# Revisiting the Ordering of Channel and Spatial Attention: A Comprehensive Study on Sequential and Parallel Designs

arXiv: https://arxiv.org/abs/2601.07310 · HTML: https://mlsanity.com/paper/2601.07310/

## Key facts

- No citing papers in the map yet (data through 7 Oct 2026).
- Builds on 18 of its 44 references that are in the map.
- First posted on arXiv 12 Jan 2026.

## Builds on (most cited first, 18 of 18)

- [Attention Is All You Need](https://mlsanity.com/paper/1706.03762/) (2017)
- [Squeeze-and-Excitation Networks](https://mlsanity.com/paper/1709.01507/) (2017)
- [CBAM: Convolutional Block Attention Module](https://mlsanity.com/paper/1807.06521/) (2018)
- [Dual Attention Network for Scene Segmentation](https://mlsanity.com/paper/1809.02983/) (2018)
- [MedMNIST v2 -- A large-scale lightweight benchmark for 2D and 3D biomedical image classification](https://mlsanity.com/paper/2110.14795/) (2021)
- [ECA-Net: Efficient Channel Attention for Deep Convolutional Neural Networks](https://mlsanity.com/paper/1910.03151/) (2019)
- [Selective Kernel Networks](https://mlsanity.com/paper/1903.06586/) (2019)
- [GCNet: Non-local Networks Meet Squeeze-Excitation Networks and Beyond](https://mlsanity.com/paper/1904.11492/) (2019)
- [Coordinate Attention for Efficient Mobile Network Design](https://mlsanity.com/paper/2103.02907/) (2021)
- [BAM: Bottleneck Attention Module](https://mlsanity.com/paper/1807.06514/) (2018)
- [EMCAD: Efficient Multi-scale Convolutional Attention Decoding for Medical Image Segmentation](https://mlsanity.com/paper/2405.06880/) (2024)
- [Polarized Self-Attention: Towards High-quality Pixel-wise Regression](https://mlsanity.com/paper/2107.00782/) (2021)
- [Spatial Group-wise Enhance: Improving Semantic Feature Learning in Convolutional Networks](https://mlsanity.com/paper/1905.09646/) (2019)
- [SCSA: Exploring the Synergistic Effects Between Spatial and Channel Attention](https://mlsanity.com/paper/2407.05128/) (2024)
- [YOLOv8-ResCBAM: YOLOv8 Based on An Effective Attention Module for Pediatric Wrist Fracture Detection](https://mlsanity.com/paper/2409.18826/) (2024)
- [Rethinking Decoder Design: Improving Biomarker Segmentation Using Depth-to-Space Restoration and Residual Linear Attention](https://mlsanity.com/paper/2506.18335/) (2025)
- [Channel Pruning Guided by Spatial and Channel Attention for DNNs in Intelligent Edge Computing](https://mlsanity.com/paper/2011.03891/) (2020)
- [FANCL: Feature-Guided Attention Network with Curriculum Learning for Brain Metastases Segmentation](https://mlsanity.com/paper/2410.22057/) (2024)

## Abstract (arXiv)

Attention mechanisms have become a core component of deep learning models, with Channel Attention and Spatial Attention being the two most representative architectures. Current research on their fusion strategies primarily bifurcates into sequential and parallel paradigms, yet the selection process remains largely empirical, lacking systematic analysis and unified principles. We systematically compare channel-spatial attention combinations under a unified framework, building an evaluation suite of 18 topologies across four classes: sequential, parallel, multi-scale, and residual. Across two vision and nine medical datasets, we uncover a "data scale-method-performance" coupling law: (1) in few-shot tasks, the "Channel-Multi-scale Spatial" cascaded structure achieves optimal performance; (2) in medium-scale tasks, parallel learnable fusion architectures demonstrate superior results; (3) in large-scale tasks, parallel structures with dynamic gating yield the best performance. Additionally, experiments indicate that the "Spatial-Channel" order is more stable and effective for fine-grained classification, while residual connections mitigate vanishing gradient problems across varying data scales. We thus propose scenario-based guidelines for building future attention modules. Code is open-sourced at https://github.com/DWlzm.