Salvato in:
Dettagli Bibliografici
Autori principali: Min, Yizhou, Lu, Yizhou, Li, Lanqi, Zhang, Zhen, Teng, Jiaye
Natura: Preprint
Pubblicazione: 2026
Soggetti:
Accesso online:https://arxiv.org/abs/2601.21455
Tags: Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
_version_ 1866912858227343360
author Min, Yizhou
Lu, Yizhou
Li, Lanqi
Zhang, Zhen
Teng, Jiaye
author_facet Min, Yizhou
Lu, Yizhou
Li, Lanqi
Zhang, Zhen
Teng, Jiaye
contents Conformal prediction (CP) has become a cornerstone of distribution-free uncertainty quantification, conventionally evaluated by its coverage and interval length. This work critically examines the sufficiency of these standard metrics. We demonstrate that the interval length might be deceptively improved through a counter-intuitive approach termed Prejudicial Trick (PT), while the coverage remains valid. Specifically, for any given test sample, PT probabilistically returns an interval, which is either null or constructed using an adjusted confidence level, thereby preserving marginal coverage. While PT potentially yields a deceptively lower interval length, it introduces practical vulnerabilities: the same input can yield completely different prediction intervals across repeated runs of the algorithm. We formally derive the conditions under which PT achieves these misleading improvements and provides extensive empirical evidence across various regression and classification tasks. Furthermore, we introduce a new metric interval stability which helps detect whether a new CP method implicitly improves the length based on such PT-like techniques.
format Preprint
id arxiv_https___arxiv_org_abs_2601_21455
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle Questioning the Coverage-Length Metric in Conformal Prediction: When Shorter Intervals Are Not Better
Min, Yizhou
Lu, Yizhou
Li, Lanqi
Zhang, Zhen
Teng, Jiaye
Machine Learning
Conformal prediction (CP) has become a cornerstone of distribution-free uncertainty quantification, conventionally evaluated by its coverage and interval length. This work critically examines the sufficiency of these standard metrics. We demonstrate that the interval length might be deceptively improved through a counter-intuitive approach termed Prejudicial Trick (PT), while the coverage remains valid. Specifically, for any given test sample, PT probabilistically returns an interval, which is either null or constructed using an adjusted confidence level, thereby preserving marginal coverage. While PT potentially yields a deceptively lower interval length, it introduces practical vulnerabilities: the same input can yield completely different prediction intervals across repeated runs of the algorithm. We formally derive the conditions under which PT achieves these misleading improvements and provides extensive empirical evidence across various regression and classification tasks. Furthermore, we introduce a new metric interval stability which helps detect whether a new CP method implicitly improves the length based on such PT-like techniques.
title Questioning the Coverage-Length Metric in Conformal Prediction: When Shorter Intervals Are Not Better
topic Machine Learning
url https://arxiv.org/abs/2601.21455