SHRUG-FM: Reliability-Aware Foundation Models for Earth Observation
Geospatial foundation models (GFMs) for Earth observation often fail to perform reliably in environments underrepresented during pretraining. We introduce SHRUG-FM, a framework for reliability-aware prediction that enables GFMs to identify and abstain from likely failures. Our approach integrates three complementary signals: geophysical out-of-distribution (OOD) detection in the input space, OOD detection in the embedding space, and task-specific predictive uncertainty. We evaluate SHRUG-FM across three high-stakes rapid-mapping tasks: burn scar segmentation, flood mapping, and landslide detection. Our results show that SHRUG-FM consistently reduces prediction risk on retained samples, outperforming established single-signal baselines like predictive entropy. Crucially, by utilizing a shallow "glass-box" decision tree for signal fusion, SHRUG-FM provides interpretable abstention thresholds. It builds a pathway toward safer and more interpretable deployment of GFMs in climate-sensitive applications, bridging the gap between benchmark performance and real-world reliability.
Code (0)
등록된 구현이 없습니다.
Similar Papers 제목 키워드 기반
TerraFlow: Multimodal, Multitemporal Representation Learning for Earth Observation
We propose TerraFlow, a novel approach to multimodal, multitemporal learning for Earth observation. TerraFlow builds on temporal training objectives that enable sequence-aware learning across space, time, and modality, w…
Representation LearningEarth Science Foundation Models: From Perception to Reasoning and Discovery
Large foundation models (FMs) are transforming Earth science by integrating heterogeneous multimodal data, such as multi-platform imagery, gridded reanalysis data, diverse geophysical and geochemical observations, and do…
Multimodal ReasoningFoundation Models Meet Agriculture: Challenges Beyond Pretraining
Global food security and sustainable climate action increasingly rely on robust, scalable agricultural monitoring. Earth observation foundation models have emerged as powerful, label-efficient tools across general remote…
Crop ClassificationLightweight Metadata-Aware Mixture-of-Experts Masked Autoencoder for Earth Observation
Recent advances in Earth Observation have focused on large-scale foundation models. However, these models are computationally expensive, limiting their accessibility and reuse for downstream tasks. In this work, we inves…
OlmoEarth: Stable Latent Image Modeling for Multimodal Earth Observation
Earth observation data presents a unique challenge: it is spatial like images, sequential like video or text, and highly multimodal. We present OlmoEarth: a multimodal, spatio-temporal foundation model that employs a nov…
Self-Supervised Learning