Hugging Face Trending Papers

SlerpFlow: Spherical Trajectory Correction for Rectified Flow Inversion

Read the original on Hugging Face Trending Papers →

Rectified-flow-based diffusion transformers, particularly FLUX, have demonstrated outstanding performance in high-quality image generation. However, achieving fast and accurate inversion--transforming images back to latent noise for faithful reconstruction and editing--remains a challenging bottleneck due to the discretization errors of linear solvers.

Summary generated by The Flow from the publisher's feed. The full article lives at Hugging Face Trending Papers.