Pith. sign in

REVIEW 1 cited by

Trust, or Don't Predict: Introducing the CWSA Family for Confidence-Aware Model Evaluation

Not yet reviewed by Pith; the record is open.

This paper has not been read by Pith yet. Machine review is queued; the pith claim, tier, and objections will appear here once it completes.

SPECIMEN: schema-true, not a live event

T0 review · schema-true

One-sentence machine reading of the paper's core claim.

pith:XXXXXXXX · record.json · timestamp

arxiv 2505.18622 v1 pith:2E52HR3J submitted 2025-05-24 cs.LG cs.AIstat.ML

Trust, or Don't Predict: Introducing the CWSA Family for Confidence-Aware Model Evaluation

classification cs.LG cs.AIstat.ML
keywords cwsametricsaccuracyconfidencemodeloverconfidentpredictionselective
verification ladder T0 review T1 audit T2 compute T3 formal T4 reserved
0 comments
Share X Bluesky LinkedIn Reddit HN
read the original abstract

In recent machine learning systems, confidence scores are being utilized more and more to manage selective prediction, whereby a model can abstain from making a prediction when it is unconfident. Yet, conventional metrics like accuracy, expected calibration error (ECE), and area under the risk-coverage curve (AURC) do not capture the actual reliability of predictions. These metrics either disregard confidence entirely, dilute valuable localized information through averaging, or neglect to suitably penalize overconfident misclassifications, which can be particularly detrimental in real-world systems. We introduce two new metrics Confidence-Weighted Selective Accuracy (CWSA) and its normalized variant CWSA+ that offer a principled and interpretable way to evaluate predictive models under confidence thresholds. Unlike existing methods, our metrics explicitly reward confident accuracy and penalize overconfident mistakes. They are threshold-local, decomposable, and usable in both evaluation and deployment settings where trust and risk must be quantified. Through exhaustive experiments on both real-world data sets (MNIST, CIFAR-10) and artificial model variants (calibrated, overconfident, underconfident, random, perfect), we show that CWSA and CWSA+ both effectively detect nuanced failure modes and outperform classical metrics in trust-sensitive tests. Our results confirm that CWSA is a sound basis for developing and assessing selective prediction systems for safety-critical domains.

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Forward citations

Cited by 1 Pith paper

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score.

  1. Beyond ECE: Calibrated Size Ratio, Risk Assessment, and Confidence-Weighted Metrics

    cs.LG 2026-05 unverdicted novelty 7.0

    Introduces Calibrated Size Ratio (CSR) and confidence-weighted metrics to better detect overconfidence risk and calibration issues beyond the limitations of ECE.