AI Paper Atlas

Research map / Architectures and efficiency / Attention and transformers

MG-ViT: A Multi-Granularity Method for Compact and Efficient Vision Transformers

Yu Zhang, Yepeng Liu, Duoqian Miao, Qi Zhang, Yiwei Shi, Liang Hu

NeurIPS 2023 · Architectures and efficiency › Attention and transformers · approach: Vision transformer design · 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
use global-attention-class (CASR across layers) division instead of random division (ablation)MG-ViT (DeiT-S) · ImageNet MGIStop-1 accuracy: 78.9 -> 80.8Improved
use CASR from last layer only vs lambda-weighted global CASR (ablation)MG-ViT (DeiT-S) · ImageNet MGIStop-1 accuracy: 80.5 -> 80.8Improved
lambda sweep for global class attention (0 to 0.999) (ablation)MG-ViT (DeiT-S) · ImageNettop-1 accuracy: 80.3 -> 80.8No clear effect
add Feature Reuse module (ablation)MG-ViT (DeiT-S) · ImageNet MGIStop-1 accuracy: 79.9 -> 80.8Improved
use zero-padding of tail tokens in feature reuse (ablation)MG-ViT (DeiT-S) · ImageNet MGIStop-1 accuracy: 80.7 -> 80.8No clear effect
match-merge boundary tokens instead of pruning (ablation)MG-ViT (DeiT-S) · ImageNettop-1 accuracy: 79.5 -> 80.8Improved
match-merge instead of K-means token clustering (ablation)MG-ViT (DeiT-S) · ImageNettop-1 accuracy: 79.8 -> 80.8Improved
weighted-average merging instead of keep-one (ablation)MG-ViT (DeiT-S) · ImageNettop-1 accuracy: 80.3 -> 80.8Improved
soft (three-way) slimming instead of hard slimming with merging (ablation)MG-ViT (DeiT-S) · ImageNettop-1 accuracy: 80.3 -> 80.8Improved
merge tokens instead of pruning under soft slimming (ablation)MG-ViT (DeiT-S) · ImageNettop-1 accuracy: 80.2 -> 80.8Improved
multi-granularity splitting on all images (eps 1) vs DeiT-SDeiT-S · ImageNettop-1 accuracy: 79.8 -> 81.0Improved
extend to hierarchical ViT (MG-ViT-A) on PVT-SmallPVT-Small · COCO Mask R-CNNbox AP: 40.4 -> 40.9Improved

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