arXiv cs.AI / cs.LG / cs.CL·18d agoDeCAL: Towards Physically-Grounded Dexterous Vision-Language-Action Models via Contact-Aware Latent Co-Imagination#dexterous-manipulation#mixture-of-transformers#multimodalAI research1
arXiv cs.AI / cs.LG / cs.CL·22d agoRoboSPA: Can VLA Models Go Beyond Simple Scenes and Short-Horizon Tasks?#benchmark#embodied-ai#long-horizon-planningAI research