Research map / Vision-language and multimodal / Human and embodied modeling
Attention Branch Network: Learning of Attention Mechanism for Visual Explanation
Check your idea against 108,967 papers
AI-extracted summary. Written by AI from the paper's title only (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 changed | Where | Measured | Effect |
|---|---|---|---|
| Add an attention branch to ResNeXt | ResNeXt · CIFAR10 | top-1 error: 3.84 -> 3.80 | Helped |
| Add an attention branch to ResNeXt | ResNeXt · CIFAR100 | top-1 error: 18.32 -> 17.70 | Helped |
| Add an attention branch to ResNet | ResNet · SVHN | top-1 error: 2.18 -> 1.86 | Helped |
| Add an attention branch to SENet | SENet · ImageNet | top-1 error: 22.19 -> 20.77 | Helped |
| Add an attention branch to ResNet101 | ResNet101 · CompCars model recognition | accuracy: 90.2 -> 97.1 | Helped |
| Add an attention branch to ResNet101 | ResNet101 · CompCars maker recognition | accuracy: 90.1 -> 98.1 | Helped |
| Add task-specific attention maps and an attention branch | ResNet101 multi-task facial attribute model · CelebA | average attribute accuracy: 90.69 -> 91.07 | Helped |
| Add multiplicative attention g(x)·M(x) (ablation) | ResNet20 · CIFAR100 | top-1 error: 31.47 -> 30.61 | Helped |
| Add residual-style attention g(x)·(1 + M(x)) instead of multiplicative attention (ablation) | ResNet20 · CIFAR100 | top-1 error: 30.61 -> 30.46 | Helped |
| Add multiplicative attention g(x)·M(x) (ablation) | ResNet32 · CIFAR100 | top-1 error: 30.13 -> 28.34 | Helped |
| Add residual-style attention g(x)·(1 + M(x)) instead of multiplicative attention (ablation) | ResNet44 · CIFAR100 | top-1 error: 24.83 -> 25.59 | Hurt |
| Add multiplicative attention g(x)·M(x) (ablation) | ResNet44 · CIFAR100 | top-1 error: 25.90 -> 24.83 | Helped |
Closest papers by meaning
- One Explanation is Not Enough: Structured Attention Graphs for Image Classification (NeurIPS 2021 · similarity 0.88)
- Residual Attention Network for Image Classification (CVPR 2017 · similarity 0.87)
- Learn to Pay Attention (ICLR 2018 · similarity 0.87)
- Learning what and where to attend (ICLR 2019 · similarity 0.87)
- Show and Tell: Visually Explainable Deep Neural Nets via Spatially-Aware Concept Bottleneck Models (CVPR 2025 · similarity 0.85)
- EAN: An Efficient Attention Module Guided by Normalization for Deep Neural Networks (AAAI 2024 · similarity 0.85)
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.