Multi-Agent Egocentric World Model with Fine-Grained Embodied Interaction Paper • 2610.12299 • Published 5 days ago • 58
RobotWorld: Benchmarking Multimodal Agents for Robot Use Across Diverse Tasks and Embodiments Paper • 2610.10409 • Published 6 days ago • 39
GRACE: Generation-aware latent compression for efficient video generation Paper • 2610.10524 • Published 6 days ago • 88
World Observer: Joint Actor-Observer Generation for Persistent World Modeling Paper • 2610.02162 • Published 12 days ago • 90