Bill Psomas @billpsomas.bsky.social · Dec 27 1 / 2
10/n Faster convergence🔥 REGLUE (SiT-B/2) achieves 12.9 and 28.7 FID at 400K iterations in conditional and unconditional generation, respectively, outperforming REPA, ReDi, and REG. REGLUE (SiT-XL/2) matches 1M-step SOTA performance in just 700k iterations (~30% fewer steps).
0 likes 1 replies
?
Replies
11/n Summary🏁 REGLUE shows that the way we leverage VFM semantics matters for diffusion. Combining compact local semantics with global context yields faster convergence and state-of-the-art image generation. 📄arXiv: arxiv.org/abs/2512.16636 💻Project: reglueyourlatents.github.io