AI Paper Atlas

Research map / Architectures and efficiency / Attention and transformers

Semantic Alignment and Reinforcement for Data-Free Quantization of Vision Transformers

Yunshan Zhong, Yuyao Zhou, Yuxin Zhang, Wanchen Sui, Shen Li, Yong Li, Fei Chao, Rongrong Ji

ICCV 2025 · Architectures and efficiency › Attention and transformers · approach: Token merging and masked pretraining · 4 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 Attention Priors Alignment to one-hot baseline (ablation)SARDFQ · W4/A4 DeiT-S ImageNettop-1 accuracy: 51.73 -> 60.26Improved
add Multi-Semantic Reinforcement alone to one-hot baseline (ablation)SARDFQ · W4/A4 DeiT-S ImageNettop-1 accuracy: 51.73 -> 50.75Hurt
add Softlabel Learning alone to one-hot baseline (ablation)SARDFQ · W4/A4 DeiT-S ImageNettop-1 accuracy: 51.73 -> 52.02Improved
add Softlabel Learning to MSR (ablation)SARDFQ · W4/A4 DeiT-S ImageNettop-1 accuracy: 50.75 -> 56.08Improved
add MSR to APA (ablation)SARDFQ · W4/A4 DeiT-S ImageNettop-1 accuracy: 60.26 -> 61.58Improved
add MSR to APA+SL (full method) (ablation)SARDFQ · W4/A4 DeiT-S ImageNettop-1 accuracy: 60.51 -> 62.29Improved
add SL to APA+MSR (full method) (ablation)SARDFQ · W4/A4 DeiT-S ImageNettop-1 accuracy: 61.58 -> 62.29Improved
apply APA only to deeper blocks (S=L/2) instead of all blocks (ablation)SARDFQ · W4/A4 DeiT-S ImageNettop-1 accuracy: 61.96 -> 62.29Improved
add depth weighting l/L to APA loss (ablation)SARDFQ · W4/A4 DeiT-S ImageNettop-1 accuracy: 61.32 -> 62.29Improved
use real attention priors instead of GMM priors (ablation)SARDFQ · W4/A4 DeiT-S ImageNettop-1 accuracy: 62.29 -> 63.19Improved
use Laplace instead of GMM attention priors (ablation)SARDFQ · W4/A4 DeiT-S ImageNettop-1 accuracy: 62.29 -> 62.16No clear effect
set K_APA to 5 instead of 1 (ablation)SARDFQ · W4/A4 DeiT-S ImageNettop-1 accuracy: 61.13 -> 62.29Improved

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