Canonicalizing inputs before conformal prediction preserves coverage and shrinks prediction sets under rotation shifts, without retraining the underlying model.
Online conformal prediction with decaying step sizes
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
stat.ML 1years
2025 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
CP$^2$: Leveraging Geometry for Conformal Prediction via Canonicalization
Canonicalizing inputs before conformal prediction preserves coverage and shrinks prediction sets under rotation shifts, without retraining the underlying model.