The field of image restoration and generation is moving towards the use of diffusion models and transformers to improve the quality and efficiency of image processing tasks. Researchers are exploring the use of dual prompting, image quality predictors, and physics-informed diffusion models to enhance the restoration and generation of images. These approaches have shown promising results in achieving high-quality image restoration, super-resolution, and generation. Notably, the use of diffusion models has improved the balance between perceptual fidelity and human-tuned image quality assessment measures. Noteworthy papers include: Dual Prompting Image Restoration with Diffusion Transformers, which introduces a novel image restoration method that effectively extracts conditional information of low-quality images from multiple perspectives. Diff-Prompt: Diffusion-Driven Prompt Generator with Mask Supervision, which proposes a diffusion-driven prompt generator that achieves state-of-the-art results in referring expression comprehension. DGSolver: Diffusion Generalist Solver with Universal Posterior Sampling for Image Restoration, which introduces a diffusion generalist solver that outperforms state-of-the-art methods in restoration accuracy, stability, and scalability.