Research map / Probabilistic and scientific modeling / Molecules and proteins
Masked Mineral Modeling: Continent-Scale Mineral Prospecting via Geospatial Infilling
Sujay Nair, Evan Austen Coleman, Sherrie Wang, Elsa Olivetti
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 |
|---|---|---|---|
| use ResNet U-Net instead of ViT (ablation) | M3 infilling model · US mineral infilling, ID and OOD (McDermitt Caldera) | Dice coefficient | Improved |
| use Gaussian process (kriging) instead of M3 (ablation) | infilling baseline · US mineral infilling | Dice coefficient | Hurt |
| coarse-grain inputs 10x for ViT/ResNet comparison (ablation) | M3 vs ViT · US mineral infilling | Dice coefficient | Hurt |
| increase training set size from 1K to 100K (ablation) | M3 · US mineral infilling, ID test | Dice coefficient and recall | Improved |
| increase backbone from ResNet50 to ResNet101 (ablation) | M3 · US mineral infilling, ID test | Dice coefficient | No clear effect |
| increase backbone to ResNet152 (ablation) | M3 · US mineral infilling, ID test | Dice coefficient | Improved |
| add geophysical input layers (ablation) | M3 · US mineral infilling, ID test | Dice coefficient | Improved |
| add agronomic (hyperspectral soil) input layers (ablation) | M3 · US mineral infilling, ID test | Dice coefficient | No clear effect |
| unmask more mineral input layers (ablation) | M3 · US mineral infilling, single-mineral Dice | Dice coefficient | Improved |
| add iron as an input layer for silver prediction (ablation) | M3 · Silver prediction | Dice coefficient: +0.10 | Improved |
Closest papers by meaning
- Towards Automated Petrography (NeurIPS 2025 · similarity 0.85)
- GeoCore-9B: Towards Geo-Aware Generative Foundation Models in Earth Observation (NeurIPS 2026 · similarity 0.83)
- PEACE: Empowering Geologic Map Holistic Understanding with MLLMs (CVPR 2025 · similarity 0.83)
- A Unified Framework for Forward and Inverse Problems in Subsurface Imaging using Latent Space Translations (ICLR 2025 · similarity 0.82)
- Mars-Bench: A Benchmark for Evaluating Foundation Models for Mars Science Tasks (NeurIPS 2025 · similarity 0.81)
- Infusing Synthetic Data with Real-World Patterns for Zero-Shot Material State Segmentation (NeurIPS 2024 · similarity 0.81)
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.
