Fetching the paper…

OpenFMNav: Towards Open-Set Zero-Shot Object Navigation via Vision-Language Foundation Models · Around