GeoAgent: Evaluating VLM Geolocalization Through Embodied Navigation
This work introduces GeoAgent, an agentic environment-based benchmark that requires agents to navigate Street View environments to refine their geolocalization through sequential reasoning, and establishes the challenges of embodied navigation and geospatial reasoning.
Arka Mukherjee, Soham Roy, Kartikeya Trivedi et al.
· 0 citations