Bo-Ruei Huang✈️ CoRL

Robotics PhD Student @CSatUSC | Prev. Intern @MIT_CSAIL & @Caltech | B.S. @NTU_TW CS & EE

[Bo-Ruei Huang✈️ CoRL @borueihuang]

I'm presenting WACL today in #CoRL2026!
Poster Session 3 #58, 11/11 10:30 am

[Bo-Ruei Huang✈️ CoRL @borueihuang]

Coffee in one hand, a door in your way. You bump it open with your elbow. Can a robot do that? 🤔

We introduce WACL, which learns whole-arm manipulation from human demos. No teleop or simulation. Zero-shot transfer to a new robot. Thread 👇
https://t.co/kC1XBkrxij (1/9)

[Bo-Ruei Huang✈️ CoRL @borueihuang]

To learn more, check out our paper and chat with us at CoRL 2026 next week!

Huge thanks to @daniel_t_seita and @ebiyik_, USC LiraLab 💛
📄 https://t.co/kBmli8WWac
🌍 https://t.co/kC1XBkrxij
CoRL 2026 Austin, Poster Session 3 #58, 11/11 10:30 am. (9/9)

[Bo-Ruei Huang✈️ CoRL @borueihuang]

Coffee in one hand, a door in your way. You bump it open with your elbow. Can a robot do that? 🤔

We introduce WACL, which learns whole-arm manipulation from human demos. No teleop or simulation. Zero-shot transfer to a new robot. Thread 👇
https://t.co/kC1XBkrxij (1/9)

[Bo-Ruei Huang✈️ CoRL @borueihuang]

Does it transfer to another robot? 🤖

We swapped the xArm for a Franka FR3. Same policy, same human demos, no new data. With the robot-agnostic abstractions, WACL generalizes zero-shot to the FR3, with a 68% success rate under constraints. (8/9)

[Jesse Zhang ✈️ CoRL 2026 @Jesse_Y_Zhang]

A reward model that works, zero-shot, across robots, tasks, and scenes?

Introducing Robometer: Scaling general-purpose robotic reward models with 1M+ trajectories.

Enables zero-shot: online/offline/model-based RL, data retrieval + IL, automatic failure detection, and more!

🧵 (1/12)

[Bo-Ruei Huang✈️ CoRL @borueihuang]

Haoshu is fantastic and will definitely be a good advisor. Join his lab to work on cool robotics!

[Hao-Shu Fang @haoshu_fang]

Life update ✨
I’ll be joining @umdcs as faculty starting Spring 2026!
I’m looking for motivated students (undergrad, MS, PhD) and postdocs interested in robotic manipulation, tactile sensing, and learning-driven hardware design.

Excited to build my group at UMD—let’s connect!

[Hao-Shu Fang @haoshu_fang]

How do we unlock the full dexterity of robot hands with data, even beyond what teleoperation can achieve?

DEXOP captures natural human manipulation with full-hand tactile & proprio sensing, plus direct force feedback to users, without needing a robot👉https://t.co/rjfQ9nzofm

[Bo-Ruei Huang✈️ CoRL @borueihuang]

@haoshu_fang @umdcs Congrats Haoshu!

[Bo-Ruei Huang✈️ CoRL @borueihuang]

I'll present my work "Diffusion Imitation from Observation" with @shaohua0116 tomorrow at #NeurIPS2024. Please come to have a chat!

Thu 12/11 11am-2pm West Ballroom A-D #6307

[Project Page] https://t.co/AvsNNwCPRn

[Shao-Hua Sun @shaohua0116]

#NeurIPS2024 @xinyun_chen_ (Google DeepMind), @levilelis (U of Alberta), and I will be giving a tutorial on programmatic reinforcement learning and code generation on Dec 10 (Tue) 1:30-4 PM @ East Exhibition Hall A. Come say hi!

[Hao-Shu Fang @haoshu_fang]

A good hand can push intelligence development. Introducing Eyesight Hand, equipped with full-hand high-res tactile sensors and proprioceptive actuators. It is compliant, agile, and powerful. Good tactile sensing makes learning more efficient and robust. Shout out to Branden!