ImagineNav: Prompting Vision-Language Models as Embodied Navigator through Scene Imagination

Open in new window