Synapse
⌘+K
Synapse
PulseExploreClubsResearchersJournals
Instagram
HomeClubsExplore
July 23, 2025Open Access Research Journal of Science and Technology

GeoLLMs in action: A systematic review of multimodal models for satellite image captioning and geospatial understanding

View Full Paper
Ask AI
Bookmark
Share

Authors

JEJohn Adeyemi EyinadeAAAdebisi Joseph Ademusire

Discussion

Loading...

Member takes

Overview

This review examines large language models' impact on satellite imagery captioning, indicating gaps in geographic representation and evaluation metrics.

Key Points

  • GeoLLMs combine satellite imagery and large language models to enhance geospatial understanding.
  • A review of 42 studies highlights the emergence of architectural patterns in multimodal models for Earth observation.
  • Findings reveal challenges in geographic generalization and representation of non-Western locales within geospatial AI.
  • Research suggests new directions, like geometry-aware embeddings and multilingual fine-tuning, to advance GeoLLMs.

Cite This Study

Eyinade et al. (2025) studied this question.

synapsesocial.com/papers/689a0621e6551bb0af8cdc50https://doi.org/10.53022/oarjst.2025.14.2.0093
View Full Paper
Ask AI
Bookmark
Share

Also Consider

Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context:

  1. 1A review of large language models in geomatics: integrating multimodal data, addressing challenges, and exploring synergies2026 · 4 citations
  2. 2Mapping with Words: Integrating Large Language Models into Geospatial Practice2026
  3. 3Large language models for knowledge-centric scientific intelligence: methods, challenges, and lessons from geoscience2026
  4. 4A Comprehensive Evaluation of Multimodal Large Language Models in Hydrological Applications2024 · 5 citations
  5. 5GeoKGM: A Multimodal Large Language Model for Zero-Shot Knowledge Graph Completion in Geospatial Databases2025 · 5 citations