Metraipolis › archive › dc_a67f0c83c5da

A Composition-Aware Pretraining Framework for Geospatial Foundation Models — arXiv:2608.30817

Source: arXiv (en)
Original: https://arxiv.org/abs/2608.30817
License (as archived): CC0-1.0
Archived: 2026-10-17T07:52:58.106Z
SHA-256 of the archived text: 0e8625b30a98f7691ba956f6079c3ca8b327dd6143c8adba1999431f615ed8d7
Stable URL of this page: https://metraipolis.com/documento/dc_a67f0c83c5da

How to cite

A Composition-Aware Pretraining Framework for Geospatial Foundation Models — arXiv:2608.30817 — arXiv (en), https://arxiv.org/abs/2608.30817 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_a67f0c83c5da (Metraipolis archive dc_a67f0c83c5da, sha256 0e8625b30a98f7691ba956f6079c3ca8b327dd6143c8adba1999431f615ed8d7)

BibTeX

@misc{dca67f0c83c5da,
  title = {A Composition-Aware Pretraining Framework for Geospatial Foundation Models — arXiv:2608.30817},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2608.30817},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_a67f0c83c5da; sha256 0e8625b30a98f7691ba956f6079c3ca8b327dd6143c8adba1999431f615ed8d7},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2608.30817. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Geospatial foundation models have emerged as state-of-the-art methods for downstream Earth observation tasks. However, existing pretraining methodologies process imagery through a single-concept lens, failing to capture the highly compositional nature of complex satellite scenes. We propose a composition-aware pretraining framework that explicitly encodes fractional land-cover mixtures. Each satellite image cell is mapped to a histogram representing its fractional land-cover distribution, which we term the "composition target". These targets serve as the primary prediction objective and are distilled into the backbone using Earth Mover's Distance. Experimental evaluation shows that composition-aware pretraining yields substantial gains on region-level understanding tasks requiring semantic similarity judgment, including zero-shot image retrieval and scene classification, while remaining competitive on tasks requiring fine-grained spatial precision, such as segmentation and object detection. With a 36.8M-parameter backbone, our framework outperforms SatMAE and Prithvi-EO-2.0, which contain 303M and 600M parameters, respectively, in most retrieval and scene classification settings. On the fine-grained ForestNet-12 dataset, a rigorous testbed for compositional discrimination, our method boosts baseline mAP@10 from 0.279 to 0.434, a 55.6% relative improvement, providing direct evidence for the effectiveness of explicit composition modeling. The code implementation can be found at this https URL

Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-09-01 in cs.AI).
Categorie arXiv: Computer Vision and Pattern Recognition (cs.CV) ; Artificial Intelligence (cs.AI).
Pagina del lavoro: https://arxiv.org/abs/2608.30817 — autori e testo completo sono lì (qui omessi i nomi: nessun dato personale nel deposito).
Licenza: titolo e abstract sono metadati arXiv distribuiti in CC0 1.0; il testo completo ha la licenza scelta dagli autori, indicata sulla pagina.
Letto il 2026-10-17T07:28:26.911Z dalla pagina https://arxiv.org/catchup/cs.AI/2026-09-01?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).