LATIDIA · Robótica

GemNav: Navegación de robots visuales de tokens discretos utilizando un modelo de lenguaje grande multimodal

arXiv: 2607.06882v3Announce Type: replace Abstract: Las políticas de navegación visual basadas en grandes modelos preentrenados se basan cada vez más en grandes conjuntos de datos de encarnaciones cruzadas y mecanismos expresivos de generación de acciones. En este pa

La noticia

arXiv:2607.06882v3 Announce Type: replace Abstract: Visual navigation policies built on large pretrained models increasingly rely on large cross-embodiment datasets and expressive action-generation mechanisms. In this paper, we ask whether a pretrained Multimodal Large Language Model (MLLM) can be adapted efficiently for robot navigation by rethinking the representation and construction of navigation actions. We introduce GemNav, a visual navigation policy that represents each robot-local waypoint coordinate as a categorical distribution over discrete metric value bins and soft-decodes it into a continuous coordinate for execution. This probabilistic representation retains likelihood information over alternative coordinate values while providing a compact action space suitable for robot navigation. We further construct training trajectories using metric-distance