Learning Sample Difficulty from Pre-trained Models for Reliable Prediction
–Neural Information Processing Systems
Equipped with the knowledge of sample difficulty, we further propose to use it for regularizing the prediction confidence.
Neural Information Processing Systems
Feb-11-2026, 17:38:33 GMT
- Country:
- Asia > China
- Beijing > Beijing (0.04)
- Shandong Province (0.04)
- Shanghai > Shanghai (0.04)
- Europe
- Germany > Baden-Württemberg
- Tübingen Region > Tübingen (0.04)
- Slovenia > Drava
- Municipality of Benedikt > Benedikt (0.04)
- Germany > Baden-Württemberg
- North America > United States
- Asia > China
- Technology:
- Information Technology
- Artificial Intelligence
- Machine Learning
- Neural Networks (0.70)
- Performance Analysis > Accuracy (1.00)
- Natural Language (0.93)
- Vision (1.00)
- Machine Learning
- Data Science > Data Mining (0.93)
- Sensing and Signal Processing > Image Processing (0.93)
- Artificial Intelligence
- Information Technology