Rui Chen @ray695_c

Research Assistant Professor at Tsinghua University. Embodied AI, robot learning, 3D vision, tactile sensing. callmeray.github.io/homepage/ Joined October 2017

Tweets

27
Followers

190
Following

172
Likes

110

Fangchen Liu @fangchenliu_

6 months ago

1/N Most Vision-Language-Action models need tons of data for finetuning, and still fail for new objects and instructions. Introducing OTTER, a lightweight, easy-to-train model that uses text-aware visual features to nail unseen tasks out of the box! Here's how it works 👇

12 65 313 67K 191

Download Video

Li Yi @ericyi0124

7 months ago

Excited to share our #ICLR2025 work, DexTrack, where we introduce a neural controller that mimics a wide range of human manipulations. This unlocks the potential for large-scale conversion of internet hand motion into robot manipulation trajectories.

Xueyi Liu @xymeow7

7 months ago

7 26 155 22K 71

Download Video

3 5 78 8K 22

Rui Chen @ray695_c

8 months ago

🎉 Happy New Year 2025! 🚀 Stage 1 of ManiSkill-ViTac Challenge 2025 is officially LIVE! 🦾 Compete in contact-rich challenging vision-tactile manipulation tasks for a $14,500 prize pool! 📅 Phase 1 DDL: Feb 15, 2025 🔗 Join now: github.com/cyliizyz/ManiS… #ICRA2025 #EmbodiedAI

Rui Chen @ray695_c

8 months ago

0 5 20 8K 8

Download Image

0 4 15 4K 4

Rui Chen @ray695_c

8 months ago

Our ViTac Workshop has been accepted for #ICRA2025! 🚀 Join us and explore the future of touching and seeing—where vision meets tactile sensing! 👀🤝 Stay tuned for updates! #Robotics #EmbodiedAI

Rui Chen @ray695_c

9 months ago

2 14 81 11K 26

Download Gif

0 5 20 8K 8

Download Image

Hao Su @haosu_twitr

11 months ago

I was recently interviewed by AsiaTechDaily around Hillbot's work and vision for the robotic future: asiatechdaily.com/the-next-ai-wa…

0 11 133 9K 20

Stone Tao @Stone_Tao

11 months ago

This work now has a paper you can cite on arxiv now! arxiv.org/abs/2410.00425

0 1 9 515 3

Rui Chen @ray695_c

12 months ago

Physics simulation is a great way to model the world since the generalizability to unseen actions is guaranteed. Once we build the precise world model in simulation, we can use MPC to find feasible trajectories for diverse task goals without any RL training. #WorldModel #MPC

TAORAN JIANG @TaoranJiang

12 months ago

1 1 11 4K 4

Download Video

0 3 23 3K 11

Stone Tao @Stone_Tao

a year ago

📢 ManiSkill 3 beta is out! Simulate everything everywhere all at once 🥯 - 18K RGBD FPS on 1 GPU, 3K on Colab! - Diverse parallel GPU sim - Tons of new robots/tasks All open-sourced: github.com/haosulab/ManiS… Photo: MS3 Tasks w/ scenes from AI2THOR and ReplicaCAD 🧵(1/6)

9 65 263 58K 121

Download Image

Yuzhe Qin @QinYuzhe

2 years ago

We tackle robotic manipulation with real-world imitation learning or sim2real approaches. But what if we merge the two? Introducing CyberDemo: a novel pipeline that leverages simulated human demos to master real-world dexterous manipulation tasks.