AI Paper Atlas

Research map / Vision-language and multimodal / Multimodal LLMs

GEOBench-VLM: Benchmarking Vision-Language Models for Geospatial Tasks

Muhammad Danish, Muhammad Akhtar Munir, Syed Roshaan Ali Shah, Kartik Kuckreja, Fahad Shahbaz Khan, Paolo Fraccaro, Alexandre Lacoste, Salman Khan

ICCV 2025 · Vision-language and multimodal › Multimodal LLMs · approach: Spatial reasoning in VLMs · 49 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
use geospatial-specific EarthDial instead of GPT-4oVLM on geospatial MCQ · Scene Understandingaccuracy: 0.7114 -> 0.7705Improved
use geospatial-specific EarthDial instead of GPT-4oVLM on geospatial MCQ · Object Classificationaccuracy: 0.5863 -> 0.4039Hurt
use geospatial-specific EarthDial instead of GPT-4oVLM on geospatial MCQ · Event Detectionaccuracy: 0.4726 -> 0.5418Improved
use geospatial-specific GeoChat instead of generic LLaVA-1.5VLM · Image CaptioningBERTScore: 0.6346 -> 0.4395Hurt
use generic LLaVA-OneVision instead of domain models for countingVLM · Countingaccuracy: 0.3626 -> 0.4377Improved
use SPHINX instead of GPT-4o for referring detectionVLM · Referring Expression Detectionprecision at IoU 0.5: 0.0087 -> 0.3408Improved
use multi-temporal instead of single-temporal input for land use classificationGPT-4o, EarthDial, Qwen2-VL, LLaVA-OneVision · Land Use ClassificationaccuracyImproved
use multi-temporal instead of single-temporal input for crop classificationVLMs · Crop ClassificationaccuracyHurt
use multi-temporal input for disaster classificationVLMs · Disaster ClassificationaccuracyMixed
use multispectral instead of RGB inputsGPT-4o and Qwen2-VL · Crop Type ClassificationaccuracyHurt

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy