Adam Rashid retweeted
Is this machine playing? 🏀🪁🛝
We left Eko, an embodied coding agent, on an island for 30 hours with no task or goal, studying it like ethologists study a new species
Eko explored, built, climbed, made games, and learned completely new techniques (1/9)
machineplaying.org
43
53
521
59,045
Adam Rashid retweeted
Excited to release EyeRobot 2.0! We noticed it is better at dealing with distractors. When the gaze policy looks at the right target, the policy can recover
Almost all bimanual robots include wrist cameras. They add bulk, cost, and complexity, but are critical for performance. Are they really needed?
Excited to finally share EyeRobot 2.0, where we use active gaze from an ego stereo camera instead 🧵 (with @kushtimusprime)
3
10
90
6,894
What can Astra do when given a humanoid embodiment?
We built HomeBody to find out. Controlled by GPT Astra, it carries out long-horizon tasks in a previously unseen kitchen—from tidying up across the room to retrieving remembered objects from ambiguous requests—without environment-specific training data or additional policy learning.
Here's how we did it 👀: tml.stanford.edu/homebody/
107
327
2,413
670,391
Happy to announce the full ABC release! We’re also excited that ABC was accepted to CoRL 2026!
Check out our website for code, 400+ hours of sim data on 24 tasks, and 5,850 labeled policy-evaluation episodes.
@arthurallshire @Cinnabar233 @ritvik_singh9 @redstone_hong @davidrmcall
6
33
259
77,148
Adam Rashid retweeted
Recently, there have been a lot of impressive demos of AI agents, like Claude, controlling robots.
I wrote a short blog post with my thoughts on the advent of these "robot-use agents."
web.mit.edu/phillipi/www/wri…
I think it's an important change in the trajectory of robotics!
44
113
757
206,867
Adam Rashid retweeted
There's a really simple recipe behind much of modern AI:
1. Pre-train on massive proxy data
2. Post-train on a small amount of the real thing
Here we apply this recipe to robot hands:
1. Pre-train on lots of sim
2. Post-train on a bit of real
Very simple and works!
Excited to share SPD: simulation pre-training for dexterity. We pre-trained a policy in simulation and fine-tuned with less than 2 hours of real data (with @sarthakkamat)
8
30
370
43,810
Adam Rashid retweeted
here’s an uncut, real-time video of our policy stacking and destacking a cup tower
Excited to share SPD: simulation pre-training for dexterity. We pre-trained a policy in simulation and fine-tuned with less than 2 hours of real data (with @sarthakkamat)
3
7
80
9,346
Adam Rashid retweeted
Great collaboration with @sarthakkamat and @adamrasb!
Excited to share SPD: simulation pre-training for dexterity. We pre-trained a policy in simulation and fine-tuned with less than 2 hours of real data (with @sarthakkamat)
2
5
621
Excited to share SPD: simulation pre-training for dexterity. We pre-trained a policy in simulation and fine-tuned with less than 2 hours of real data (with @sarthakkamat)
17
71
539
139,612
You can read more about it and check out the full videos here at spd.bot
1
24
4,510
Adam Rashid retweeted
My favorite part: ABC is a reproducible robotics stack. The website demos are not just cool videos: the underlying data, recipes, and infra are open, so people can actually inspect and build on them.
Also excited about the >400h of sim teleop data coming soon. This should help the community iterate faster and study questions like sim-real co-training. One encouraging signal from our experiments: even with only ~2% simulated data mixed into co-training, we observed strong sim-real performance correlations.
Excited to see what the community builds on top!
Introducing ABC: open data, training, and infrastructure for robotics.
We release the largest teleop dataset to date, and extensively investigate design decisions, pretraining, and post-training techniques.
@arthurallshire @Cinnabar233 @adamrasb @redstone_hong @davidrmcall
9
80
15,760
Adam Rashid retweeted
also check out our code here github.com/amazon-far/abc
we haven't released everything yet but there's enough there that you can train a policy, eval it in sim, and view the rollouts live in viser!
2
5
54
15,781
Adam Rashid retweeted
So excited for this to finally be out:
A fully open-source data & stack released so we can learn the ABCs of bimanual manipulation together!
This was a huge team effort, and many long days and nights of work. Congrats to the amazing team, I'm very proud of you!
Introducing ABC: open data, training, and infrastructure for robotics.
We release the largest teleop dataset to date, and extensively investigate design decisions, pretraining, and post-training techniques.
@arthurallshire @Cinnabar233 @adamrasb @redstone_hong @davidrmcall
4
21
182
30,820
Super excited to be able to release this project we've been working on. The largest open robotics dataset to date and policies capable of dexterous manipulation!
Introducing ABC: open data, training, and infrastructure for robotics.
We release the largest teleop dataset to date, and extensively investigate design decisions, pretraining, and post-training techniques.
@arthurallshire @Cinnabar233 @adamrasb @redstone_hong @davidrmcall
1
9
54
9,977