AI Paper Atlas

Research map / Architectures and efficiency / Attention and transformers

Stripe Observation Guided Inference Cost-free Attention Mechanism

Zhongzhan Huang, Shanshan Zhong, Wushao Wen, Jinghui Qin, Liang Lin

ECCV 2024 · Architectures and efficiency › Attention and transformers · approach: Sparse and structured attention · 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
add ASR (SE/ECA/SRM) re-parameterized attention to ResNet at no inference costResNet34/50/101 · ImageNet-1k valtop-1 accuracy: 76.13 -> 76.87Improved
add ASR to ShuffleNetV2 with SEShuffleNetV2 · CIFAR10top-1 accuracy: 91.68 -> 91.68No clear effect
add ASR on top of an existing attention moduleSE in ResNet164 · CIFAR100top-1 accuracy: 75.29 -> 77.32Improved
replace standalone attention with ASR counterpartSRM in ResNet164 · CIFAR100top-1 accuracy: 75.38 -> 75.23Hurt
add ASR to RepVGG and ACNetRepVGG, ACNet · STL10 / CIFAR100top-1 accuracy: 83.54 -> 86.13Improved
insert ASR multiple times (delta from 1 to 2, then 4) (ablation)ResNet164 ASR (SE) · CIFAR100top-1 accuracy: 75.36 -> 75.87 -> 75.22Mixed
make input psi learnable rather than fixed constant or random (ablation)ASR (SE) · CIFAR100top-1 accuracy: 75.18 -> 75.36Improved
add ASR under constant noise injected into BN layersResNet164 · CIFAR100 noise (0.5,0.5)top-1 accuracy: 35.77 -> 72.18Improved
use ASR with constant-vector attention in object detection and segmentationMask R-CNN ResNet50 · COCO2017 segmentation APAP: 35.2 -> 35.1No clear effect

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 0 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