Unified Sequence Modeling for Remote Sensing: A Parameter-Efficient Foundation Model via Prompt-Driven Granularity Alignment

Liu, Yang; Luo, Weixing; Qi, Huaizhou; Jia, Suisui; Guo, Yongjing

doi:10.24963/ijcai.2026/162

Unified Sequence Modeling for Remote Sensing: A Parameter-Efficient Foundation Model via Prompt-Driven Granularity Alignment

Yang Liu, Weixing Luo, Huaizhou Qi, Suisui Jia, Yongjing Guo

Download BibTeX Cite

Proceedings of the Thirty-Fifth International Joint Conference on Artificial Intelligence. Main Track. Pages 1451-1459.

https://doi.org/10.24963/ijcai.2026/162

Keywords

Computer Vision: Machine learning for vision
Machine Learning: Classification
Computer Vision: Recognition (object detection, categorization)

Abstract

Current remote sensing (RS) perception systems suffer from task heterogeneity, necessitating distinct architectures for classification, localization, and reasoning. While vision--language models (VLMs) offer a route toward unification, their computational cost can hinder deployment. In this work, we propose RS-Florence, a compact unified model that addresses these tasks through a Prompt-Driven Sequence-to-Sequence framework. Unlike traditional approaches that segregate semantic understanding and geometric localization, RS-Florence maps images and task-specific prompts into a unified sequence of natural language and discrete geometric tokens. This formulation helps bridge high-level semantics and low-level pixel perception. Experiments across 4 task families and 8 benchmarks show that our 0.23B model remains competitive with task-specific specialists. These results also suggest that multitask joint training improves performance on several benchmarks over single-task fine-tuning, indicating that a shared prompt-driven interface can serve both language and geometry tasks.

Yang Liu, Weixing Luo, Huaizhou Qi, Suisui Jia, and Yongjing Guo. 2026. Unified Sequence Modeling for Remote Sensing: A Parameter-Efficient Foundation Model via Prompt-Driven Granularity Alignment. In Proceedings of the Thirty-Fifth International Joint Conference on Artificial Intelligence, IJCAI-26. IJCAI, 1451–1459. https://doi.org/10.24963/ijcai.2026/162

@inproceedings{ijcai2026p162,
  title     = {Unified Sequence Modeling for Remote Sensing: A Parameter-Efficient Foundation Model via Prompt-Driven Granularity Alignment},
  author    = {Liu, Yang and Luo, Weixing and Qi, Huaizhou and Jia, Suisui and Guo, Yongjing},
  booktitle = {Proceedings of the Thirty-Fifth International Joint Conference on
               Artificial Intelligence, {IJCAI-26}},
  publisher = {International Joint Conferences on Artificial Intelligence Organization},
  editor    = {Diego Calvanese},
  pages     = {1451--1459},
  year      = {2026},
  month     = {8},
  note      = {Main Track},
  doi       = {10.24963/ijcai.2026/162},
  url       = {https://doi.org/10.24963/ijcai.2026/162},
}

TY  - CPAPER
TI  - Unified Sequence Modeling for Remote Sensing: A Parameter-Efficient Foundation Model via Prompt-Driven Granularity Alignment
AU  - Liu, Yang
AU  - Luo, Weixing
AU  - Qi, Huaizhou
AU  - Jia, Suisui
AU  - Guo, Yongjing
T2  - Proceedings of the Thirty-Fifth International Joint Conference on Artificial Intelligence, IJCAI-26
PB  - International Joint Conferences on Artificial Intelligence Organization
PY  - 2026
SP  - 1451
EP  - 1459
DO  - 10.24963/ijcai.2026/162
UR  - https://doi.org/10.24963/ijcai.2026/162
ER  -