Loading video...
Video Failed to Load
The robot is learning several novel tasks instantly, after just ONE demonstration each... Instant Policy makes it possible: no extra training, no weight updates, just pure in-context learning. It just got accepted at ICLR 2025, and it’s changing how robots learn. With just a single demo, a robot can... show more
46,285 views • 1 year ago •via X (Twitter)
6 Comments

Mark Ruby1 year ago
I'm waiting for the robot to become sentient, then impatient, and slap him.

Luke Liem1 year ago
This combines imitation learning and graph representation. Apparently a transformer is also involved in the architecture - to learn context?

The Rundown AI1 year ago
If you're not learning AI in 2025, you're falling behind. Join 1,000,000+ early adopters reading and learn AI in just 5 minutes a day (for free).

Dayder1 year ago
Dum-E from Iron Man is so close😁 Amazing progress.

Jamal Jalal1 year ago
يتعلم ذراع الروبوت من توجيه واحد فقط ولايتأثر بتغير البيئه , اعتقد انهم اقتربوا من صناعة هواتف الموبايل بالروبوتات ,حيث كان الهدف قبل سنوات ان ينفذ الروبوت ما يراه اي يتم تعليمه كما يتعلم البشر من النظر,كان ينبغي على الدول العربيه السير بهذا الطريق وهو مربح للغايه ايضا

Renato Pudo1 year ago
🤯

