Dual-IFM delivers state-of-the-art-comparable performance on retinal fundus images with inherent local and global interpretability through evidence maps and 2D projections after training on over 800,000 images.
Advances in neural information processing systems31 (2018)
2 Pith papers cite this work. Polarity classification is still indexing.
2
Pith papers citing it
years
2026 2verdicts
UNVERDICTED 2representative citing papers
Neural networks trained via supervised contrastive learning yield feature attributions that are more faithful, less complex, and more continuous than those from cross-entropy trained networks.
citing papers explorer
-
Towards Interpretable Foundation Models for Retinal Fundus Images
Dual-IFM delivers state-of-the-art-comparable performance on retinal fundus images with inherent local and global interpretability through evidence maps and 2D projections after training on over 800,000 images.
-
On the Properties of Feature Attribution for Supervised Contrastive Learning
Neural networks trained via supervised contrastive learning yield feature attributions that are more faithful, less complex, and more continuous than those from cross-entropy trained networks.