AI Paper Atlas

Research map / Architectures and efficiency / Attention and transformers

Vision Transformer Slimming: Multi-Dimension Searching in Continuous Optimization Space

Arnav Chavan, Zhiqiang Shen, Zhuang Liu, Zechun Liu, Kwang-Ting Cheng, Eric P. Xing

CVPR 2022 · Architectures and efficiency › Attention and transformers · approach: Efficient and robust vision transformers · 69 citations (OpenAlex)

Check 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
Search MHSA dimensions only instead of retaining all dimensions (ablation)DeiT-S · ImageNet-1KTop-1 accuracy: 79.90 -> 80.90Improved
Search MLP dimensions only instead of retaining all dimensions (ablation)DeiT-S · ImageNet-1KTop-1 accuracy: 79.90 -> 80.80Improved
Use joint MHSA and MLP dimension search at 70% budget (ablation)DeiT-S · ImageNet-1KTop-1 accuracy: 79.90 -> 80.03Improved
Reduce the joint MHSA and MLP budget from 60% to 50% (ablation)DeiT-S ViT-Slim-PS · ImageNet-1KTop-1 accuracy: 79.20 -> 77.94Hurt
Search MLP dimensions rather than MHSA dimensions at equal FLOPs (ablation)DeiT-S · ImageNet-1KTop-1 accuracy: MHSA 40%: 79.61; MLP 60%: 80.39Mixed
Add patch selection and reduce active patches from 100% to 80% (ablation)DeiT-S ViT-Slim-JS, MHSA and MLP fixed at 70% · ImageNet-1KTop-1 accuracy: 80.03 -> 79.91No clear effect
Reduce active patches from 100% to 70% (ablation)DeiT-S ViT-Slim-JS, MHSA and MLP fixed at 70% · ImageNet-1KTop-1 accuracy: 80.03 -> 79.72No clear effect
Reduce active patches from 100% to 60% (ablation)DeiT-S ViT-Slim-JS, MHSA and MLP fixed at 70% · ImageNet-1KTop-1 accuracy: 80.03 -> 79.51Hurt
Tune the MHSA and MLP sparsity weights to 2e-04 and 5e-05 (ablation)DeiT-S ViT-Slim-PS at 60% budget · ImageNet-1Kfinal Top-1 accuracy: 79.10 -> 79.20Improved
Use joint MHSA and MLP search at 60% budget (ablation)DeiT-S ViT-Slim-PS · ImageNet-1KTop-1 accuracy: 79.90 -> 79.20Hurt
Use joint MHSA and MLP search at 50% budget (ablation)DeiT-S ViT-Slim-PS · ImageNet-1KTop-1 accuracy: 79.90 -> 77.94Hurt
Compress DeiT-B using ViT-Slim at 60% budgetDeiT-B · ImageNet-1KTop-1 accuracy: 81.8 -> 82.4Improved

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