Textual Supervision Enhances Geospatial Representations in Vision-Language Models
Universidade Federal de Minas Gerais · Max Planck Institute for Security and Privacy · Institute for Basic Science · Changwon National University, South Korea · Universidade Federal de Minas Gerais, Universidade Federal de Minas Gerais · MPI-SP & KAIST
PDF 由论文原始站点提供，PaperCompass 不保存论文文件。
摘要
Geospatial understanding is a critical yet underexplored dimension in the development of machine learning systems for tasks such as image geolocation and spatial reasoning. In this work, we analyze the geospatial representations acquired by three model families: vision-only architectures (e.g., ViT), vision-language models (e.g., CLIP), and large-scale multimodal foundation models (e.g., LLaVA, Qwen, and Gemma). By evaluating across image clusters, including people, landmarks, and everyday objects, grouped based on the degree of localizability, we reveal systematic gaps in spatial accuracy and show that textual supervision enhances the learning of geospatial representations. Our findings suggest the role of language as an effective complementary modality for encoding spatial context and multimodal learning as a key direction for advancing geospatial AI.