Research map / Architectures and efficiency / Attention and transformers
SG-Former: Self-guided Transformer with Evolving Token Reallocation
Sucheng Ren, Xingyi Yang, Songhua Liu, Xinchao Wang
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 changed | Where | Measured | Effect |
|---|---|---|---|
| use hybrid-scale attention instead of single-scale local attention (ablation) | SG-Former · ImageNet-1K val | Top-1 accuracy (%): 81.6 -> 82.6 | Improved |
| use hybrid-scale attention instead of single-scale global attention (ablation) | SG-Former · ImageNet-1K val | Top-1 accuracy (%): 82.0 -> 82.6 | Improved |
| use learned significance map for token reallocation instead of uniform manual map (ablation) | SG-Former · ImageNet-1K val | Top-1 accuracy (%): 82.3 -> 83.2 | Improved |
| use hybrid-scale significance instead of local-only significance (ablation) | SG-Former · ImageNet-1K val | Top-1 accuracy (%): 82.9 -> 83.2 | Improved |
| use global-only significance instead of manual uniform map (ablation) | SG-Former · ImageNet-1K val | Top-1 accuracy (%): 82.3 -> 82.4 | No clear effect |
| SG-Former-S versus Swin-T at 224 resolution | backbone · ImageNet-1K val | Top-1 accuracy (%): 81.2 -> 83.2 | Improved |
| SG-Former-B versus CSWin-B at 224 resolution | backbone · ImageNet-1K val | Top-1 accuracy (%): 84.2 -> 84.7 | Improved |
| SG-Former-S versus CSWin-T on COCO Mask R-CNN 1x | backbone · COCO Mask R-CNN 1x (small) | box AP: 46.7 -> 47.4 | Improved |
| SG-Former-B versus Swin-B on ADE20K UperNet | backbone · ADE20K UperNet 160k (base) | mIoU: 48.1 -> 52.0 | Improved |
Closest papers by meaning
- Shunted Self-Attention via Multi-Scale Token Aggregation (CVPR 2022 · similarity 0.90)
- Focal Attention for Long-Range Interactions in Vision Transformers (NeurIPS 2021 · similarity 0.90)
- Token Selection is a Simple Booster for Vision Transformers (TPAMI 2022 · similarity 0.90)
- Vision Transformers Need Registers (ICLR 2024 · similarity 0.89)
- Skip-Attention: Improving Vision Transformers by Paying Less Attention (ICLR 2024 · similarity 0.89)
- You Only Need Less Attention at Each Stage in Vision Transformers (CVPR 2024 · similarity 0.89)
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.
