SOURCE-LINKED INTELLIGENCE

Split Conformal Prediction with Label-Shift-Adjusted Bayesian Scores

arXiv · AI, language, vision and robotics · article · Sep 11, 2026 · UTC

Conformal prediction provides distribution-free uncertainty quantification under exchangeability. However, this assumption is violated by label shift, where the marginal distribution of labels changes while the conditional distribution of inputs given labels remains stable. Under such shifts, standard conformal procedures no longer maintain their intended coverage behavior. Existing approaches address this via importance weighting. They pair the reweighting with residual-based nonconformity scores that ignore predictive uncertainty. The resulting intervals have uniform width. Bayesian conforma

Read original source ↗ Open in workspace

recordType

paper

region

Global