboticist.
Lock in, build with us!
Deep dive in thread:
Heading to Austin for #CoRL2026 ✈️ Excited to meet up with old and new friends! 🤠
Come check out our EgoWAM poster and talk with us:
📍 Poster Session 2, #62 · JW Marriott, Lone Star (3rd floor)
🕔 Tue, Nov 10 · 5:00–6:30 PM
Also excited to chat about world (action) models, cross-embodiment learning, agentic robotics, software-hardware co-design, and anything else about robot learning during the conference. DMs open, come say hi! 👋
Still remember the excitement when seeing sign of life of X–Adaptabillity from GROOT2.
More to come soon!
Meet GR00T 2, our new generation of world-action models (WAMs): it imagines how the physical world will unfold and generates the actions that get it there. One model runs on parallel grippers, dexterous hands, and humanoids.
research.nvidia.com/labs/gea…
15
479
Zhenyang Chen retweeted
Can one world model handle rigid boxes, plush toys, and cloth?
Introducing MoRE: a Mixture-of-Experts 3D world model that predicts how objects move across materials and plans real-robot actions with MPC.
See us at #CoRL2026: Monday, Spotlight Session 2 & Poster Session 1 🤠!
The era of System One Vision Model is here! Introducing VFlash: parsing the visual world in parallel at unmatched speeds for Physical AI.
Most embodied prediction tasks like object localization, robot trajectory planning, and dense 3D motion are fixed-length float vectors. For example, a bounding box is always 6 tokens, a 10-waypoint robot arm trajectory is 54: we know these beforehand. Instead of decoding one token at a time, we can output the entire geometric shape in one go.
VFlash enables up to 12x faster action decoding and a 5.1x overall robot inference speedup on edge GPUs like the Jetson Thor and DGX Spark. What’s surprising is that it ends up even more accurate than the VLM backbone, with up to 61% lower 3D trajectory error. The intuition is that VFlash avoids the cascading errors of autoregressive decoding, where every coordinate is written in ink and every later point loses the chance to backspace.
Everything that moves will be autonomous, and everything autonomous will need to think on its own chip. The next inference revolution is physical reasoning at the speed of reflex.
And as always, it’s open paper, open source! Link in thread:

























