# A data-centric approach to class-specific bias in image data augmentation

Source: [A data-centric approach to class-specific bias in image data augmentation](<https://www.sota2.com/research/paper/68fa72791e209131cea56967>)

Collection index: [llms.txt](<https://www.sota2.com/research/llms.txt>)

Data updated: 2025-10-28T13:57:04.875Z

Authors: Athanasios Angelakis, Andrey Rass

Published: 2024-03-07T00:32:47.000Z

## Abstract

Data augmentation (DA) enhances model generalization in computer vision but may introduce biases, impacting class accuracy unevenly. Our study extends this inquiry, examining DA's class-specific bias across various datasets, including those distinct from ImageNet, through random cropping. We evaluated this phenomenon with ResNet50, EfficientNetV2S, and SWIN ViT, discovering that while residual models showed similar bias effects, Vision Transformers exhibited greater robustness or altered dynamics. This suggests a nuanced approach to model selection, emphasizing bias mitigation. We also refined a "data augmentation robustness scouting" method to manage DA-induced biases more efficiently, reducing computational demands significantly (training 112 models instead of 1860; a reduction of factor 16.2) while still capturing essential bias trends.

## Recorded benchmark results

A research result applies to its stated dataset, split, metric and evaluation configuration. Compare like-for-like settings, preserve the metric's unit and direction, and distinguish paper-reported results from independent reproduction. The best result recorded here is not a claim of exhaustive, current global coverage.

## Original sources and implementation

[Original paper](<https://arxiv.org/abs/2403.04120>)

[PDF](<https://arxiv.org/pdf/2403.04120.pdf>)