cs.LGSep 7, 2026

Prevalence calibration as shortcut mitigation

Authors: Mohamed Amine KinaEike Petersen

Organizations: Universität Bremen, Bremen, Germany · Fraunhofer Institute for Digital Medicine MEVIS, Bremen, Germany

Abstract

Shortcut learning denotes the widespread situation in which a classifier exploits spurious correlations rather than diagnostic features. Existing mitigation strategies mostly aim to learn shortcut-invariant representations; their empirical success is limited and they cannot be applied to classifiers using frozen foundation model encoders. We propose to reframe shortcut learning as fundamentally a calibration problem: unconstrained learning implicitly calibrates each shortcut group to its training set disease prevalence, rendering the resulting classifier necessarily over-confident in one group and under-confident in the other. Building on this insight, we prevalence-equalize calibration between shortcut groups through two encoder-agnostic methods, an in-processing regularizer and a post-hoc prevalence-equalized recalibration step. Across chest-drain-pneumothorax benchmarks on CheXpert and SIIM-ACR, spanning fine-tuned CNNs and frozen foundation-model backbones, both methods substantially outperform all baselines. Post-hoc recalibration of a standard ERM-trained DenseNet raises misaligned-group AUROC from 0.23 to 0.73, indicating that shortcut reliance degrades the classification head rather than the underlying representation. Besides two new state-of-the-art shortcut mitigation approaches, our findings more fundamentally connect shortcut learning to calibration theory and algorithmic fairness.

Explore similar work

CardsList
  1. Look What the Probes Dragged In! Real-World Chest X-ray Shortcuts in MedCLIP

    Aug 12, 2026Nikolette Pedersen, Regitze Sydendal, Veronika Cheplygina +1Chest X-RayChest Radiography

  2. Shortcut Mitigation via Spurious-Positive Samples

    May 13, 2026Phuong Quynh Le, Jörg Schlötterer, Sari Sadiya +2Training Data AttributionMitigation

  3. Unbiased Prevalence Estimation with Multicalibrated LLMs

    Apr 23, 2026Fridolin Linder, Thomas Leeper, Daniel Haimovich +3PrevalenceImplicit Bias