Senior Research Scientist @ NVIDIA | PhD in Robotics @ CMU
Seattle, WA
Joined March 2009
- Tweets32
- Following104
- Followers506
- Likes136
Pinned Tweet
When every generalist robot model scores 95%+ on a benchmark, the numbers become meaningless.
What if we built a photorealistic benchmark that never saturates and can generate new scenes and tasks with AI Workflows in minutes?
We introduce RoboLab! 🧵(1/6)
Jenai Xuning Yang retweeted
🚀 Flexible test-time scaling with a 4B model and just 8K-token context windows!
One of my favorite things about 🪽 Hermes is how many simple but powerful ideas come together to tackle a much broader challenge:
How do we teach a small model an understudied agentic capability and inadequately measured by existing benchmarks?
In our case, that capability is contextual reasoning: knowing when and how to delegate, what to preserve in compact reasoning summaries, and how to continue reasoning from them.
🚀 We gave a 4B model with an 8K context window more test-time compute. Instead of improving, it got worse.
Scaling across context windows requires more than inline reasoning. It takes contextual reasoning: knowing how to delegate, preserve reasoning in compact summaries, and continue reasoning from them.
(1/10)
Jenai Xuning Yang retweeted
🎓 On-policy distillation has a teacher problem. SCOUT fixes it.
Then irony? OPD lets students learn from reasoning paths they actually visit. But their teachers must provide feedback on paths they rarely take themselves.
As these paths grow longer, even strong teachers struggle to provide reliable supervision.
Our idea is simple: Why not let the teacher and student learn together?
Jenai Xuning Yang retweeted
Introducing MILO, a step toward recursively self-improving agents for long-horizon, reasoning-intensive problems! 🧬
TL;DR: MILO evolves harnesses, but also the search process that discovers better harnesses.
It beats Claude Code and Codex on Terminal-Bench, outperforms methods like GEPA and Meta-Harness, and that's not all! It also discovers tighter upper bounds on three open math problems.
🧵👇
Can and should a general-purpose model serve as a robot policy?
In Astranomy, we study GPT-6 Astra for general pick-and-place manipulation. Astra sets a new SOTA on RoboLab-120, but real-time robot control remains an open challenge.
Full tech report:
research.nvidia.com/labs/srl…
Native VLA/WAMs have very weak language-grounding and reasoning capabilities. With ARC, we uplift existing VLA/WAMs to gain ~30pp performance with no new robot action data. The capability is already in these models; we just haven’t fully unlocked it yet.
arc-robot-reasoning.github.i…
Today we are introducing ARC, a reasoning recipe for robot foundation models that sets a new state of the art on RoboLab-120 and MolmoSpaces.
The prevailing way to improve robot foundation models is brute force: larger models, more robot demonstrations, costly training at scale. ARC needs none of that. With no new robot data and no foundation-scale training, π0.5 and Cosmos3-Nano-Policy gain up to:
📈 +50.0 pp on RoboLab-Reasoning-50
📈 +29.8 pp on RoboLab-120
📈 +27.2 pp on MolmoSpaces
To our knowledge, these gains are unprecedented.
📄 Paper: arxiv.org/abs/2610.12386
🌐 Website: arc-robot-reasoning.github.i…
🧵 (1/7)
Jenai Xuning Yang retweeted
Excited to share that VoLo: A Physical Orchestrator for Open-Vocabulary Long-Horizon Manipulation has been accepted to #CoRL2026! 🎉🤖
VoLo explores agentic robotics for long-horizon, open-vocabulary tasks, where a VLM acts as the robot’s brain with memory to reason, plan, monitor, detect failures, recover, and orchestrate tools—including VLA/WAMs, vision models, and action primitives.
We also introduce RoboVoLo, a high-fidelity simulation benchmark with 126 tasks spanning common sense, memory/state tracking, complex references, and world knowledge, together with detailed diagnosis of current system limitations.
This work was done during my internship at @nvidia @NVIDIAAI @NVIDIARobotics. Huge thanks to all collaborators @HugoHadfield1, Alexander Zook, @mikacuy, @luke_ch_song, @erwincoumans, @xuningy, Faisal Ladhak, @BirchfieldStan, Jonathan Tremblay, and @robovalts!
🔗 Project: chicychen.github.io/VoLo/
💻 Code: github.com/NVlabs/VoLoAgent
📊 RoboVoLo Benchmark: github.com/NVlabs/RoboVoLo
Looking forward to discussions at CoRL 2026 — hope to meet many of you in person!
#CoRL2026 #AgenticRobotics #Robotics #EmbodiedAI #RobotLearning #PhysicalAI #NVIDIA
Jenai Xuning Yang retweeted
CPT is an important bridge between academic and industrial research.
I have benefited from it on both sides: as a graduate student, learning how frontier research is done in industry, and now as an industry researcher, working with graduate students to advance fundamental research.
Weakening that bridge would hurt both academia and industry.
ICE just published a new memo (BCM 2608-01) targeting international students. It effectively blocks universities from authorizing CPT internships that are not strictly mandatory for every student to graduate.
Even though internships in CS are not technically required for a degree, they greatly benefit both employers and students. They help students get familiar with the U.S. professional system. They also give American employers early access to a vital talent pipeline from top American universities.
This policy change cuts U.S. employers off from vital STEM talent pipelines and tells the world’s brightest minds to go elsewhere. This is terrible for our national interest and global competitiveness.
We need to retain top talent, not criminalize academic internships.
Jenai Xuning Yang retweeted
We have just open sourced Droid+, a really simple and easy way to drive the Droid hardware system using modern web technologies and very few dependencies! I hope people find it useful! github.com/NVlabs/DroidPlus
We released DROID+, a new and improved DROID control stack from SRL. We believe robots (and hardware) should be easy to setup and easy to use, and so we updated the DROID stack to be the same. @HugoHadfield1
github.com/nvlabs/droidplus
Here's pi0.5 doing our "hello world" task:
Our team at SRL is hiring!
We're looking for people who dive deep, think systematically, and care deeply about doing rigorous work. If you're excited about understanding the science behind building foundation models, please apply!
We're hiring! The Seattle Robotics Lab (SRL) at NVIDIA is hiring a research scientist to work on robot foundation models. The focus will be on investigating the science behind building VLAs, WAMs, and world models for robotics, and leveraging the findings to propose the next generation of models and training paradigms.
We're seeking a world-class researcher (PhD grad) or senior researcher (3+ years of post-PhD experience) with expertise in training generalist policies. The researcher should aim to question conventional wisdom, push beyond the status quo, and dive into the foundations of general-purpose robot intelligence.
NVIDIA SRL provides a rare environment to conduct open research on fundamental and applied robotics, with the freedom to publish, collaborate broadly, and tackle big scientific questions at scale.
If this sounds like a strong fit, please reach out to us and apply here!
nvidia.eightfold.ai/careers/…
nvidia.eightfold.ai/careers/…
There's a growing gap between how fast generalist robot policies are improving and how well we can actually evaluate that progress, and understand when and why they fail.
I wrote a blog on our take: developer.nvidia.com/blog/ho…
Want to discuss? Find me at #RSS2026!
Jenai Xuning Yang retweeted
The next wave of robot foundation models is here. ⚡
World-Action Models are emerging as a powerful new approach for physical AI, combining pretrained scene dynamics with action generation.
Learn more ➡️ nvda.ws/43C1abm
A fantastic technical deep dive into what's happening now and next in robot foundation models from @moritz_reuss. We're thinking deeply about *how* to build an efficient task generalist policy. Check it out!
World-Action Models (WAMs) have become the second dominant recipe for robot foundation models, next to classical VLAs.
So where do they come from, and how do they compare vs VLAs?
I wrote an small overview of the WAM landscape, with some personal takes:
developer.nvidia.com/blog/pr…
VoLo is a really good way to address long horizon manipulation problems is to combine a variety of tools and capabilities in robotics into a single system. We call this a physical orchestrator.
We've come a long way since code-as-policies.
Wonderful to be back from #CVPR2026, and excited to share the release of our follow-up work:
VoLo: A Physical Orchestrator for Open-Vocabulary Long-Horizon Manipulation
VoLo introduces the idea of a physical orchestrator for open-vocabulary, long-horizon manipulation. Our goal is to move toward robots that can reason, plan, act, monitor, and recover by adaptively using VLA/WAMs, vision models, and action primitives as tools.
We introduce three main contributions:
🤖 VoLoAgent — a physical orchestrator that plans, monitors, and recovers by adaptively using, halting, and redirecting robot actions with tools.
📊 RoboVoLo — a high-fidelity benchmark with 126 open-vocabulary long-horizon manipulation tasks spanning common sense, memory/state tracking, complex references, and world knowledge.
📈 A large-scale empirical study comparing action models, code-as-policy systems, TAMP-style systems, and ablations of the VoLoAgent orchestrator, complemented by real-robot experiments.
This work was done during my internship at @NVIDIA and would not have been possible without my brilliant collaborators: Hugo Hadfield, Alexander Zook, @mikacuy, @luke_ch_song, @erwincoumans, @xuningy, Faisal Ladhak, @qu_1006, @BirchfieldStan, Jonathan Tremblay, and @robovalts. Huge thanks to everyone!
🔗 Project: chicychen.github.io/VoLo/
🔗 Previous work, SpaceTools: spacetools.github.io/
#Robotics #EmbodiedAI #VisionLanguageModels #VLAModels #RobotLearning #NVIDIA #CVPR2026 #LongHorizonManipulation #AI #ComputerVision
Jenai Xuning Yang retweeted
Cosmos3 (post-trained on DROID) surpassed strong VLA & WAM baselines to rank #1 on RoboLab
All the compute FLOPs invested during the massive Cosmos3 pre-training and mid-training contribute to unlocking a better robot foundation model.😄
🎉 We added 2 SOTA WAMs to the RoboLab Leaderboard 🎉
Current leaders on RoboLab-120 (specific instr.):
🥇Cosmos3-Nano-Policy (39.7%)
🥈π0.5 (28.1%)
🥉DreamZero (28.1%)
→ See full results at: research.nvidia.com/labs/srl…
→ All policy clients available at: github.com/NVlabs/RoboLab/
🎉 We added 2 SOTA WAMs to the RoboLab Leaderboard 🎉
Current leaders on RoboLab-120 (specific instr.):
🥇Cosmos3-Nano-Policy (39.7%)
🥈π0.5 (28.1%)
🥉DreamZero (28.1%)
→ See full results at: research.nvidia.com/labs/srl…
→ All policy clients available at: github.com/NVlabs/RoboLab/
Jenai Xuning Yang retweeted
a new form of greeting has dropped: see you at icml🇰🇷
1/1 accepted to ICML. more details soon
Jenai Xuning Yang retweeted
We are still far from zero-shot policy deployment on new tasks
Evaluation is a critical bottleneck in building robot foundation models. Check out our latest work RoboLab, led by @xuningy, which addresses this exact challenge.
Its a high-fidelity simulation environment for testing these models. A truly generalist policy should be able to complete these tasks zero-shot, and this benchmark highlights exactly how far we still have to go. More info 👇
Jenai Xuning Yang retweeted
Generalist robot policies need a benchmark that works across any robot and any policy. 🦾
Introducing RoboLab, a high‑fidelity simulation benchmark built on NVIDIA Isaac and Omniverse to evaluate generalist robot policies in diverse, photoreal, physics‑based environments.
Coming soon to the NVIDIA Isaac Lab‑Arena roadmap for large‑scale, robotic policy evaluation. 📖 nvda.ws/47RbOgX
#NationalRoboticsWeek
→ Customization: Comes with 200+ objects, 100+ backgrounds, lighting, camera poses… don’t like it? No problem, add your own
→ Diagnostics: motion quality, failure events, + sensitivity analysis for failure attribution
(5/6)
RoboLab comes with RoboLab-120 — a curated, diverse benchmark of 120 tasks to get started.
Set up and run in <20 min. (6/6)
Try it out 👇
🌐 research.nvidia.com/labs/srl…
📄 arxiv.org/abs/2604.09860
💻 github.com/NVLabs/RoboLab