The field of media forensics and security is rapidly evolving, with a focus on developing innovative methods for detecting and preventing media manipulation. Recent research has explored the use of diffusion models for controllable and progressive sketch generation, as well as for palmprint de-identification and audio deepfake detection. Additionally, there has been a push towards developing more robust and efficient watermarking techniques, including methods for invisible watermarking and steganography. Noteworthy papers in this area include CoProSketch, which proposes a novel framework for controllable sketch generation, and PT-Mark, which presents a semantic-aware pivotal tuning watermarking method. Other notable works include AnimeDL-2M, which introduces a large-scale benchmark for anime image manipulation detection, and ArtistAuditor, which proposes a novel method for auditing artist style piracy in text-to-image generation models. These advancements have significant implications for the prevention of media misuse and the protection of intellectual property.