Steganographic Embeddings As an Effective Data Augmentation

Nicholas DiSalvo · 2025

Read Steganographic Embeddings As an Effective Data Augmentation by Nicholas DiSalvo (2025) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2502-15245.txt.

Description

Nicholas DiSalvo turns an old hiding trick, in a 2025 preprint, into a way to train image classifiers better. Steganography hides one image inside another by swapping the least important bits of each pixel. The cover image looks unchanged to the eye. DiSalvo asks what happens if such merged images feed a neural network as training data. Each training picture then quietly carries a second one. The network sees a richer mix without any new data collection.

On the CIFAR-10 benchmark, the trick sped up training a great deal. It also acted like a bundle of classic color tweaks, such as brightness and hue shifts. Those tweaks usually need fussy tuning of extra settings. The steganographic version needed none. One training run did the work of a jointly tuned augmentation pipeline. The author presents the method as nearly free improvement for image training.

Catalog record

CREATORNicholas DiSalvo
YEAR2025
COLLECTIONPAPERS
FORMATTXT
FILE SIZE21 KB
VERSION1.0
LICENSEhttp://creativecommons.org/licenses/by/4.0/
RIGHTSCC BY
SOURCEarXiv

Source: https://arxiv.org/abs/2502-15245

Browse the full library · About the Tower of Babel