|
|
--- |
|
|
license: mit |
|
|
datasets: |
|
|
- notadib/NASA-Power-Daily-Weather |
|
|
language: |
|
|
- en |
|
|
metrics: |
|
|
- r_squared |
|
|
tags: |
|
|
- weather, |
|
|
- agriculture, |
|
|
--- |
|
|
|
|
|
# VITA: Variational Pretraining of Transformers for Climate-Robust Crop Yield Forecasting |
|
|
|
|
|
This is the official pretrained model weights for the paper [arXiv:2508.03589](https://arxiv.org/abs/2508.03589). VITA is a variational pretraining framework that learns weather representations from rich satellite data and transfers them to yield prediction tasks with limited ground-based measurements. |
|
|
|
|
|
## Overview |
|
|
|
|
|
VITA addresses the data asymmetry problem in agricultural AI: pretraining uses 31 meteorological variables from NASA POWER satellite data, while deployment relies on only 6 basic weather features. Through variational pretraining with a seasonality-aware sinusoidal prior, VITA achieves state-of-the-art performance in predicting corn and soybean yields across 763 U.S. Corn Belt counties, particularly during extreme weather years. |
|
|
|
|
|
## Citation |
|
|
|
|
|
```bibtex |
|
|
@inproceedings{hasan2026vita, |
|
|
title={VITA: Variational Pretraining of Transformers for Climate-Robust Crop Yield Forecasting}, |
|
|
author={Adib Hasan and Mardavij Roozbehani and Munther Dahleh}, |
|
|
booktitle={Proceedings of the 40th AAAI Conference on Artificial Intelligence}, |
|
|
year={2026}, |
|
|
url={https://arxiv.org/abs/2508.03589}, |
|
|
} |
|
|
``` |
|
|
|
