The field of image generation and restoration is moving towards more sophisticated and frequency-aware approaches. Recent developments have focused on improving the fidelity and realism of generated images, particularly in degraded or low-quality conditions. This has led to the proposal of new paradigms and frameworks that incorporate frequency knowledge and contextual disentanglement to enhance image generation and restoration. Notably, innovative methods have been introduced to suppress interference in ultrasound images, reconstruct fine details in images, and adapt diffusion models for blind image restoration. These advancements have significant implications for various applications, including content creation, medical imaging, and real-time monitoring. Noteworthy papers include: FICGen, which proposes a frequency-inspired contextual disentanglement generative paradigm to improve layout-driven degraded image generation. HIFU-ILDiff, which leverages latent diffusion models to suppress HIFU-induced interference in ultrasound images for real-time monitoring. Missing Fine Details in Images, which introduces a wavelet-based frequency-aware variational autoencoder framework to improve the reconstruction of fine textures in images. BIR-Adapter, which presents a low-complexity diffusion model adapter for blind image restoration that achieves competitive performance with significantly lower complexity.