Saved in:
| Main Authors: | , , , , , , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2025
|
| Subjects: | |
| Online Access: | https://arxiv.org/abs/2505.20740 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866909010788089856 |
|---|---|
| author | Zhao, Xiangyu Xu, Wanghan Liu, Bo Zhou, Yuhao Ling, Fenghua Fei, Ben Yue, Xiaoyu Bai, Lei Zhang, Wenlong Wu, Xiao-Ming |
| author_facet | Zhao, Xiangyu Xu, Wanghan Liu, Bo Zhou, Yuhao Ling, Fenghua Fei, Ben Yue, Xiaoyu Bai, Lei Zhang, Wenlong Wu, Xiao-Ming |
| contents | The rapid advancement of multimodal large language models (MLLMs) offers new opportunities for complex scientific challenges, yet their application in earth science-especially at the graduate level-remains underexplored due to a lack of benchmarks reflecting the depth and complexity of geoscientific reasoning. Existing datasets often rely on synthetic data or simple figure-caption pairs, failing to capture the nuanced reasoning required for real-world applications. To address this, we introduce MSEarth, a multimodal scientific dataset and benchmark curated from high-quality, open-access publications. Covering the five major spheres of Earth science-atmosphere, cryosphere, hydrosphere, lithosphere, and biosphere-MSEarth features over 289K figures with refined captions enriched by contextual discussions and reasoning from the original papers. The benchmark supports tasks such as scientific figure captioning, multiple choice questions, and open-ended reasoning, providing a scalable, high-fidelity resource for developing and evaluating MLLMs in scientific reasoning. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2505_20740 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | MSEarth: A Multimodal Benchmark for Earth Science Phenomenon Discovery with MLLMs Zhao, Xiangyu Xu, Wanghan Liu, Bo Zhou, Yuhao Ling, Fenghua Fei, Ben Yue, Xiaoyu Bai, Lei Zhang, Wenlong Wu, Xiao-Ming Artificial Intelligence The rapid advancement of multimodal large language models (MLLMs) offers new opportunities for complex scientific challenges, yet their application in earth science-especially at the graduate level-remains underexplored due to a lack of benchmarks reflecting the depth and complexity of geoscientific reasoning. Existing datasets often rely on synthetic data or simple figure-caption pairs, failing to capture the nuanced reasoning required for real-world applications. To address this, we introduce MSEarth, a multimodal scientific dataset and benchmark curated from high-quality, open-access publications. Covering the five major spheres of Earth science-atmosphere, cryosphere, hydrosphere, lithosphere, and biosphere-MSEarth features over 289K figures with refined captions enriched by contextual discussions and reasoning from the original papers. The benchmark supports tasks such as scientific figure captioning, multiple choice questions, and open-ended reasoning, providing a scalable, high-fidelity resource for developing and evaluating MLLMs in scientific reasoning. |
| title | MSEarth: A Multimodal Benchmark for Earth Science Phenomenon Discovery with MLLMs |
| topic | Artificial Intelligence |
| url | https://arxiv.org/abs/2505.20740 |