Machine learning approaches achieve equivalent AUROC of 1.0 and near-perfect accuracy to deep learning for OOD detection on over 60,000 medical images but with substantially lower end-to-end latency.
Ethical approval was not required as confirmed by the license attached with the open access data
2 Pith papers cite this work. Polarity classification is still indexing.
2
Pith papers citing it
citation-role summary
background 1
dataset 1
citation-polarity summary
fields
cs.CV 2years
2026 2verdicts
UNVERDICTED 2representative citing papers
SLA detects noisy labels task-agnostically by standardizing and aggregating validation losses across repeated cross-validation folds, generalizing hard-counting into a continuous estimator that outperforms baselines on fundus data.
citing papers explorer
-
A Comparative Study of Machine Learning and Deep Learning for Out-of-Distribution Detection
Machine learning approaches achieve equivalent AUROC of 1.0 and near-perfect accuracy to deep learning for OOD detection on over 60,000 medical images but with substantially lower end-to-end latency.
-
Task-Agnostic Noisy Label Detection via Standardized Loss Aggregation
SLA detects noisy labels task-agnostically by standardizing and aggregating validation losses across repeated cross-validation folds, generalizing hard-counting into a continuous estimator that outperforms baselines on fundus data.