AI Paper Atlas

Research map / Architectures and efficiency / Quantization

QuEST: Low-bit Diffusion Model Quantization via Efficient Selective Finetuning

Haoxuan Wang, Yuzhang Shang, Zhihang Yuan, Junyi Wu, Junchi Yan, Yan Yan

ICCV 2025 · Architectures and efficiency › Quantization · approach: Post-training quantization of ViTs · 39 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-embedding layer finetuning (TLA) to direct quantization (ablation)QuEST · LSUN-Bedrooms LDM-4 W4A8FID: 6.95 -> 4.41Improved
add attention-related layer finetuning (CMA) after TLA (ablation)QuEST · LSUN-Bedrooms LDM-4 W4A8FID: 4.41 -> 3.26Improved
add global loss supervision from full-precision output to TLA (ablation)QuEST · LSUN-Bedrooms LDM-4 W4A8FID: 8.99 -> 6.41Improved
add global loss supervision to CMA (ablation)QuEST · LSUN-Bedrooms LDM-4 W4A8FID: 11.41 -> 6.20Improved
supervise TLA with global loss only (no local loss) (ablation)QuEST · LSUN-Bedrooms LDM-4FID: -7.13 FID, -9.39 sFIDHurt
finetune only time-embedding and attention layers instead of full finetuning (ablation)QuEST · LDM-4 W4A8FID: 5.38 -> 3.82Improved
integrate QALoRA finetuning instead of finetuning original layers (ablation)QuEST · ImageNet 256x256FID: +5.62 FIDHurt
optimize time embeddings with prediction-error loss instead of reusing precomputed features (ablation)QuEST · LDM-4FIDImproved
quantize ImageNet LDM-4 to fewer bits (lower FID under quantization)all methods · ImageNet 256x256 LDM-4FID: 11.28 (FP) -> 5.98 (QuEST W4A4)Mixed
apply QuEST instead of PTQ for efficiencyQuEST · LSUN-Bedrooms LDM-4 W4A8time cost (GPU hours): 23.08 -> 15.25Improved

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