AI Paper Atlas

Research map / Image generation / Autoregressive and latent generation

DiffiT: Diffusion Vision Transformers for Image Generation

Ali Hatamizadeh, Jiaming Song, Guilin Liu, Jan Kautz, Arash Vahdat

ECCV 2024 · Image generation › Autoregressive and latent generation · approach: High-resolution diffusion synthesis · 102 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
add time-dependent multi-head self-attention (TMSA) in place of standard self-attention in DDPM++ (VE) (ablation)DDPM++ (VE) · CIFAR-10FID: 3.77 -> 3.49Improved
add time-dependent multi-head self-attention in DDPM++ (VP) (ablation)DDPM++ (VP) · CIFAR-10FID: 3.01 -> 2.76Improved
add DiffiT transformer blocks to encoder (ablation)multi-resolution U-shaped ViT · CIFAR-10FID: 3.71 -> 2.27Improved
add DiffiT transformer blocks to decoder (ablation)multi-resolution U-shaped ViT · CIFAR-10FID: 2.27 -> 1.95Improved
use multi-resolution (downsampling/upsampling) stages instead of isotropic ViT (ablation)ViT encoder and SETR-MLA decoder · CIFAR-10FID: 5.34 -> 3.71Improved
inject time token into q,k,v projections rather than only relative position bias (ablation)DiffiT · CIFAR-10FID: 3.97 -> 1.95Improved
inject time token into attention rather than only the MLP layer (ablation)DiffiT · CIFAR-10FID: 3.81 -> 1.95Improved
mix time and spatial tokens in the same attention rather than attending separately (ablation)DiffiT · CIFAR-10FID: 2.28 -> 1.95Improved
use positional instead of Fourier time embedding (ablation)DiffiT · CIFAR-10FID: 2.02 -> 1.95Improved
increase classifier-free guidance scale beyond 4.6 (256) / 1.49 (512) (ablation)latent DiffiT · ImageNet-256 / ImageNet-512FID: best 1.73 (256) and 2.67 (512)Hurt
use power-cosine improved classifier-free guidance schedule (ablation)latent DiffiT · ImageNet-512FIDNo clear effect
use DiffiT instead of DiT-XL/2-G backbone designlatent diffusion on ImageNet-256 · ImageNet-256parameters (M) / FID: 675 -> 561 params; FID 2.27 -> 1.73Improved

Later papers that compared with this method

7 later accepted papers report DiffiT as a baseline in their results tables; 4 of them report a better number on most of the results they share with it.

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