Distilling Cross-Modal Knowledge via Feature Disentanglement

Knowledge distillation (KD) has proven highly effective for compressing large models and enhancing the performance of smaller ones. However, its effectiveness diminishes in cross-modal scenarios, such as vision-to-language distillation, where inconsistencies in representation across modalities lead to difficult knowledge transfer. To address this challenge, we propose frequency-decoupled cross-modal knowledge distillation, a method designed to decouple and balance knowledge transfer across modalities by leveraging frequency-domain features. We observed that low-frequency features exhibit high consistency across different modalities, whereas high-frequency features demonstrate extremely low cross-modal similarity. Accordingly, we apply distinct losses to these features: enforcing strong alignment in the low-frequency domain and introducing relaxed alignment for high-frequency features. We also propose a scale consistency loss to address distributional shifts between modalities, and employ a shared classifier to unify feature spaces. Extensive experiments across multiple benchmark datasets show our method substantially outperforms traditional KD and stateof-the-art cross-modal KD approaches.