cs.CVJan 23, 2026

GRASP: Guided Region-Aware Sparse Prompting for Adapting MLLMs to Remote Sensing

Authors: Qigan Sun, Chaoning Zhang, Jianwei Zhang, Xudong Wang, Jiehui Xie, Pengcheng Zheng, Haoyu Wang, Sungyoung Lee, 

Organizations: School of Computing, Kyung Hee University, Yongin-si, South Korea · School of Computer Science and Engineering, University of Electronic Science and Technology of China, Chengdu, China · School of Information and Software Engineering, University of Electronic Science and Technology of China, Chengdu, China · College of Computer Science and Information Engineering, Harbin Normal University, Harbin, 150025, China · 

Abstract

In recent years, Multimodal Large Language Models (MLLMs) have made significant progress in visual question answering tasks. However, directly applying existing fine-tuning methods to remote sensing (RS) images often leads to issues such as overfitting on background noise or neglecting target details. This is primarily due to the large-scale variations, sparse target distributions, and complex regional semantic features inherent in RS images. These challenges limit the effectiveness of MLLMs in RS tasks. To address these challenges, we propose a parameter-efficient fine-tuning (PEFT) strategy called Guided Region-Aware Sparse Prompting (GRASP). GRASP introduces spatially structured soft prompts associated with spatial blocks extracted from a frozen visual token grid. Through a question-guided sparse fusion mechanism, GRASP dynamically aggregates task-specific context into a compact global prompt, enabling the model to focus on relevant regions while filtering out background noise. Extensive experiments on multiple RSVQA benchmarks show that GRASP achieves competitive performance compared to existing fine-tuning and prompt-based methods while maintaining high parameter efficiency.

Figures & tables

Explore similar work

CardsList
  1. Multimodal Large Language Models for Remote Sensing Image Understanding: Domain-Specific or General-Purpose?

    Jul 22, 2026Qiwei Ma, Chunping Qiu, Xinjun Cheng +5Remote Sensing Image UnderstandingRemote Sensing Visual Question Answering

  2. RS-Agent: Automating Remote Sensing Tasks through Intelligent Agent

    Jun 11, 2024Wenjia Xu, Zijian Yu, Boyang Mu +3Multimodal RAGMultimodal Large Language Models

  3. Evaluating Remote Sensing Image Captions Beyond Metric Biases

    Apr 22, 2026Ziyun Chen, Fan Liu, Liang Yao +3VLM EvaluationRemote Sensing Image Understanding