Referring expression understanding in remote sensing poses unique challenges, as it requires reasoning over complex object-context relationships. While supervised fine-tuning (SFT) on multimodal large language models achieves strong performance with massive labeled datasets, they struggle in data-scarce scenarios, leading to poor generalization. To address this limitation, we propose Geo-R1, a reasoning-centric reinforcement fine-tuning (RFT) paradigm for few-shot geospatial referring. Geo-R1 enforces the model to first generate explicit, interpretable reasoning chains that decompose referring expressions, and then leverage these rationales to localize target objects. This "reason first, then act" process enables the model to make more effective use of limited annotations, enhances generalization, and provides interpretability. We validate Geo-R1 on three carefully designed few-shot geospatial referring benchmarks, where our model consistently and substantially outperforms SFT baselines. It also demonstrates strong cross-dataset generalization, highlighting its robustness. Code and data will be released at: https://github.com/Geo-R1/geo-r1.
@article{arxiv.2509.21976,
title = {Geo-R1: Improving Few-Shot Geospatial Referring Expression Understanding with Reinforcement Fine-Tuning},
author = {Zilun Zhang and Zian Guan and Tiancheng Zhao and Haozhan Shen and Tianyu Li and Yuxiang Cai and Zhonggen Su and Zhaojun Liu and Jianwei Yin and Xiang Li},
journal= {arXiv preprint arXiv:2509.21976},
year = {2026}
}