AI Paper Atlas

Research map / Image generation / Fast diffusion sampling

LiT: Delving into a Simple Linear Diffusion Transformer for Image Generation

Jiahao Wang, Ning Kang, Lewei Yao, Mengzhao Chen, Chengyue Wu, Songyang Zhang, Shuchen Xue, Yong Liu, Taiqiang Wu, Xihui Liu, Kaipeng Zhang, Shifeng Zhang and 3 more

ICCV 2025 · Image generation › Fast diffusion sampling · approach: Training-free diffusion acceleration · 9 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 depth-wise convolution on value (ablation)ReLU linear attention DiT-S/2 · ImageNet 256 DiT-S/2FID: 88.46 -> 63.66Improved
add depth-wise convolution on value (ablation)ReLU linear attention DiT-B/2 · ImageNet 256 DiT-B/2FID: 56.92 -> 42.11Improved
add focused function on top of DWC (ablation)linear DiT-S/2 · ImageNet 256 DiT-S/2FID: 63.66 -> 63.05No clear effect
replace ReLU kernel with GELU (ablation)focused linear DiT-S/2 · ImageNet 256 DiT-S/2FID: 63.05 -> 70.83Hurt
use few heads (2) instead of default 6 (ablation)linear DiT-S/2 · ImageNet 256 DiT-S/2FID: 63.66 -> 63.24Improved
use 96 heads instead of 2 (ablation)linear DiT-S/2 · ImageNet 256 DiT-S/2FID: 63.24 -> 116.00Hurt
use 2 heads instead of 16 (ablation)linear DiT-XL/2 · ImageNet 256 DiT-XL/2FID: 21.69 -> 20.66Improved
initialize from DiT pretrained 200K steps (vs random) (ablation)linear DiT-S/2 · ImageNet 256 DiT-S/2FID: 63.24 -> 57.84Improved
initialize from DiT pretrained 800K vs 200K steps (ablation)linear DiT-S/2 · ImageNet 256 DiT-S/2FID: 57.84 -> 53.83Improved
also load Q,K,V attention projections from DiT (ablation)linear DiT-S/2 · ImageNet 256 DiT-S/2FID: 54.80 -> 55.29Hurt
add noise distillation from larger DiT-XL/2 teacher (lambda1=0.5) (ablation)linear DiT-S/2 · ImageNet 256 DiT-S/2FID: 53.83 -> 51.13Improved
add variance distillation lambda2=0.05 to noise distillation (ablation)linear DiT-S/2 · ImageNet 256 DiT-S/2FID: 51.13 -> 50.79Improved

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