Dyna Robotics放出了个叫Dyna-2的世界动作模型，用100万小时人类视频训的，能预测未来再行动，还发现了数据量越大越强的缩放规律，值得看看。
Dyna Robotics发布世界动作模型Dyna-2，在100万小时人类第一视角视频上预训练。该模型能联合预测未来视频和未来动作，在行动前推理结果。研究首次发现，从1000到100万小时的人类数据上，世界动作模型呈现四个数量级的缩放定律。该缩放定律还能迁移到从未见过的机器人数据上，且视频数据与世界建模对跨具身缩放迁移至关重要。
This is huge! Interesting new scaling laws discovered. Dyna-2 is a world-action model trained on o...
This is huge! Interesting new scaling laws discovered. Dyna-2 is a world-action model trained on over 1M hours of egocentric human video. It jointly predicts future video and future actions, so it reasons about outcomes before acting. Dyna Robotics @DynaRobotics 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 to 1,000,000 hours, • this 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 🧵 Your browser does not support the video tag. 🔗 View on Twitter 🔗 View Quoted Tweet 💬 3 🔄 3 ❤️ 7 👀 2385 📊 4 ⚡