Conformal Prediction under Partial Verification
Abstract
Conformal prediction provides prediction sets with finite-sample guarantees, but the label verification required for calibration can be expensive. We develop a partial verification method that returns exactly the same prediction sets as complete verification. We characterize calibration certificates, the verified information sufficient to determine the conformal threshold, and design a procedure that coordinates verification across calibration examples. For finite thresholds at high coverage, its verification cost is less than twice the minimum certificate cost when candidates are checked in order. Across retrieval, mathematical solutions, and configuration evaluation, it reduces verification cost by 15-82% compared with verifying calibration examples one at a time, while producing identical prediction sets.
Related
- Design-Based Validation Method of EHR Algorithms to Address Partial Verification Bias
- Generalized Hyperrectangular Prediction Sets in Multivariate Conformal Prediction
- Prediction with Five Experts and Geometric Stopping: A Probabilistic Construction and Analytic Verification
- Multi-Variable Conformal Prediction: Optimizing Prediction Sets using Multiple Calibration Variables
- Conformal Prediction and Conditional Coverage for Tabular Foundation Models
BibTeX
@misc{yu2026conformal,
title = {{Conformal Prediction under Partial Verification}},
author = {Zijun Yu and Yu Gu and Vahid Partovi Nia and Masoud Asgharian},
howpublished = {arXiv},
year = {2026},
month = oct,
doi = {10.48550/arxiv.2610.10829},
eprint = {2610.10829},
archivePrefix = {arXiv},
url = {https://arxiv.org/abs/2610.10829},
}
RIS
TY - UNPB
TI - Conformal Prediction under Partial Verification
AU - Zijun Yu
AU - Yu Gu
AU - Vahid Partovi Nia
AU - Masoud Asgharian
PY - 2026
DA - 2026/10/07
PB - arXiv
DO - 10.48550/arxiv.2610.10829
UR - https://arxiv.org/abs/2610.10829
AB - Conformal prediction provides prediction sets with finite-sample guarantees, but the label verification required for calibration can be expensive. We develop a partial verification method that returns exactly the same prediction sets as complete verification. We characterize calibration certificates, the verified information sufficient to determine the conformal threshold, and design a procedure that coordinates verification across calibration examples. For finite thresholds at high coverage, its verification cost is less than twice the minimum certificate cost when candidates are checked in order. Across retrieval, mathematical solutions, and configuration evaluation, it reduces verification cost by 15-82% compared with verifying calibration examples one at a time, while producing identical prediction sets.
ER -