AI Paper Atlas

Research map / Architectures and efficiency / CNN design

Cross-channel Communication Networks

Jianwei Yang, Zhile Ren, Chuang Gan, Hongyuan Zhu, Devi Parikh

NeurIPS 2019 · Architectures and efficiency › CNN design · approach: Efficient CNN scaling and dynamic inference · 17 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 C3 blockResNet-110 · CIFAR-100accuracy (%): 72.01 -> 73.36Improved
add C3 blockResNet-20 · CIFAR-100accuracy (%): 67.73 -> 69.34Improved
add C3 blockResNet-101 · ImageNet valtop-1 error (%): 22.48 -> 21.93Improved
add C3 blockResNet-50 · ImageNet valtop-1 error (%): 23.61 -> 23.19Mixed
add C3 blockFaster R-CNN · COCO detectionmAP@0.5:0.95: 33.9 -> 34.8Improved
add C3 blockDeeplabv2 · PASCAL VOC 2012 segmentationmean IoU: 75.2 -> 75.7Improved
C3 vs SE blockWide-ResNet · CIFAR-100accuracy (%): 78.57 vs 78.34No clear effect
add message passing to encoder/decoder (ablation)C3 block · CIFAR-100accuracy (%): 72.65 -> 73.36Improved
add encoder/decoder to message passing (ablation)C3 block · CIFAR-100accuracy (%): 72.74 -> 73.36Improved
insert C3 at residual layer 2 instead of layer 1 (ablation)ResNet-110 · CIFAR-100accuracy (%): 72.28 -> 72.78Improved
add C3 block to ResNet-74 to match ResNet-110 baseline (ablation)ResNet-74 · CIFAR-100accuracy (%)Improved

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