Full spectrum Unlearnable Examples via Spectral Equalization

2026-06-25

AI Summary

Existing unlearnable examples suffer from limited robustness due to perturbations concentrated in high-frequency components, rendering them ineffective after low-pass filtering and lacking full-spectrum resilience. This work proposes the first spectrum-agnostic method for generating unlearnable samples by introducing Random Spectrum Masking (RSM) and Cross-Band Guidance (CBG) strategies. These techniques balance the contribution of perturbations across both low- and high-frequency bands while enforcing spectral consistency in the frequency domain. The proposed approach significantly enhances protection in the low-frequency regime and preserves image semantic fidelity, consistently outperforming existing methods across diverse datasets, model architectures, and spectral filtering conditions in terms of robust data protection.

Abstract

Unlearnable examples (UEs) protect training data by injecting imperceptible perturbations so that models fail to extract exploitable representations. In this paper, we reveal that existing UEs exhibit a critical failure once low-pass filtering is applied, indicating that the effective perturbation signals for unlearnability concentrate predominantly in high frequencies. Hence, we argue that reliable UEs should remain effective across the full spectrum. To this end, we propose Full-spectrum Unlearnable examples via Spectral Equalization (FUSE), which aims to generate spectrum-agnostic perturbations by equalizing the contributions from different bands and enforcing cross-band consistency. Specifically, FUSE adopts a Random Spectral Masking (RSM) strategy during generator training, which randomly removes a contiguous frequency band, forcing the remaining bands to maintain unlearnability. In addition, FUSE further integrates Cross-Band Guidance (CBG), which enforces mutual consistency between high- and low-frequency components, thereby further enhancing low-frequency unlearnability and regulating high-frequency perturbations to preserve the semantic fidelity of images. Extensive experiments across multiple datasets, architectures, and spectral filtering demonstrate the strong protection achieved by FUSE.

Problem

• Unlearnable Examples
• Spectral Equalization
• Full-spectrum
• Frequency Robustness
• Data Protection

Innovation

• Unlearnable Examples
• Spectral Equalization
• Random Spectral Masking
• Cross-Band Guidance
• Frequency Robustness