Research map / Label-efficient and robust learning / Recognition and long-tail
GMMamba: Group Masking Mamba for Whole Slide Image Classification
Tingting Zheng, Hongxun Yao, Kui Jiang, Yi Xiao, Sicheng Zhao
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 |
|---|---|---|---|
| add location-based grouping to BiMamba with max-pooling (ablation) | GMMamba · TCGA-ESCA | accuracy: 0.833 -> 0.885 | Improved |
| add intra-group masking Mamba (IMM) over unmasked grouping (ablation) | GMMamba · TCGA-ESCA | accuracy: 0.898 -> 0.924 | Improved |
| add cross-group super-feature sampling (CSS) on top of IMM (ablation) | GMMamba · TCGA-ESCA | accuracy: 0.924 -> 0.949 | Improved |
| add IMM on top of CSS (ablation) | GMMamba · TCGA-ESCA | accuracy: 0.936 -> 0.949 | Improved |
| use location-based grouping instead of random grouping (ablation) | GMMamba · TCGA-ESCA | accuracy: 0.898 -> 0.949 | Improved |
| use location-based grouping instead of feature-based clustering (ablation) | GMMamba · TCGA-ESCA | accuracy: 0.911 -> 0.949 | Improved |
| increase group number G from 2 to 10 (ablation) | GMMamba · TCGA-ESCA (Mr=0) | accuracy: 0.860 -> 0.936 | Improved |
| increase group number G from 10 to 20 (ablation) | GMMamba · TCGA-ESCA (Mr=0) | accuracy: 0.936 -> 0.910 | Hurt |
| raise mask ratio Mr from 0 to 20% (ablation) | GMMamba · TCGA-ESCA | accuracy: 0.844 -> 0.938 | Improved |
| raise mask ratio Mr from 0 to 20% (ablation) | GMMamba · TCGA-Lung | accuracy: 0.792 -> 0.807 | Mixed |
| add CSS to ABMIL | ABMIL · TCGA-ESCA | accuracy: 0.827 -> 0.866 | Improved |
| add CSS to DTFD-AFS | DTFD-AFS · TCGA-ESCA | accuracy: 0.872 -> 0.879 | Improved |
Closest papers by meaning
- M3amba: Memory Mamba is All You Need for Whole Slide Image Classification (CVPR 2025 · similarity 0.89)
- 2DMamba: Efficient State Space Model for Image Representation with Applications on Giga-Pixel Whole Slide Image Classification (CVPR 2025 · similarity 0.88)
- GroupMamba: Efficient Group-Based Visual State Space Model (CVPR 2025 · similarity 0.87)
- MobileMamba: Lightweight Multi-Receptive Visual Mamba Network (CVPR 2025 · similarity 0.85)
- S³-Mamba: Small-Size-Sensitive Mamba for Lesion Segmentation (AAAI 2025 · similarity 0.84)
- Dynamic Policy-Driven Adaptive Multi-Instance Learning for Whole Slide Image Classification (CVPR 2024 · similarity 0.84)
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.
