AI Paper Atlas

Research map / Architectures and efficiency / Attention and transformers

AutoFormer: Searching Transformers for Visual Recognition

Minghao Chen, Houwen Peng, Jianlong Fu, Haibin Ling

ICCV 2021 · Architectures and efficiency › Attention and transformers · approach: Efficient and robust vision transformers · 346 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 weight entanglement instead of classical weight sharing, with evolution search (ablation)AutoFormer-S supernet · ImageNet valtop-1 accuracy (inherited): 71.5% -> 81.7%Improved
use weight entanglement instead of classical weight sharing, with evolution search (ablation)AutoFormer-S · ImageNet valtop-1 accuracy (retrain): 80.4% -> 81.7%Improved
use weight entanglement instead of classical weight sharing, with random search (ablation)AutoFormer-S · ImageNet valtop-1 accuracy (inherited): 69.7% -> 81.3%Improved
use evolution search instead of random search with weight entanglement (ablation)AutoFormer-S · ImageNet valtop-1 accuracy (inherited): 81.3% -> 81.7%Improved
use supernet-based search instead of plain random search (ablation)AutoFormer-S · ImageNet valtop-1 accuracy (retrain): 79.4% -> 81.7%Improved
finetune searched subnet 40 epochs after inheriting weightsAutoFormer-T · ImageNet valtop-1 accuracy: 74.7% -> 74.9%No clear effect
finetune searched subnet 40 epochs after inheriting weightsAutoFormer-B · ImageNet valtop-1 accuracy: 82.4% -> 82.6%No clear effect
retrain searched subnet from scratch for 300 epochsAutoFormer-S · ImageNet valtop-1 accuracy: 81.7% -> 81.7%No clear effect
add knowledge distillation from RegNetY-32GF teacherAutoFormer-T/S/B · ImageNet valtop-1 accuracy: 74.7%/81.7%/82.4% -> 75.7%/82.4%/82.9%Improved
fine-tune at 384 resolutionAutoFormer-S · ImageNet valtop-1 accuracy: 81.7% -> 83.4Improved

Later papers that compared with this method

1 later accepted paper reports AutoFormer-tiny as a baseline in their results tables; 1 of them report a better number on most of the results they share with it.

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