Saved in:
| Main Authors: | , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2026
|
| Subjects: | |
| Online Access: | https://arxiv.org/abs/2605.18911 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866913153774780416 |
|---|---|
| author | Xu, Yangshuang Dai, Yuyang Chang, Liling Wang, Qi Dong, Yushun |
| author_facet | Xu, Yangshuang Dai, Yuyang Chang, Liling Wang, Qi Dong, Yushun |
| contents | Wildfire prediction is important for early warning and resource allocation, yet existing Earth foundation models (Earth FMs) are pretrained for general atmospheric and geophysical objectives rather than wildfire forecasting. To address this gap, we introduce WILDFIRE-FM, the first foundation model pretrained specifically for wildfire prediction using weather, active-fire observations, topography, vegetation, and static environmental data. However, introducing a domain-specific backbone alone does not solve the evaluation problem: wildfire events are sparse in space and time, making transfer conclusions highly sensitive to matching rules and evaluation settings. To address this problem, we introduce a fixed-contract evaluation framework with two controlled checks: a fixed-output check for matching-rule effects and a fixed-feature check for head-selection effects. Under matched contracts, we compare WILDFIRE-FM with ten Earth-FM baselines across occupancy, spread, retrieval, and regression tasks. Our results show that wildfire transfer conclusions depend strongly on evaluation design and task formulation. We hope this framework and WILDFIRE-FM provide a foundation for future wildfire-specific Earth-FM research and benchmarking. Our code is available at https://anonymous.4open.science/r/Wildfire-fm-evaluation-contracts-5AE9/. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2605_18911 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | Does Your Wildfire Prediction Model Actually Work, or Just Score Well? Xu, Yangshuang Dai, Yuyang Chang, Liling Wang, Qi Dong, Yushun Machine Learning Artificial Intelligence Wildfire prediction is important for early warning and resource allocation, yet existing Earth foundation models (Earth FMs) are pretrained for general atmospheric and geophysical objectives rather than wildfire forecasting. To address this gap, we introduce WILDFIRE-FM, the first foundation model pretrained specifically for wildfire prediction using weather, active-fire observations, topography, vegetation, and static environmental data. However, introducing a domain-specific backbone alone does not solve the evaluation problem: wildfire events are sparse in space and time, making transfer conclusions highly sensitive to matching rules and evaluation settings. To address this problem, we introduce a fixed-contract evaluation framework with two controlled checks: a fixed-output check for matching-rule effects and a fixed-feature check for head-selection effects. Under matched contracts, we compare WILDFIRE-FM with ten Earth-FM baselines across occupancy, spread, retrieval, and regression tasks. Our results show that wildfire transfer conclusions depend strongly on evaluation design and task formulation. We hope this framework and WILDFIRE-FM provide a foundation for future wildfire-specific Earth-FM research and benchmarking. Our code is available at https://anonymous.4open.science/r/Wildfire-fm-evaluation-contracts-5AE9/. |
| title | Does Your Wildfire Prediction Model Actually Work, or Just Score Well? |
| topic | Machine Learning Artificial Intelligence |
| url | https://arxiv.org/abs/2605.18911 |