PhD student @NYUDataScience @CILVRatNYU @agentic_ai_lab
New York
- Tweets 146
- Following 628
- Followers 2,240
- Likes 933
Warm take: Your world model should never stop learning
Introducing AdaJEPA, an adaptive WM that plans, acts, and adapts in a closed loop. Every action leads to a new observation, and every transition refines the latent representation and prediction.
📝: agenticlearning.ai/adajepa/
30
137
932
120,528
Ying Wang retweeted
[1/n] Can a world model drive without training a driving policy?
Introducing AD-E2E-JEPA, a JEPA-based world model for end-to-end autonomous driving.
📄: arxiv.org/abs/2609.34085
💻: github.com/HaoranZhuExplorer…
Joint work with @kevinghstz, Prof. @ylecun, and Prof. Anna Choromanska
5
18
103
17,797
Join us on Oct 10 if you’re a world model lover in SF! I will share our ICML paper on representation learning for WMs and recent work AdaJEPA. Excited to attend my first reading club 📚!
📚 Saturday Robotics @saturdayrobotic & World Models Reading Club #33: Perceiving, Predicting, and Planning for Physical Interaction + AdaJEPA
San Francisco · October 10 · 2–5 PM
👉 RSVP: luma.com/y9omqzz0
What does a robot actually need to understand before it can reliably act in the physical world?
For Reading Club #33, we’re bringing together two perspectives on robot world models: grounding intelligence in physical interaction and building latent spaces that make planning and adaptation work.
🎤 @Hongyu_Lii — Robotics Researcher, @NVIDIARobotics, @nvidia
Perceiving, Predicting, and Planning for Physical Interaction
Robots manipulating the real world need to do more than recognize objects. They need to perceive contact, predict physical change, and plan under uncertainty.
Hongyu will present a unified line of work spanning:
• NovaFlow — zero-shot manipulation by distilling actionable 3D flow from generated video
• NovaPlan — closing the loop on long-horizon tasks through video-language planning
• Deform360 — large-scale multi-view visuotactile data for deformable world models
• Hydra-0 — a generalist world model conditioned on action flow, representing robot actions as pixel motion
The bigger question: can a shared visual interface connect touch, vision, actions, world models, policy evaluation, and control across different robots, tasks, and environments?
🎤 @yingwww_ — @NYUDataScience
AdaJEPA: Adaptive World Models for Planning Under Distribution Shift
JEPA-style world models predict the consequences of actions in a learned representation space—but what makes that latent space actually useful for planning?
Ying will discuss temporal straightening, which encourages locally straight latent trajectories so Euclidean distance becomes a better proxy for geodesic distance and makes gradient-based planning more stable.
She will also introduce AdaJEPA, an adaptive world model that plans, acts, and adapts in a closed loop: every action produces a new observation, which updates the latent representation and prediction when reality diverges from the model.
Two complementary questions:
🧠 What should a robot represent about the physical world?
🔄 What should it do when its world model is wrong?
📍 San Francisco
🗓 Saturday, October 10, 2026 | 2–5 PM
2:00–2:30 — Doors Open & Social 🍓🧋
2:30–3:30 — Keynote
3:30–5:00 — Q&A + Open-Floor Technical Roundtable
Hosted by @junfanzhu98 & @aurorafeng_01 & Audrey Pe (@sievedata).
Come discuss robot world models, visuotactile perception, deformable objects, action-conditioned video models, latent planning, distribution shift, test-time adaptation, and what it actually takes to build robots that can perceive → predict → plan → act → adapt.
👉 RSVP: luma.com/y9omqzz0
#Robotics #WorldModels #EmbodiedAI #PhysicalAI #RobotLearning #JEPA #RobotManipulation #NVIDIA #NYU
3
16
206
38,598
Ying Wang retweeted
Right. Dario was already claiming that GPT2 was too dangerous to open source back in 2019.
I made fun of them then.
Everyone should make fun of them now.
395
3,191
26,030
1,984,860
Ying Wang retweeted
Really excited to be part of this workshop! What makes a pretrained policy easy to adapt is still an open question, and we'd love for you to join the discussion! Submission deadline is Oct 4 ✨
🤖 Excited to announce our CoRL 2026 workshop: Pretrain to Adapt! 🔗 corl2026-robotcl.github.io
How should we pretrain robot policies not just to perform well today, but to adapt effectively to new tasks, environments, embodiments, sensing modalities, etc?
To keep the conversation going beyond the workshop, we’re hosting a Slack community for anyone interested in robot pretraining + adaptation. Join our Slack to connect with fellow researchers!
3
19
1,878
Ying Wang retweeted
Introducing Outerloop: open autoresearch on your own cluster. My lab develops and uses it in our own research.
Outerloop 0.1.0 is out. A fleet of AI agents that improve your benchmarks on your code. Agents propose changes, run them on your cluster, and open pull requests when the metrics improve. We are open source. Let's embrace open science! Get started at outerloop.science
2
5
51
5,986
Ying Wang retweeted
🤖 Excited to co-organize our #CoRL2026 Workshop on Bringing Physics Simulation and World Models Together for Robotic Manipulation!
Come join us in Austin and submit your paper by Sept. 30!
🔗Website: corl26ws-physwm.github.io
🚀 Excited to announce our #CoRL2026 Workshop: Bringing Physics Simulation and World Models Together for Robotic Manipulation! corl26ws-physwm.github.io
🤖 We’ll explore a central question: How can insights from modern physics simulation help build world models for better physical reasoning and robotic control?
🎤 Thrilled to bring together an amazing lineup of invited speakers & panelists: Yuval Tassa (@yuvaltassa), Erwin Coumans (@erwincoumans), Yann LeCun (@ylecun), Yunzhu Li (@YunzhuLiYZ), Lucy Shi (@lucy_x_shi), and Sherry Yang (@sherryyangML)
📢 Call for Submissions! Submit your work to our #CoRL2026 Workshop! Papers up to 4 pages; dual submissions welcome. 🏆 $500 Best Paper Award. ⏰ Deadline: Sept. 30. Submit via OpenReview: openreview.net/group?id=robo…
Join us in Austin! 🤖🌎
#CoRL2026 #WorldModels #PhysicsSimulation #RobotLearning #Robotics #EmbodiedAI #PhysicalAI #AI
5
25
1,953
Ying Wang retweeted
New positions open! We’re looking for full-time researchers and interns with backgrounds in geometry and 3D vision to join our team at AMI Labs.
The roles can be based in NYC, Paris, Montreal, or Singapore.
If you’re interested, apply through the link below:
jobs.ashbyhq.com/ami/0d7332d…
17
48
692
94,151
I published a new post on my personal blog!
Joining AMI to work on World Models
lihaoyi.com/post/JoiningAMIt…
7
23
345
26,820
Ying Wang retweeted
Introducing LpWM: A Case for Sparse Representations in World Models
Dense Gaussian representations are a choice, not a requirement. We find that sparse representations can make latent dynamics easier to model for planning.
📄arxiv.org/abs/2608.22764
💻github.com/YilunKuang/lpworl…
14
50
266
46,229
Ying Wang retweeted
What if robots, humans, and video models could all describe action in the same language? Introducing Hydra-0🐍: a generalist world model that represents robot actions as action flow. One interface. Many embodiments, tasks, and environments.
🧵1/7
8
76
469
81,669
Ying Wang retweeted
CFP: Continual World Models Workshop @ NeurIPS 2026
🗓 Submission deadline: Aug 29, 2026
📍 Sydney, Australia
🌐 continual-world-models-works…
World models are increasingly capable of representing complex environments, yet most are still trained largely as static models. A central challenge is building world models that continually learn and adapt to a changing world.
At CWM @ NeurIPS 2026, we bring together researchers across world models, continual learning, robotics, multimodal learning, memory, and agents.
We are also recruiting reviewers! If you are interested in becoming a reviewer, please submit your information through this form:
docs.google.com/forms/d/e/1F…
We hope to see your work and have you join us at the workshop!
#NeurIPS2026 #WorldModels #ContinualLearning #EmbodiedAI #Robotics
14
102
6,844
Ying Wang retweeted
For about 10 years now, I have argued that the *only* way forward is for AI technology to be widely available, shared, and open.
Like the printing press and the Internet, AI amplifies human intelligence and efficiency by improving access to knowledge.
To empower individuals, societies require a high diversity of AI systems with different value systems, linguistic abilities, philosophical/political biases, and specific expertise.
We need diverse AIs for same reason we need a diverse press.
Given the cost and complexity, this can only be achieved through open foundation models on top of which anyone can build systems with their languages, biases, expertise, and value systems.
I have been more vocal about this over the last 4 years, since AI popped into the public discourse.
I have made the argument in various forums: corporate C-suites, AI safety discussion groups, professional meeting, the US Senate, the UN Security Council, and the public sphere through media interviews, podcasts and social media posts.
I totally agree with @finkd Mark Zuckerberg's recent piece in which he writes: "the notion that AI is so dangerous that the only safe path is an extreme concentration of power seems inherently problematic. Historically, hoping that an absolute power will benevolently provide for humanity if sufficiently enlightened has not led to safe or positive outcomes.”
When @DarioAmodei writes: “some may object that we can simply keep AIs in check with a balance of power between many AI systems, as we do with humans", he is talking about me, among (thankfully) many others. It is the only good path forward.
There will be nefarious uses of AI, as there have been with every technology ever invented.
But it will be your Bad AI against my Good AI.
118
261
1,802
198,696
Ying Wang retweeted
We are organizing Scaling H2R 🦾workshop at @corl_conf this year!
Submit papers & join us if you are interested in exploring human data for robot learning
scaling-h2r-corl.github.io/
4
39
10,344
Ying Wang retweeted
We have extended the submission deadline for the HRD workshop @ IROS to Aug 17th!
Website: human-robot-dialogue.github.…
Announcing the Human-Robot Dialogue Workshop #IROS2026 (Pittsburgh, Oct 1)!
Can dialogue improve robot learning, human-robot interaction, and effective and safe deployment in human environments?
We're bringing together robotics, NLP, HRI, CV, foundation models, & cognitive science researchers to discuss human-robot dialogue.
Speakers and Panelists: @_jessethomason_ (Georgia Tech), @imankitgoyal (NVIDIA), @jacobandreas (MIT), @BaharIrfan_ (Familiar Machines & Magic), @GabrielSkantze (KTH/Furhat) & @matt_marge (DARPA)
📝 Call for short and long papers due on Aug 3, 2026
🔗 human-robot-dialogue.github.…
We welcome short and long papers on human-robot dialogue, language grounding, embodied communication, dialogue policy learning, evaluation, datasets, and adjacent topics.
8
7
1,687
We're hosting the World Model + Uncertainty workshop at #CoRL2026 🤖 If you're working on world models, uncertainty, planning..., submit your work by September 23!
Announcing Modeling Uncertainty in Robotic World Models Workshop at CoRL 2026.
📍 Austin, TX · Nov 12, 2026 📄 Submissions due Sept 23 (AoE)
Learned world models are getting powerful. They're still bad at knowing what they don't know. 🧵
4
11
116
13,675
Ying Wang retweeted
Hong Wang received the Fields Medal at the Opening Ceremony of ICM 2026 this morning—becoming only the third female mathematician to be honored with the prize. Congratulations, Professor Wang!
12
107
904
93,506
Ying Wang retweeted
Pretrained ViTs see the world in rich, dense detail. Most policies pool it to a single vector before acting, discarding most of it.
We introduce Patch Policy: a minimal architectural extension that enables transformer-based policies to consume dense tokens directly, no billion-param VLM required. It outperforms a fine-tuned 7B VLA by 18% with ~0.7% of its parameters, enabling robust, precise manipulation.
17
40
281
55,116
#icml2026
- Excited to present our icml paper on representation learning for WMs. Thx everyone who stopped by!
- Had 10+ coffee chats about WMs and JEPA 🥳
- Met old friends and made many new ones
- A lot of amazing food 🤩
- Feeling energized for what’s next 👀 Stay tuned!
What is a good latent space for world modeling and planning? 🤔
Inspired by the perceptual straightening hypothesis in human vision, we introduce temporal straightening to improve representation learning for latent planning.
📑: agenticlearning.ai/temporal-…
3
2
62
4,977
Warm take: Your world model should never stop learning
Introducing AdaJEPA, an adaptive WM that plans, acts, and adapts in a closed loop. Every action leads to a new observation, and every transition refines the latent representation and prediction.
📝: agenticlearning.ai/adajepa/
30
137
932
120,528
Thanks for reproducing our work! Official code coming after ICML 🫶
Your world model might be wrong the moment it's deployed 👀
AdaJEPA's answer: let it keep learning while it acts 🕹️
Reproduced it 1 grad step per replan and the model literally un-fails itself mid-episode 🚀
p.s., Ada Lovelace paper should be next ❤️
2
35
7,279