The field of deepfake detection and media forensics is rapidly advancing, with a focus on developing robust and universal methods for identifying synthetic content. Recent research has explored the use of multi-modal models, uncertainty-guided approaches, and retrieval-augmented generation to improve detection accuracy and efficiency. These innovations have the potential to mitigate the risks associated with deepfakes and enhance the reliability of digital media. Noteworthy papers in this area include Unraveling Hidden Representations, which proposes a multi-modal layer analysis for better synthetic content forensics, and RAVID, which introduces a retrieval-augmented visual detection framework for AI-generated image identification. Additionally, FaceAnonyMixer presents a cancelable face generation framework that leverages latent space mixing to synthesize privacy-preserving face images, and FLUX-Makeup proposes a high-fidelity, identity-consistent, and robust makeup transfer framework via diffusion transformer. These advancements demonstrate significant progress in the field and highlight the importance of continued research in deepfake detection and media forensics.