Newest

LightNav-0 Proposes Generalist Embodied Navigation Model Leveraging VLM Spatial Intelligence

Robots can now borrow a vision-language model's sense of space to navigate different bodies and environments without task-specific retraining.