English

ST-BCP: Tightening Coverage Bound for Backward Conformal Prediction via Non-Conformity Score Transformation

Machine Learning 2026-05-19 v2 Machine Learning

Abstract

Conformal Prediction (CP) provides a statistical framework for uncertainty quantification that constructs prediction sets with coverage guarantees. While CP yields uncontrolled prediction set sizes, Backward Conformal Prediction (BCP) inverts this paradigm by enforcing a predefined upper bound on set size and estimating the resulting coverage guarantee. However, the looseness induced by Markov's inequality within the BCP framework causes a significant gap between the estimated coverage bound and the empirical coverage. In this work, we introduce ST-BCP, a novel method that introduces a data-dependent transformation of nonconformity scores to narrow the coverage gap. In particular, we develop a computable transformation and prove that it outperforms the baseline identity transformation. Extensive experiments demonstrate the effectiveness of our method, reducing the average coverage gap from 4.20\% to 1.12\% on common benchmarks.

Keywords

Cite

@article{arxiv.2602.01733,
  title  = {ST-BCP: Tightening Coverage Bound for Backward Conformal Prediction via Non-Conformity Score Transformation},
  author = {Junxian Liu and Hao Zeng and Hongxin Wei},
  journal= {arXiv preprint arXiv:2602.01733},
  year   = {2026}
}