RT-X: generalist AI models lead to 50% improvement over RT-1 and 3x improvement over RT-2, our previous best models. 🔥🥳🧵 Project website: robotics-transformer-x.github.io

Very excited to release the Open X-Embodiment Dataset today — the largest robot dataset to date with 1M+ trajectories! Robotics needs more data & this is a big step! robotics-transformer-x.github.io There’s lots to unpack here, so let’s do a deep dive into the dataset! 🧵1/15

There’s a new bot in town 🤖 Check this out (until the very end)! tesla.com/ai

We are now having full conversations with Figure 01, thanks to our partnership with OpenAI. Our robot can: - describe its visual experience - plan future actions - reflect on its memory - explain its reasoning verbally Technical deep-dive 🧵:

Humanoids 🤖 will do anything humans can do. But are state-of-the-art algorithms up to the challenge? Introducing HumanoidBench, the first-of-its-kind simulated humanoid benchmark with 27 distinct whole-body tasks requiring intricate long-horizon planning and coordination. 🧵👇

We're diving into the world of humanoid robotics 🤖! We were curious how to train humanoid robots and ended up making HumanoidBench, a simulated benchmark for humanoid robots! Give it a try and let us know what you think!

After two years, it is my pleasure to introduce “DROID: A Large-Scale In-the-Wild Robot Manipulation Dataset” DROID is the most diverse robotic interaction dataset ever released, including 385 hours of data collected across 564 diverse scenes in real-world households and offices

Access to *diverse* training data is a major bottleneck in robot learning. We're releasing DROID, a large-scale in-the-wild manipulation dataset. 76k trajectories, 500+ scenes, multi-view stereo, language annotations etc Check it out & download today! 💻: droid-dataset.github.io

I'll be presenting DreamSmooth at #ICLR2024! 😊 Reward prediction is vital to model-based RL. Yet, many algorithms struggle to predict sparse rewards accurately. By addressing this, DreamSmooth dramatically improves task performance in many sparse-reward environments. 🧵👇

My first project at Yonsei is finally out :) We introduce a simple yet effective conservative Q-learning formulation, and enable offline model-based RL to solve long-horizon tasks! Kudos to my undergraduate student Kwanyoung Park , applying for PhD this year!

Another paper release :) This paper tackles unsupervised skill discovery (especially, goal-conditioned RL) and achieves diverse skills without supervision in complex AntMazes. Kudos to my amazing undergrads Junik Bae and Kwanyoung Park (they are applying for PhD this year!)

I will be presenting HumanoidBench today at #RSS2024 in the Locomotion and Manipulation session starting at 11am! You will find me at the poster session (Commissiekamer 2) throughout the day 🤖

Excited to be a part of the #CoRL2024 workshop on Mastering Robot Manipulation in a World of Abundant Data!

Want to shape the future of humanoid robots? Join our #CoRL2024 workshop on Whole-body Control & Bimanual Manipulation! Featuring amazing talks from world-renowned experts from both academia and industry.

📢 Announcement: deadline extension! 📢 We have extended the paper submission deadline to October 18. Looking forward to your submissions!

On Saturday, we are organizing the workshop on Whole-Body Control and Bimanual Manipulation: Applications in Humanoids and Beyond. We have an exciting full-day schedule, join us in room Taurus 2: wcbm-workshop.github.io

Join us at Conference on Robot Learning in the afternoon for an inspiring lineup of talks and posters on robotic assembly! Check out how manufacturing and construction are the perfect arenas for roboticists to tackle complex manipulation challenges and drive real-world impact.

#RSS2025 Workshop on Whole-Body Control & Bimanual Manipulation: Applications in Humanoids & Beyond 🤖 Join us with our amazing lineup of speakers from academia & industry! Share your latest work by May 23 wcbm-workshop.github.io Robotics: Science and Systems

We’re releasing the RoboArena today!🤖🦾 Fair & scalable evaluation is a major bottleneck for research on generalist policies. We’re hoping that RoboArena can help! We provide data, model code & sim evals for debugging! Submit your policies today and join the leaderboard! :) 🧵