The paper derives that calibration-conditional coverage follows a Beta(k, n+1-k) law under continuous i.i.d. exchangeability and quantifies non-i.i.d. departures via Wasserstein distances on transported beta laws, yielding explicit bounds in scale-shift, clustered, and mixing regimes.
Journal of Machine Learning Research , volume=
3 Pith papers cite this work. Polarity classification is still indexing.
3
Pith papers citing it
years
2026 3verdicts
UNVERDICTED 3representative citing papers
Trimming helps conformal prediction under contamination precisely when the anomaly score separates retention probabilities without biasing clean scores, otherwise the retained mixture coefficient prevents substantial decontamination.
A classification-integrated conformal framework for zero-inflated outcomes that guarantees marginal coverage and asymptotic minimal length under exchangeability, independent of the underlying models.
citing papers explorer
-
Conformal Prediction via Transported Beta Laws
The paper derives that calibration-conditional coverage follows a Beta(k, n+1-k) law under continuous i.i.d. exchangeability and quantifies non-i.i.d. departures via Wasserstein distances on transported beta laws, yielding explicit bounds in scale-shift, clustered, and mixing regimes.
-
When Does Trimming Help Conformal Prediction? A Retained-Law Diagnostic under Calibration Contamination
Trimming helps conformal prediction under contamination precisely when the anomaly score separates retention probabilities without biasing clean scores, otherwise the retained mixture coefficient prevents substantial decontamination.
-
Classification-Powered Conformal Inference for Zero-inflated Outcomes
A classification-integrated conformal framework for zero-inflated outcomes that guarantees marginal coverage and asymptotic minimal length under exchangeability, independent of the underlying models.