Weijun Dong (@dwjshift) profile picture

Ph.D. student @ucsd_cse | Robot Learning | Prev: M.Eng. & B.Eng. @Tsinghua_IIIS

Joined November 2020
Photo by Weijun Dong (@dwjshift)Photo by Weijun Dong (@dwjshift)Photo by Weijun Dong (@dwjshift)Photo by Weijun Dong (@dwjshift)
What makes reinforcement learning efficient for training generalist robot policies? A key ingredient is the action representation: what choices do we give the learner to explore? A suitable action representation can make the learning problem substantially easier. 🧵(1/6)
5
12
2
83
18,070
Honored to co-lead RAX with Xiangcheng (@xczhangxx)! Excited to share what robots can learn through scaling exploration—no demos, no model finetuning! We’re open-sourcing the full framework and every experiment session. Please fell free to try out! Project + sessions: robot-agent-exploration.com/ Paper: assets.robot-agent-explorati… GitHub: github.com/robot-agent-explo… Huge thanks to @du_yilun for your advice and support, and to all our amazing collaborators—it’s been a joy working with you! @TonyC56348978 @yitongli165665 @wangyenjen @Lyubh22 @haotian_yeee @ChuyeHong @YXWangBot @Huangyu58589918 @Haoyu_Xiong_ @yangzhen04 @xieenze_jr @MandersonTravis @StefanoErmon @pabbeel Antonio Torralba @dawnsongtweets @Weirui_Ye
Robotics has spent years scaling demonstration data and model size. What happens when we scale EXPLORATION for robot agents? We're incredibly excited to share RAX, an agentic robotics system that learns by scaling online exploration. A coding agent connects directly to the robot’s perception and control APIs, learns from feedback, and builds programs that perceive, plan, and act. The model stays frozen. As exploration scales, the agent improves its programs and builds up reusable skills. The results: • ZERO teleoperation or action demonstrations on real robots—from insertion with 0.1 mm clearance and air-fryer manipulation to deformable objects and long-horizon mobile whole-body control. • +37 percentage points on average over leading learned policies across five simulation benchmarks: RoboDojo, BEHAVIOR-1K, HumanoidArena, DexJoCo’s bimanual tasks, and LIBERO-Pro. • One framework spanning single-arm, bimanual, dexterous, humanoid, and mobile robots, with new embodiments connected through APIs. The craziest part: on each real-world embodiment, going from scratch to mastering the tasks took roughly two hours. No teleop. No model finetuning. We’re open-sourcing the full framework AND every experiment session. We can’t wait to see what people build with it! Project + sessions: robot-agent-exploration.com/ Paper: assets.robot-agent-explorati… GitHub: N Project fully led by @RunhanH and @xczhangxx who are absolutely awesome. How it works, experiments, and robot videos below 🧵 (1/N)
5
11
2
74
3,959
Weijun Dong retweeted
🤖 Why do generative policies beat regression policies? Multi-modality? Manifold adherence? Or more? Our analysis suggests: highly imbalanced gradients under MSE! A loss change closes the gap: no denoising, competitive performance, faster inference. 📄 arxiv.org/abs/2610.12231
22
38
5
289
30,192
Flexible Locomotion Learning with Diffusion Model Predictive Control Excited to share that our paper has been accepted to #ICRA2026 @ieee_ras_icra! A diffusion-planning framework for flexible real-world quadruped locomotion. Instead of learning a fixed RL policy or relying on hand-crafted dynamics for MPC, we train a diffusion trajectory prior that jointly predicts future states and actions. Key Ideas: Diffusion-MPC: A diffusion planner unlocks flexible locomotion through test-time reward and constraint adaptation Interactive reward-weighted finetuning enables continual behavior refinement from online environment feedback Real-world deployment on Unitree Go2 with efficient and adaptive planning The same planner can adapt at test time to height changes, posture/joint constraints, balancing under external disturbances, energy-aware locomotion, and zero-shot outdoor walking on grass and slopes. 🌐Homepage: flexible-diffusion-mpc.githu… 📖Paper: arxiv.org/abs/2510.04234 🔗Code: github.com/hrh6666/Flexible-… This work is by @RunhanH, Haldun Balim, @hankyang94 , and @du_yilun. #ICRA2026 #Robotics #LeggedRobots #RobotLearning #DiffusionModels #MPC #MachineLearning
3
26
4
84
17,561
Proud to be a co-first author of this work! 🎉 Really excited to share our #ICLR2026 paper — HinFlow: Translating Flow to Policy via Hindsight Online Imitation. Great collaboration with the entire team!
Learning from vast video data allows point-flow-based planners to create generalizable task plans, guiding robots via future keypoint trajectories. But how can we ensure low-level execution doesn't become the system's bottleneck? Introducing our #ICLR2026 paper, HinFlow: Translating Flow to Policy via Hindsight Online Imitation
2
19
2,642
Learning from vast video data allows point-flow-based planners to create generalizable task plans, guiding robots via future keypoint trajectories. But how can we ensure low-level execution doesn't become the system's bottleneck? Introducing our #ICLR2026 paper, HinFlow: Translating Flow to Policy via Hindsight Online Imitation
3
15
7
94
26,578
We hope our work hints at the broader potential of hindsight relabeling — perhaps as a useful ingredient in the era of scaling up robotic foundation models. Check for more details! Webpage: dwjshift.github.io/HinFlow Arxiv: arxiv.org/abs/2512.19269 Code: github.com/yzc0731/HinFlow
1
1
4
405
Huge thanks to collaborators and supervisors: Yitian Zheng, @YeZhangchen0731, @ShengjieWa34067, Yuyang Liu, Chongjie Zhang, @ChuanWen15, and @gao_young!
5
362