Dyna-2：具备小时级自主性的通用物理智能体
原标题：Dyna-2: General Purpose Physical Agents with Hour-Long Autonomy — Jason Ma, Dyna Robotics
AI 摘要
Dyna Robotics 联合创始人兼首席科学家 Jason Ma 在 Latent Space 播客中介绍了 Dyna-2 世界动作模型及其后续物理机器人智能体 Dyna-2.1。Dyna-2 基于超过一百万小时人类视频预训练，结合未来视频预测与机器人动作生成，并声称在人类数据上呈现跨四个数量级的新缩放规律。Dyna-2.1 在餐厅折餐巾任务中实现 24 小时连续运行、99.4% 成功率，并已在鼎泰丰等真实客户场景中部署。
正文节选
Watch along AIE NYC and join us at AIE CODE in SF next month! This episode includes a live demo of the Taku robot that is best enjoyed on YouTube. Last month, Dyna made waves by announcing Dyna-2, a world-action model pre-trained on one million hours of human video, claiming several new scaling laws on human data across four orders of magnitude. They have now followed up the model with a physical robot agent, Dyna-2.1, that has to be seen to be believed: In his talk at AIE this year, Jason talke