Research map / Label-efficient and robust learning / Recognition and long-tail
DTFD-MIL: Double-Tier Feature Distillation Multiple Instance Learning for Histopathology Whole Slide Image Classification
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 |
|---|---|---|---|
| Use DTFD-MIL with aggregated feature selection instead of Trans-MIL | WSI classification · CAMELYON-16 test | AUC: 0.906 -> 0.946 | Helped |
| Use DTFD-MIL with MaxMinS instead of Trans-MIL | WSI classification · TCGA lung cancer | AUC: 0.949 -> 0.961 | Helped |
| Add pseudo-bags and a second-tier MIL over distilled features | DTFD-MIL · CAMELYON-16 | AUC | Helped |
| Distill the highest- and lowest-probability instances instead of only the highest-probability instance (ablation) | DTFD-MIL · CAMELYON-16 test | AUC: 0.907 -> 0.941 | Helped |
| Distill the aggregated feature instead of only the highest-probability instance (ablation) | DTFD-MIL · CAMELYON-16 test | AUC: 0.907 -> 0.946 | Helped |
| Distill the highest-attention instance instead of only the highest-probability instance (ablation) | DTFD-MIL · CAMELYON-16 test | AUC: 0.907 -> 0.945 | Helped |
| Distill the highest- and lowest-probability instances instead of only the highest-probability instance (ablation) | DTFD-MIL · TCGA lung cancer | AUC: 0.919 -> 0.961 | Helped |
| Distill the aggregated feature instead of only the highest-probability instance (ablation) | DTFD-MIL · TCGA lung cancer | AUC: 0.919 -> 0.951 | Helped |
| Distill the highest-attention instance instead of only the highest-probability instance (ablation) | DTFD-MIL · TCGA lung cancer | AUC: 0.919 -> 0.955 | Helped |
| Increase the number of pseudo-bags beyond 3 (ablation) | Tier-1 MIL · CAMELYON-16 | AUC | Hurt |
| Use MaxS feature distillation rather than another tested distillation strategy (ablation) | DTFD-MIL Tier-2 · TCGA lung cancer | AUC | Hurt |
Closest papers by meaning
- Dynamic Policy-Driven Adaptive Multi-Instance Learning for Whole Slide Image Classification (CVPR 2024 · similarity 0.90)
- CaMIL: Causal Multiple Instance Learning for Whole Slide Image Classification (AAAI 2024 · similarity 0.90)
- ViLa-MIL: Dual-scale Vision-Language Multiple Instance Learning for Whole Slide Image Classification (CVPR 2024 · similarity 0.89)
- How Effective Can Dropout Be in Multiple Instance Learning ? (ICML 2025 · similarity 0.89)
- Fast and Accurate Gigapixel Pathological Image Classification with Hierarchical Distillation Multi-Instance Learning (CVPR 2025 · similarity 0.88)
- Queryable Prototype Multiple Instance Learning with Vision-Language Models for Incremental Whole Slide Image Classification (AAAI 2025 · similarity 0.88)
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.