Noisy Ostracods: A Fine-Grained, Imbalanced Real-World Dataset for Benchmarking Robust Machine Learning and Label Correction Methods
–Neural Information Processing Systems
The dataset is created to addressing a real-world challenge: creating a clean fine-grained taxonomy dataset.
Neural Information Processing Systems
Feb-14-2026, 07:54:15 GMT
- Country:
- Europe > Germany (0.04)
- North America
- United States
- District of Columbia > Washington (0.04)
- California (0.04)
- Canada > Ontario
- Toronto (0.04)
- United States
- Asia
- Japan > Honshū
- Kantō > Tokyo Metropolis Prefecture > Tokyo (0.04)
- China
- Hong Kong (0.05)
- Jiangsu Province > Nanjing (0.04)
- Japan > Honshū
- Genre:
- Research Report > New Finding (0.93)
- Technology: