GeoZero: Incentivizing Reasoning from Scratch on Geospatial Scenes
Fuente:
arXiv
Saved in:
| Main Authors: | , , , , , , , , , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2025
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866911510667722752 |
|---|---|
| author | Wang, Di Liu, Shunyu Jiang, Wentao Wang, Fengxiang Liu, Yi Qin, Xiaolei Luo, Zhiming Zhou, Chaoyang Guo, Haonan Zhang, Jing Du, Bo Tao, Dacheng Zhang, Liangpei |
| author_facet | Wang, Di Liu, Shunyu Jiang, Wentao Wang, Fengxiang Liu, Yi Qin, Xiaolei Luo, Zhiming Zhou, Chaoyang Guo, Haonan Zhang, Jing Du, Bo Tao, Dacheng Zhang, Liangpei |
| contents | Multimodal large language models (MLLMs) have undergone rapid development in advancing geospatial scene understanding. Recent studies have sought to enhance the reasoning capabilities of remote sensing MLLMs, typically through cold-start training with elaborately curated chain-of-thought (CoT) data. However, this approach not only incurs substantial annotation costs but also introduces human biases that may limit the diversity of model reasoning. To address these challenges, we propose GeoZero, a framework that enables MLLMs to perform geospatial reasoning without any predefined CoT supervision. Specifically, we construct two datasets, GeoZero-Instruct and GeoZero-Hard. GeoZero-Instruct allows the model to acquire preliminary geospatial knowledge through supervised fine-tuning, while GeoZero-Hard stimulates deep reasoning during the subsequent reinforcement learning stage. Furthermore, we introduce Answer-Anchored Group Relative Policy Optimization (A$^2$GRPO), where the reasoning process is regularized by the model's own answers, encouraging diverse yet accurate thinking. Extensive experiments on multiple remote sensing vision-language benchmarks demonstrate that GeoZero not only surpasses existing state-of-the-art methods but also fosters universal emergent reasoning capabilities across diverse geospatial tasks. Code, data, and models are available at https://github.com/MiliLab/GeoZero. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2511_22645 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | GeoZero: Incentivizing Reasoning from Scratch on Geospatial Scenes Wang, Di Liu, Shunyu Jiang, Wentao Wang, Fengxiang Liu, Yi Qin, Xiaolei Luo, Zhiming Zhou, Chaoyang Guo, Haonan Zhang, Jing Du, Bo Tao, Dacheng Zhang, Liangpei Computer Vision and Pattern Recognition Multimodal large language models (MLLMs) have undergone rapid development in advancing geospatial scene understanding. Recent studies have sought to enhance the reasoning capabilities of remote sensing MLLMs, typically through cold-start training with elaborately curated chain-of-thought (CoT) data. However, this approach not only incurs substantial annotation costs but also introduces human biases that may limit the diversity of model reasoning. To address these challenges, we propose GeoZero, a framework that enables MLLMs to perform geospatial reasoning without any predefined CoT supervision. Specifically, we construct two datasets, GeoZero-Instruct and GeoZero-Hard. GeoZero-Instruct allows the model to acquire preliminary geospatial knowledge through supervised fine-tuning, while GeoZero-Hard stimulates deep reasoning during the subsequent reinforcement learning stage. Furthermore, we introduce Answer-Anchored Group Relative Policy Optimization (A$^2$GRPO), where the reasoning process is regularized by the model's own answers, encouraging diverse yet accurate thinking. Extensive experiments on multiple remote sensing vision-language benchmarks demonstrate that GeoZero not only surpasses existing state-of-the-art methods but also fosters universal emergent reasoning capabilities across diverse geospatial tasks. Code, data, and models are available at https://github.com/MiliLab/GeoZero. |
| title | GeoZero: Incentivizing Reasoning from Scratch on Geospatial Scenes |
| topic | Computer Vision and Pattern Recognition |
| url | https://arxiv.org/abs/2511.22645 |