Mehdi Esmaeilzadeh and coauthors introduce Recursive Token Mapper for iterative latent refinement in style-based generators, reporting 4.79 FID on AFHQ-v1 with StyleGAN2-ADA

One Pass Is Not Enough: Recursive Latent Refinement for Generative Models "We introduce RTM, which replaces the single-pass latent mapping in style-based generators with an iterative refinement process, and show that this consistently improves both quality and diversity" "RTM achieves the highest precision and recall among current state-of-the-art approaches while maintaining competitive FID" "Unlike flow-matching baselines that achieve competitive FID at the expense of coverage, recursive refinement improves both quality and diversity simultaneously."