White Paper Assistance: A Step Forward Beyond the Shortcut Learning

TL;DR

This paper introduces White Paper Assistance, a novel method to reduce shortcut learning in CNNs by using white paper to detect and mitigate spurious correlations, leading to improved accuracy and robustness.

Contribution

The paper proposes a new approach called White Paper Assistance that effectively reduces shortcut learning in CNNs across various tasks and architectures.

Findings

Consistent accuracy improvements across multiple datasets and models.

Enhanced robustness to data corruptions and imbalanced classes.

Versatility demonstrated in fine-grained recognition tasks.

Abstract

The promising performances of CNNs often overshadow the need to examine whether they are doing in the way we are actually interested. We show through experiments that even over-parameterized models would still solve a dataset by recklessly leveraging spurious correlations, or so-called 'shortcuts'. To combat with this unintended propensity, we borrow the idea of printer test page and propose a novel approach called White Paper Assistance. Our proposed method involves the white paper to detect the extent to which the model has preference for certain characterized patterns and alleviates it by forcing the model to make a random guess on the white paper. We show the consistent accuracy improvements that are manifest in various architectures, datasets and combinations with other techniques. Experiments have also demonstrated the versatility of our approach on fine-grained recognition, imbalanced classification and robustness to corruptions.