Build AI retweeted
I’ve left DeepMind to lead research at Build AI
Our first bet: scale monocular vision. We haven’t even scratched the surface :D
Our first bet: scale monocular vision. We haven’t even scratched the surface :D
55
35
1,167
134,408
pre-trained on Egocentric-1M data!
Today we are introducing Dyna-2, a world-action model pre-trained on one million hours of human video. At this scale, for the first time, we discovered several new scaling laws:
• world-action models exhibit scaling law on human data across four orders of magnitude, from 1000 his human data scaling law implied a scaling law on never seen robot data,
• both data and objective matter; world modeling and scaling on video data are essential for cross-embodiment scaling transfer to emerge
🧵
4
11
89
33,310
