EarthShift Benchmark
Explore score history ↗A public testbed that measures how Earth observation models perform in- versus out-of-distribution across shifts in source, time, geography, and sensor, covering 11 tasks and 5 shift types.
- Released
- 2026-05-28
- Institutions
- Arizona State University
- Evaluates
- General AI, Robustness, Safety & Trustworthiness, cs.CV
- Openness
- unknown
- Importer
- Claire Radar
- Review status
- ai-reviewed
EarthShift paper
No reported scores are on record for this benchmark yet.