Human Aligned Compression for Robust Models

ICLR 2026 Conference Submission16637 Authors

Keywords: Compression, adversarial purification, realism in compression, adversarial defense, adversarial attacks

Abstract:

Adversarial attacks on image models threaten system robustness by introducing imperceptible perturbations that cause incorrect predictions.
We investigate human-aligned learned lossy compression as a defense mechanism, comparing two learned models (HiFiC and ELIC) against traditional JPEG across various quality levels.
Our experiments on ImageNet subsets demonstrate that learned compression methods outperform JPEG, particularly for Vision Transformer architectures, by preserving semantically meaningful content while removing adversarial noise.
Even in white-box settings where attackers can access the defense, these methods maintain substantial effectiveness.
We also show that sequential compression—applying rounds of compression/decompression—significantly enhances defense efficacy while maintaining classification performance.
Our findings reveal that human-aligned compression provides an effective, computationally efficient defense that protects the image features most relevant to human and machine understanding.

Primary Area: alignment, fairness, safety, privacy, and societal considerations

Submission Number: 16637