World-Models

NVIDIA GEAR Lab has released DreamDojo, an open-source world model pretrained on a massive 44,000-hour dataset of human egocentric videos. By using "latent actions" to bridge the gap between human and robot movement, the model achieves zero-shot generalization and real-time controllability for teleoperation and planning.

Waymo has unveiled its new World Model, powered by Google DeepMind’s Genie 3, to simulate rare "long-tail" driving scenarios and edge cases, signaling a major shift toward generative world models in the race for physical AI.

In a deep dive on the Robo Papers podcast, 1X Director of Evaluations Daniel Ho explains how "imagination" via video generation is allowing humanoids to perform zero-shot tasks with minimal robot-specific training data.

NVIDIA GEAR Lab has unveiled DreamZero, a 14-billion parameter World Action Model (WAM) that uses video diffusion to grant robots physical "imagination," enabling zero-shot task completion and rapid adaptation across different robotic embodiments.

In a new interview, Google DeepMind's Carolina Parada discusses the 'agentic' two-part brain of Gemini Robotics 1.5, its 'surprising' ability to transfer skills between different robots, and why she believes the home will be 'one of the last frontiers' for the technology.