Tag
This paper proposes that large-scale text-to-video generation can serve as a powerful pre-training paradigm for computer vision, introducing GenCeption which achieves state-of-the-art performance across diverse vision tasks with high data efficiency and emergent generalization to unseen domains.
RoboDojo is a unified sim-and-real benchmark for comprehensive evaluation of generalist robot manipulation policies, featuring 42 simulation tasks and 18 real-world tasks across multiple evaluation dimensions.
The article discusses how AI automation will make most high-income skills obsolete. It proposes four core skills needed in the future: initiative, taste and unique perspective, judgment, and deep generalism. It emphasizes that humans need to actively adapt to become 'sovereign individuals'.
Vesta is a unified embodied generalist model that integrates localization, spatial reasoning, navigation, and long-horizon planning into a single foundation model, outperforming specialized models by over 20% on benchmarks and over 35% on real-world robotic tasks.