AI Paper Atlas

Research map / Architectures and efficiency / Quantization

DiffPro: Joint Timestep and Layer-Wise Precision Optimization for Efficient Diffusion Inference

Farhana Amin, Sabiha Afroz, Kanchon Gharami, Mona Moghadampanah, Dimitrios Nikolopoulos

ECCV 2026 · Architectures and efficiency › Quantization · approach: Post-training quantization of ViTs · 1 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 dynamic activation quantization (DAQ) (ablation)DiffPro (DiT-XL/2) · ImageNet 256FID: 31.17 -> 28.05Improved
use learned per-layer bit plan instead of uniform W8/G32 (ablation)DiffPro (DiT-XL/2) · ImageNet 256FID: 31.84 -> 28.05Improved
add drift-guided timestep pruning (1000 -> about 500 steps) (ablation)DiffPro (DiT-XL/2) · ImageNet 256FID: FID 30.38 (no prune) vs 28.05; latency 2.65 -> 2.55 sMixed
add drift-guided timestep pruning (ablation)DiffPro (DiT-XL/2) · ImageNet 256latency (sec/image): 2.65 -> 2.55Improved
apply joint INT8 quantization plan plus pruning vs full precisionDiT-XL/2 · ImageNet 256latency and size: latency 6.88 -> 2.55 s; size 2575.42 -> 397.241 MB; FID 20 -> 28.05Mixed
compare with Q-DiT and DiffuseVAE on FIDDiffPro (DiT) · ImageNet 256FID: 28.05 vs 21.59 (QDiT), 23.20 (DiffuseVAE)Hurt

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