SkyMoE: A Vision-Language Foundation Model for Enhancing Geospatial Interpretation with Mixture of Experts
Proceedings of the AAAI Conference on Artificial Intelligence · 2026-03-14 · 40 (9) · 7168-7178
DOI: 10.1609/aaai.v40i9.37653 ↗At a glance
Journal article in Proceedings of the AAAI Conference on Artificial Intelligence, published online on 2026-03-14. Written by Jiaqi Liu and 8 co-authors. Cited 8 times so far (Crossref).
Abstract (original language)
The emergence of large vision-language models (VLMs) has significantly enhanced the efficiency and flexibility of geospatial interpretation. However, general-purpose VLMs remain suboptimal for remote sensing (RS) tasks. Existing geospatial VLMs typically adopt a unified modeling strategy and struggle to differentiate between task types and interpretation granularities, limiting their ability to balance local detail perception and global contextual understanding. In this paper, we present SkyMoE, a Mixture-of-Experts (MoE) vision-language model tailored for multimodal, multi-task RS interpretation. SkyMoE employs an adaptive router that generates task- and granularity-aware routing instructions, enabling specialized large language model experts to handle diverse sub-tasks. To further promote expert decoupling and granularity sensitivity, we introduce a context-disentangled augmentation strategy that creates contrastive pairs between local and global features, guiding experts toward level-specific representation learning. We also construct MGRS-Bench, a comprehensive benchmark covering multiple RS interpretation tasks and granularity levels, to evaluate generalization in complex scenarios. Extensive experiments on 21 public datasets demonstrate that SkyMoE achieves state-of-the-art performance across tasks, validating its adaptability, scalability, and superior multi-granularity understanding in remote sensing.
Cite
APA
Liu, J., Fu, R., Sun, L., Liu, H., Yang, X., Zhang, W., Na, X., Duan, Z., & Yang, B. (2026). SkyMoE: A Vision-Language Foundation Model for Enhancing Geospatial Interpretation with Mixture of Experts. Proceedings of the AAAI Conference on Artificial Intelligence, 40(9), 7168-7178. https://doi.org/10.1609/aaai.v40i9.37653
Vancouver
Liu J, Fu R, Sun L, Liu H, Yang X, Zhang W, et al. SkyMoE: A Vision-Language Foundation Model for Enhancing Geospatial Interpretation with Mixture of Experts. Proceedings of the AAAI Conference on Artificial Intelligence. 2026;40(9):7168-7178. doi:10.1609/aaai.v40i9.37653
BibTeX
@article{Liu2026,
author = {Liu, Jiaqi and Fu, Ronghao and Sun, Lang and Liu, Haoran and Yang, Xiao and Zhang, Weipeng and Na, Xu and Duan, Zhuoran and Yang, Bo},
title = {SkyMoE: A Vision-Language Foundation Model for Enhancing Geospatial Interpretation with Mixture of Experts},
journal = {Proceedings of the AAAI Conference on Artificial Intelligence},
year = {2026},
volume = {40},
number = {9},
pages = {7168--7178},
doi = {10.1609/aaai.v40i9.37653}
}
EUVIMED