AI Paper Atlas

Research map / Image generation / Fast diffusion sampling

Pluggable Pruning with Contiguous Layer Distillation for Diffusion Transformers

Jian Ma, Qirong Peng, Xujie Zhu, Peixing Xie, Chen Chen, Haonan Lu

CVPR 2026 · Image generation › Fast diffusion sampling · approach: Distillation for few-step generation · 6 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 linear probing layer selection (ablation)PPCL · Qwen-Image 12B ablationaverage score: 0.706 -> 0.761Improved
use first-order difference instead of CKA threshold for interval end (ablation)PPCL · Qwen-Image 12B ablationaverage score: 0.718 -> 0.761Improved
extend intervals by one layer (LP-b) (ablation)PPCL · Qwen-Image 12B ablationaverage score: 0.761 -> 0.726Hurt
replace sequential with non-sequential distillation (ablation)PPCL · Qwen-Image 12B ablationaverage score: 0.761 -> 0.848Improved
add text-stream width pruning (linear layers) (ablation)PPCL · Qwen-Image ablationaverage score: 0.848 -> 0.860Improved
replace FFN with linear layer (ablation)PPCL · Qwen-Image ablationaverage score: 0.860 -> 0.850 (1B fewer params)Mixed
add full-parameter fine-tuning (ablation)PPCL · Qwen-Image ablationaverage score: 0.850 -> 0.870Improved
PPCL vs TinyFusion/HierarchicalPrune at 8BPPCL · FLUX.1-devavg performance drop (%): 13.80 / 13.38 -> 4.03Improved
PPCL vs TinyFusion at 14BPPCL · Qwen-Imageavg performance drop (%): 8.75 -> 0.42Improved
restore teacher layers in 10B model to get 12B/14B (plug-and-play)PPCL · Qwen-Imageavg performance drop (%): 6.88 (10B) -> 3.03 (12B) -> 0.42 (14B)Improved

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