Shortcut Learning in Geospatial Foundation Model Embeddings for Cross-Disaster Damage Assessment

arXiv:2610.13766 · cs.CV, cs.LG · Submitted 2026-10-09 · Read on arXiv

cs.CV, cs.LG

Submitted: 2026-10-09

Updated: 2026-10-09

Comments: NeurIPS REO Workshop 2026

Code: https://github.com/alishibli97/gfm-shortcut-learning

License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/

Terminology

• Embeddings — 1× in this paper · explained in 2 episodes
• Foundation Model — 1× in this paper · explained in 3 episodes

Sources

• EarthShift: a benchmark for measuring robustness to real-world distribution shifts in Earth observation
• xBD: A Dataset for Assessing Building Damage from Satellite Imagery
• REOBench: Benchmarking Robustness of Earth Observation Foundation Models
• Robust Building Damage Detection in Cross-Disaster Settings Using Domain Adaptation
• DINOv3
• Cross-domain Contrastive Learning for Unsupervised Domain Adaptation
• Neural Plasticity-Inspired Multimodal Foundation Model for Earth Observation
• The Clever Hans Mirage: A Comprehensive Survey on Spurious Correlations in Machine Learning

Related papers

• Loss Knows Best: Detecting Annotation Errors in Videos via Loss Trajectories
• AnchorWeave: World-Consistent Video Generation with Retrieved Local Spatial Memories
• Benchmarking the Robustness of Foundation Models for Mammography under Domain Shift
• MambaX-Net: Dual-Input Mamba-Enhanced Cross-Attention Network for Longitudinal MRI Segmentation
• TeleOCR: Navigating Document Parsing Across Digital and Camera-Captured Documents
• A Survey on Efficient Vision-Language-Action Models