Skip to main content
Search

# Lossless Latent Diffusion Model

1 related articles

Subscribe to updates
  1. arXiv

    Lossless diffusion without touching the model: LLDM splits pixels into residuals and rescales them, cutting ImageNet gFID from 104.29 to 11.64

    The Lossless Latent Diffusion Model (LLDM) takes a different route: instead of changing the diffusion architecture or objective, it builds an invertible transform that maps pixels into a representation easier for diffusion to model. It keeps a pretrained lossy VAE's latent and concatenates the reconstruction error as a residual along the channel dimension to guarantee pixel-exact reconstruction, then applies uneven scaling to that residual. On ImageNet class-to-image, LLDM reaches gFID 11.64 versus 104.29 for pixel-space diffusion; without classifier-free guidance, SD-VAE is 26.02 and FLUX2-VAE is 10.18.

You’re all caught up