Research map / Probabilistic and scientific modeling / Molecules and proteins
YieldSAT: A Multimodal Benchmark Dataset for High-Resolution Crop Yield Prediction
Miro Miranda, Deepak Pathak, Patrick Helber, Benjamin Bischke, Hiba Najjar, Francisco Mena, Cristhian Sanchez, Akshay Pai, Diego Arenas, Matias Valdenegro-Toro, Marcela Charfuelan, Marlon Nuske and 1 more
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| add weather, soil and topography data via input fusion (ablation) | 3D-LSTM / 3D-ConvLSTM · Argentina soybean field-level | R2: 3D-LSTM 0.77 -> 0.76; 3D-ConvLSTM 0.79 -> 0.82 | Mixed |
| add ADMs via input fusion to spatial models (ablation) | 3D-LSTM · Argentina soybean pixel-level | R2: 3D-LSTM Argentina soybean pixel 0.65 -> 0.64 | No clear effect |
| use advanced feature fusion (AFF) instead of S2 only (ablation) | 3D-LSTM · Argentina soybean field-level | R2: 0.77 -> 0.84 | Improved |
| use advanced feature fusion (AFF) instead of S2 only (ablation) | 3D-LSTM · Argentina soybean pixel-level | R2: 0.65 -> 0.73 | Improved |
| model spatial neighborhood with 3D-CNN blocks (ablation) | LSTM · Argentina soybean pixel-level (S2 only) | R2: 0.60 -> 0.65 | Improved |
| evaluate under Leave-One-Year-Out instead of standard CV | LSTM · Argentina soybean | R2: 0.72 -> 0.50 (field-level) | Hurt |
| evaluate under Leave-One-Region-Out instead of standard CV | LSTM · Argentina soybean | R2: 0.72 -> 0.64 (field-level) | Hurt |
| use 5-member Deep Ensemble instead of single LSTM (ablation) | LSTM · Argentina soybean LOYO | R2: 0.50 -> 0.55 | Improved |
| use 5-member Deep Ensemble instead of single LSTM (ablation) | LSTM · Argentina corn LORO | R2: 0.47 -> 0.59 | Improved |
| use Deep Ensemble of 3D-LSTM instead of baseline LSTM (ablation) | LSTM · Argentina corn LORO | R2: 0.47 -> 0.76 | Improved |
| use Deep Ensemble of 3D-LSTM instead of baseline LSTM (ablation) | LSTM · Argentina soybean LOYO | R2: 0.50 -> 0.63 | Improved |
| use Deep Ensemble on wheat (ablation) | LSTM · Argentina wheat LOYO | R2: 0.77 -> 0.79 | No clear effect |
Closest papers by meaning
- Combining Deep Learning and Street View Imagery to Map Smallholder Crop Types (AAAI 2024 · similarity 0.88)
- Can Large Multimodal Models Understand Agricultural Scenes? Benchmarking with AgroMind (NeurIPS 2025 · similarity 0.86)
- Fields of The World: A Machine Learning Benchmark Dataset for Global Agricultural Field Boundary Segmentation (AAAI 2025 · similarity 0.86)
- Exact: Exploring Space-Time Perceptive Clues for Weakly Supervised Satellite Image Time Series Semantic Segmentation (CVPR 2025 · similarity 0.84)
- HarvestNet: A Dataset for Detecting Smallholder Farming Activity Using Harvest Piles and Remote Sensing (AAAI 2024 · similarity 0.84)
- AgMMU: A Comprehensive Agricultural Multimodal Understanding Benchmark (NeurIPS 2025 · similarity 0.83)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.
