Saved in:
| Main Authors: | , , , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2025
|
| Subjects: | |
| Online Access: | https://arxiv.org/abs/2506.01297 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866918306708979712 |
|---|---|
| author | Wen, Ya Cai, Jixuan Ma, Qiyao Li, Linyan Chen, Xinhua Webster, Chris Zhou, Yulun |
| author_facet | Wen, Ya Cai, Jixuan Ma, Qiyao Li, Linyan Chen, Xinhua Webster, Chris Zhou, Yulun |
| contents | Representation learning of geospatial locations remains a core challenge in achieving general geospatial intelligence, with increasingly diverging philosophies and techniques. While Earth observation paradigms excel at depicting locations in their physical states, we claim that a location's comprehensive "meaning" is better grounded in its internal human activity patterns and, crucially, its functional relationships with other locations, as revealed by human movement. We present MoRA, a human-centric geospatial framework that leverages a mobility graph as its core backbone to fuse various data modalities, aiming to learn embeddings that represent the socio-economic context and functional role of a location. MoRA achieves this through the integration of spatial tokenization, GNNs, and asymmetric contrastive learning to align 100M+ POIs, massive remote sensing imagery, and structured demographic statistics with a billion-edge mobility graph, ensuring the three auxiliary modalities are interpreted through the lens of fundamental human dynamics. To rigorously evaluate the effectiveness of MoRA, we construct a benchmark dataset composed of 9 downstream prediction tasks across social and economic domains. Experiments show that MoRA, with four input modalities and a compact 128-dimensional representation space, achieves superior predictive performances than state-of-the-art models by an average of 12.9%. Echoing LLM scaling laws, we further demonstrate the scaling behavior in geospatial representation learning. We open-source code and pretrained models at: https://github.com/ylzhouchris/MoRA. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2506_01297 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | MoRA: Mobility as the Backbone for Geospatial Representation Learning at Scale Wen, Ya Cai, Jixuan Ma, Qiyao Li, Linyan Chen, Xinhua Webster, Chris Zhou, Yulun Artificial Intelligence Representation learning of geospatial locations remains a core challenge in achieving general geospatial intelligence, with increasingly diverging philosophies and techniques. While Earth observation paradigms excel at depicting locations in their physical states, we claim that a location's comprehensive "meaning" is better grounded in its internal human activity patterns and, crucially, its functional relationships with other locations, as revealed by human movement. We present MoRA, a human-centric geospatial framework that leverages a mobility graph as its core backbone to fuse various data modalities, aiming to learn embeddings that represent the socio-economic context and functional role of a location. MoRA achieves this through the integration of spatial tokenization, GNNs, and asymmetric contrastive learning to align 100M+ POIs, massive remote sensing imagery, and structured demographic statistics with a billion-edge mobility graph, ensuring the three auxiliary modalities are interpreted through the lens of fundamental human dynamics. To rigorously evaluate the effectiveness of MoRA, we construct a benchmark dataset composed of 9 downstream prediction tasks across social and economic domains. Experiments show that MoRA, with four input modalities and a compact 128-dimensional representation space, achieves superior predictive performances than state-of-the-art models by an average of 12.9%. Echoing LLM scaling laws, we further demonstrate the scaling behavior in geospatial representation learning. We open-source code and pretrained models at: https://github.com/ylzhouchris/MoRA. |
| title | MoRA: Mobility as the Backbone for Geospatial Representation Learning at Scale |
| topic | Artificial Intelligence |
| url | https://arxiv.org/abs/2506.01297 |