Noise injection during training reduces the ID-OOD performance gap in COVID-19 CXR classification from 0.10-0.20 to 0.01-0.06.
Chest imaging representing a covid-19 positive rural us population
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.CV 1years
2025 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
Noise Injection: Improving Out-of-Distribution Generalization for Limited Size Datasets
Noise injection during training reduces the ID-OOD performance gap in COVID-19 CXR classification from 0.10-0.20 to 0.01-0.06.