AI RESEARCH
MMLANDMARKS: a Cross-View Instance-Level Benchmark for Geo-Spatial Understanding
arXiv CS.CV
•
ArXi:2512.17492v2 Announce Type: replace Geo-spatial analysis of our world benefits from a multimodal approach, as every single geographic location can be described in numerous ways (images from various viewpoints, textual descriptions, geographic coordinates, etc.). Current benchmarks have limited coverage across modalities, leading to specialized models that perform well in their respective domains, but do not fully take advantage of other geo-spatial modalities. We