GeoVLM-R1: Reinforcement Fine-Tuning for Improved Remote Sensing Reasoning

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Fiaz, Mustansar, Debary, Hiyam, Fraccaro, Paolo, Paudel, Danda, Van Gool, Luc, Khan, Fahad, Khan, Salman
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866917010611372032
author Fiaz, Mustansar
Debary, Hiyam
Fraccaro, Paolo
Paudel, Danda
Van Gool, Luc
Khan, Fahad
Khan, Salman
author_facet Fiaz, Mustansar
Debary, Hiyam
Fraccaro, Paolo
Paudel, Danda
Van Gool, Luc
Khan, Fahad
Khan, Salman
contents Recent advances in reinforcement learning (RL) have delivered strong reasoning capabilities in natural image domains, yet their potential for Earth Observation (EO) remains largely unexplored. EO tasks introduce unique challenges, spanning referred object detection, image or region captioning, change detection, grounding, and temporal analysis, that demand task aware reasoning. We propose a novel post training framework that incorporates task aware rewards to enable effective adaptation of reasoning based RL models to diverse EO tasks. This training strategy enhances reasoning capabilities for remote sensing images, stabilizes optimization, and improves robustness. Extensive experiments across multiple EO benchmarks show consistent performance gains over state of the art generic and specialized vision language models. Code and models will be released publicly at https://mustansarfiaz.github.io/GeoVLM-R1/ .
format Preprint
id arxiv_https___arxiv_org_abs_2509_25026
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle GeoVLM-R1: Reinforcement Fine-Tuning for Improved Remote Sensing Reasoning
Fiaz, Mustansar
Debary, Hiyam
Fraccaro, Paolo
Paudel, Danda
Van Gool, Luc
Khan, Fahad
Khan, Salman
Computer Vision and Pattern Recognition
Recent advances in reinforcement learning (RL) have delivered strong reasoning capabilities in natural image domains, yet their potential for Earth Observation (EO) remains largely unexplored. EO tasks introduce unique challenges, spanning referred object detection, image or region captioning, change detection, grounding, and temporal analysis, that demand task aware reasoning. We propose a novel post training framework that incorporates task aware rewards to enable effective adaptation of reasoning based RL models to diverse EO tasks. This training strategy enhances reasoning capabilities for remote sensing images, stabilizes optimization, and improves robustness. Extensive experiments across multiple EO benchmarks show consistent performance gains over state of the art generic and specialized vision language models. Code and models will be released publicly at https://mustansarfiaz.github.io/GeoVLM-R1/ .
title GeoVLM-R1: Reinforcement Fine-Tuning for Improved Remote Sensing Reasoning
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2509.25026