---
title: "PDAR-RSITR: A Progressive Decoupling-Aggregation-Refinement Framework for Remote Sensing Image-Text Retrieval"
authors:
  - "Shuhuai Wang"
  - "Songwei Pei"
  - "Bingfeng Liu"
  - "Duo Chai"
  - "Yuanzhou Huang"
  - "Jia Liu"
  - "Qian Li"
  - "Shangguang Wang"
doi: "10.24963/ijcai.2026/661"
url: "https://ijcai.org/proceedings/2026/661"
pdf: "https://ijcai.org/proceedings/2026/0661.pdf"
venue: "Proceedings of the Thirty-Fifth International Joint Conference on Artificial Intelligence"
year: 2026
pages: "5937-5945"
track: "Main Track"
keywords:
  - "Natural Language Processing: Information retrieval and text mining"
---

# PDAR-RSITR: A Progressive Decoupling-Aggregation-Refinement Framework for Remote Sensing Image-Text Retrieval

Shuhuai Wang, Songwei Pei, Bingfeng Liu, Duo Chai, Yuanzhou Huang, Jia Liu, Qian Li, Shangguang Wang

*Proceedings of the Thirty-Fifth International Joint Conference on Artificial Intelligence*, pages 5937-5945. DOI: [10.24963/ijcai.2026/661](https://doi.org/10.24963/ijcai.2026/661)

## Abstract

Remote Sensing Image-Text Retrieval (RSITR) aims to achieve precise retrieval between remote sensing images and textual descriptions. However, existing methods neglect the multi-dimensional cognitive attributes inherent in remote sensing data and struggle to handle them simultaneously, leading to suboptimal retrieval performance. In this paper, we propose a novel Progressive Decoupling-Aggregation-Refinement framework for RSITR (PDAR-RSITR) to comprehensively capture multi-dimensional cognitive attributes. Specifically, we adapt Multi-dimensional Cognitive Decoupling to learn features across cognitive attributes of different dimensions via multi-process clustering. Subsequently, we utilize Salient Cognitive Aggregation to select and aggregate the most salient attributes via dynamic routing. Furthermore, we propose Expert Collaborative Refinement to enhance critical cross-modal relationships via three complementary expert perspectives. Extensive experimental results demonstrate that PDAR-RSITR significantly outperforms existing state-of-the-art methods across multiple metrics.