Jeffrey Wang (@jwang633) profile picture

Undergrad @Harvard & @Cornell Researcher @StanfordAILab @StanfordSVL Building | Agentic Robotics, Robotic Learning, Perception, and Control

Stanford, CA
Joined June 2026
Photo by Jeffrey Wang (@jwang633)
Learning bimanual dexterous manipulation from a single human video? Agents can be a solution! Introducing DexAgent, an agentic Human2Sim2Robot framework for dexterous manipulation with a self-evolving tool library. We evaluated DexAgent on 11 real-world, long-horizon dexterous manipulation tasks spanning rigid, articulated, and deformable objects. 🤖 63.6% policy rollout success rate, compared with 18.2% for the strongest baseline ⚡ 2.1 hours average inference time with the tool library, compared with 3.3 hours for the baseline 🧰 A self-evolving library with 103 skills + 188 verifiers, designed to keep growing as DexAgent encounters new tasks and objects Learn more and contribute to the growing tool library: Project Website: dexagent1.github.io/ Arxiv: arxiv.org/abs/2609.35318
3
11
2
78
12,282
Jeffrey Wang retweeted
Long-horizon, whole-body robot policies still fail during deployment. How do we best close this last-mile gap? 🤖 🚨 Introducing I3L: Iterative Interactive Imitation Learning for Whole-Body Mobile Manipulation Website: i-3l.github.io/ Paper: arxiv.org/abs/2610.17748
8
39
11
219
62,629
Jeffrey Wang retweeted
🤖Robots need world models for visual planning, but they don't need to imagine every frame. Introducing DreamCan: Sparse Visual Planning for Long-Horizon Robot Manipulation. Project: dreamcan.github.io/ Code: github.com/HeegerGao/DreamCa… Paper: arxiv.org/abs/2610.24981
15
43
9
288
54,894
Can a robot learn dexterous manipulation from just one human video? Check out our latest SAIL blog post on DexAgent, an agentic framework that turns a single egocentric video into robot training data using a self-evolving tool library! ai.stanford.edu/blog/dexagen…
4
6
45
10,119
Jeffrey Wang retweeted
Human video is an undiscovered goldmine for robot learning. Excited to see @jwang633 and the team exploring its potential with Sharpa’s dexterous hands. The human-like form of Sharpa W01 makes it well suited to translating natural hand motion into robotic movement. The team takes this further, showing how human video can become a powerful source for robots to learn real-world, long-horizon, contact-rich manipulation.
Learning bimanual dexterous manipulation from a single human video? Agents can be a solution! Introducing DexAgent, an agentic Human2Sim2Robot framework for dexterous manipulation with a self-evolving tool library. We evaluated DexAgent on 11 real-world, long-horizon dexterous manipulation tasks spanning rigid, articulated, and deformable objects. 🤖 63.6% policy rollout success rate, compared with 18.2% for the strongest baseline ⚡ 2.1 hours average inference time with the tool library, compared with 3.3 hours for the baseline 🧰 A self-evolving library with 103 skills + 188 verifiers, designed to keep growing as DexAgent encounters new tasks and objects Learn more and contribute to the growing tool library: Project Website: dexagent1.github.io/ Arxiv: arxiv.org/abs/2609.35318
1
4
18
1,417
Jeffrey Wang retweeted
Check out RoboRender! our new work on video generation for Sim2Real robot data. We trained our own video model to turn simulation into real-world robot videos, so policies trained entirely in simulation transfer zero-shot to the real world. Website: robo-render.github.io Joint work with @RavenHuang4, @wensi_ai, @jwang633, Zijian Du, Yang Liu, Jiaolong Yang, @drfeifei, and @jiajunwu_cs. Thanks to all coauthors!
Real2sim has made impressive progress with coding agents. But what about sim2real, especially the visual gap? Meet RoboRender: turning simulated trajectories into photorealistic videos for policy training, achieving 71% average success in zero-shot real-world robot deployment.
1
5
31
2,469
Jeffrey Wang retweeted
This is so cool – at Palo Alto, CA
Real2sim has made impressive progress with coding agents. But what about sim2real, especially the visual gap? Meet RoboRender: turning simulated trajectories into photorealistic videos for policy training, achieving 71% average success in zero-shot real-world robot deployment.
3
13
1,422
Jeffrey Wang retweeted
Check out RoboRender, our work on turning simulation trajectories into photorealistic multi-view videos for robot policy training!
Real2sim has made impressive progress with coding agents. But what about sim2real, especially the visual gap? Meet RoboRender: turning simulated trajectories into photorealistic videos for policy training, achieving 71% average success in zero-shot real-world robot deployment.
3
22
1,073
Sim2real visual gap has always been a challenge in robotic learning. Can video generation models be a solution? Introducing RoboRender — a generative video framework that transforms simulated robot trajectories into realistic training data, enabling zero-shot transfer to the real world. Across 8 tasks and 3 robot embodiments, RoboRender achieves 71% real-world success — 7.1× over raw simulation.
Real2sim has made impressive progress with coding agents. But what about sim2real, especially the visual gap? Meet RoboRender: turning simulated trajectories into photorealistic videos for policy training, achieving 71% average success in zero-shot real-world robot deployment.
1
5
18
933
Great work done with @RavenHuang4, @wensi_ai @ziyuchen_ , Zijian Du, Yang Liu, Jiaolong Yang, @drfeifei, and @jiajunwu_cs
1
189
Jeffrey Wang retweeted
Real2sim has made impressive progress with coding agents. But what about sim2real, especially the visual gap? Meet RoboRender: turning simulated trajectories into photorealistic videos for policy training, achieving 71% average success in zero-shot real-world robot deployment.
14
25
7
282
24,360
Mobile manipulation policies can break from base pose errors of just a few cm, which are common after navigation. Can we get pose generalization without additional demos? Introducing MobileVISTA: collect demos at one base pose, get a policy that works from many 🧵👇 Website: sdwistreich.github.io/mobile… Paper: arxiv.org/abs/2610.07511
12
18
4
92
7,682
Jeffrey Wang retweeted
Do coding agents understand the dynamics of the world well enough to reconstruct it from video? We introduce 4DCodeBench to evaluate this ability through 4D inverse graphics. 4DCodeBench.com/ <🧵>
26
106
33
628
67,464
Teaching robots dexterous skills from human video is difficult when objects are articulated, deformable, or physically diverse. DexAgent addresses this challenge with an adaptive Human2Sim2Robot framework that turns a single egocentric human video and task prompt into physically grounded robot trajectories for policy training. The system reconstructs the task in simulation, selects or develops task-specific tools, verifies each stage, and uses feedback to correct errors before they propagate. It can also vary object and robot states to generate diverse training data, while retaining newly developed skills and verifiers for future tasks. Across 11 real-world tasks, policies trained on DexAgent-generated data achieved a 3.5× higher success rate than competing baselines. Title: DexAgent: An Agentic Human2Sim2Robot Framework for Dexterous Manipulation with Self-Evolving Tool Library Project URL: dexagent123.github.io/ Paper: arxiv.org/abs/2609.35318v1
1
5
20
2,051
Jeffrey Wang retweeted
Astra already shows exciting potential for real2sim2real for solving robotics tasks. Our DexAgent harness takes it further: no task-specific robot demos, 3.9× the success rate of zero-shot Astra across 11 real-world tasks (63.6% vs. 16.4%)! Check it out!
Learning bimanual dexterous manipulation from a single human video? Agents can be a solution! Introducing DexAgent, an agentic Human2Sim2Robot framework for dexterous manipulation with a self-evolving tool library. We evaluated DexAgent on 11 real-world, long-horizon dexterous manipulation tasks spanning rigid, articulated, and deformable objects. 🤖 63.6% policy rollout success rate, compared with 18.2% for the strongest baseline ⚡ 2.1 hours average inference time with the tool library, compared with 3.3 hours for the baseline 🧰 A self-evolving library with 103 skills + 188 verifiers, designed to keep growing as DexAgent encounters new tasks and objects Learn more and contribute to the growing tool library: Project Website: dexagent1.github.io/ Arxiv: arxiv.org/abs/2609.35318
1
4
49
5,999
This work was done at the wonderful @StanfordSVL with wonderful people! Also, biggest thanks to the supervision of @RavenHuang4, @jiajunwu_cs, @drfeifei, and @YunzhuLiYZ, who provided me with tremendous and constructive advice!
3
529