AI Paper Atlas

Research map / Image generation / Autoregressive and latent generation

Reviving ConvNeXt for Efficient Convolutional Diffusion Models

Taesung Kwon, Lorenzo Bianchi, Lennart Wittke, Felix Watine, Fabio Carrara, Jong Chul Ye, Romann Weber, Vinicius Azevedo

CVPR 2026 · Image generation › Autoregressive and latent generation · approach: High-resolution diffusion synthesis · 2 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
use FCDM-XL instead of DiT-XL/2latent diffusion, ImageNet 256x256, no guidance · ImageNet 256x256FID: DiT-XL/2 400K 19.47 -> FCDM-XL 400K 10.72Improved
use FCDM-XL at 1M iterations vs DiT-XL/2 at 7Mlatent diffusion, ImageNet 256x256 · ImageNet 256x256FID: 9.62 -> 7.91Improved
use FCDM backbone instead of DiT at equal parameter countall scales S/B/L/XL · ImageNet 256x256FLOPs (G): XL 118.6 -> 64.6Improved
use FCDM-L instead of DiCo-LL scale, 400K iters · ImageNet 256x256FID: 13.66 -> 13.83No clear effect
use FCDM-XL at 512x512latent diffusion, ImageNet 512x512 · ImageNet 512x512FID: DiT-XL/2 3M 12.03 -> FCDM-XL 1M 7.46Improved
shrink depthwise kernel from 7x7 (ablation)FCDM-L · ImageNet 256x256 200K itersFID: 19.97 -> 20.48 (5x5), 21.28 (3x3)Hurt
replace GRN with compact channel attention from DiCo (ablation)FCDM-L · ImageNet 256x256 200K itersFID: 19.97 -> 23.85Hurt
add DiCo-style feedforward module (ablation)FCDM-L · ImageNet 256x256 200K itersFID: 19.97 -> 28.52Hurt
add inverted bottleneck (ablation)FCDM-L · ImageNet 256x256 200K itersFID: 28.76 -> 19.97Improved
use FCDM block instead of ResNet block (ablation)FCDM-L · ImageNet 256x256 200K itersFID: 31.14 -> 19.97Improved
scale model size from S to XL (ablation)FCDM · ImageNet 256x256 400K itersFID: 48.52 -> 26.21 -> 13.83 -> 10.72 (S, B, L, XL)Improved
use FCDM-XL with guidance instead of latest state of the art (EDM-2, Simpler Diffusion)FCDM-XL · ImageNet 256x256FIDNo clear effect

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.

Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy