Applying learned test-time augmentation before conformal scoring reduces prediction set sizes by 10-14% with no loss of nominal coverage.
Test-time Data Augmentation for Estimation of Heteroscedastic Aleatoric Uncertainty in Deep Neural Networks
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
citation-role summary
background 1
citation-polarity summary
fields
cs.LG 1years
2025 1verdicts
CONDITIONAL 1roles
background 1polarities
background 1representative citing papers
citing papers explorer
-
Test-time augmentation improves efficiency in conformal prediction
Applying learned test-time augmentation before conformal scoring reduces prediction set sizes by 10-14% with no loss of nominal coverage.