AI RESEARCH
LLaVA-LE: Large Language-and-Vision Assistant for Lunar Exploration
arXiv CS.CV
•
ArXi:2603.24696v1 Announce Type: new Recent advances in multimodal vision-language models (VLMs) have enabled joint reasoning over visual and textual information, yet their application to planetary science remains largely unexplored. A key hindrance is the absence of large-scale datasets that pair real planetary imagery with detailed scientific descriptions. In this work, we