RECORD
SLPD: Slide-level Prototypical Distillation for WSIs
Computer Vision and Pattern Recognition · Submitted 20 Jul 2023
abstract compiled · 2026-09-23 · no language-model review
The authors' main claim SLPD achieves state-of-the-art results on multiple slide-level benchmarks and demonstrates that representation learning of semantic structures of slides can make a suitable proxy task for WSI analysis.
The compiled record
The paper's own sentences, arranged. Each label is drawn from a closed vocabulary; the number beside it is the probability the classifier gave that label. Numbers in the tables are the authors', checked by code. No sentence in this record was written by a model.
Open the compiled record
The abstract, sentence by sentence
Improving the feature representation ability is the foundation of many whole slide pathological image (WSIs) tasks. Recent works have achieved great success in pathological-specific self-supervised learning (SSL). However, most of them only focus on learning patch-level representations, thus there is still a gap between pretext and slide-level downstream tasks, e.g., subtyping, grading and staging. Aiming towards slide-level representations, we propose Slide-Level Prototypical Distillation (SLPD) to explore intra- and inter-slide semantic structures for context modeling on WSIs. Specifically, we iteratively perform intra-slide clustering for the regions (4096x4096 patches) within each WSI to yield the prototypes and encourage the region representations to be closer to the assigned prototypes. By representing each slide with its prototypes, we further select similar slides by the set distance of prototypes and assign the regions by cross-slide prototypes for distillation. SLPD achieves state-of-the-art results on multiple slide-level benchmarks and demonstrates that representation learning of semantic structures of slides can make a suitable proxy task for WSI analysis. Code will be available at https://github.com/Carboxy/SLPD.
background contribution method result interpretation main claim
| # | sentence | role | the authors' own work | states the main conclusion | strength |
|---|---|---|---|---|---|
| s0 | Improving the feature representation ability is the foundation of many whole slide pathological image (WSIs) tasks. | background 1.00 | 0.05 | 0.02 | asserted 1.00 |
| s1 | Recent works have achieved great success in pathological-specific self-supervised learning (SSL). | background 1.00 | 0.04 | 0.02 | asserted 1.00 |
| s2 | However, most of them only focus on learning patch-level representations, thus there is still a gap between pretext and slide-level downstream tasks, e.g., subtyping, grading and staging. | gap 1.00 | 0.06 | 0.03 | asserted 0.98 |
| s3 | Aiming towards slide-level representations, we propose Slide-Level Prototypical Distillation (SLPD) to explore intra- and inter-slide semantic structures for context modeling on WSIs. | contribution 0.99 | 0.86 | 0.06 | asserted 0.86 |
| s4 | Specifically, we iteratively perform intra-slide clustering for the regions (4096x4096 patches) within each WSI to yield the prototypes and encourage the region representations to be closer to the assigned prototypes. | method 1.00 | 0.94 | 0.03 | asserted 0.93 |
| s5 | By representing each slide with its prototypes, we further select similar slides by the set distance of prototypes and assign the regions by cross-slide prototypes for distillation. | method 1.00 | 0.91 | 0.04 | asserted 0.89 |
| s6 | SLPD achieves state-of-the-art results on multiple slide-level benchmarks and demonstrates that representation learning of semantic structures of slides can make a suitable proxy task for WSI analysis. | result 0.94 | 0.96 | 0.90 | asserted 1.00 |
| s7 | Code will be available at https://github.com/Carboxy/SLPD. | future work 0.85 | 0.19 | 0.02 | asserted 0.63 |
What was examined, and what was not
- arXiv listing abstract v1
- full text
- figures and tables
- references and citing sentences
- relations to other papers
- P abstract
- S sections
- T body sentences
- C citations
- R cross-paper relations
The paper
Original abstract
Improving the feature representation ability is the foundation of many whole slide pathological image (WSIs) tasks. Recent works have achieved great success in pathological-specific self-supervised learning (SSL). However, most of them only focus on learning patch-level representations, thus there is still a gap between pretext and slide-level downstream tasks, e.g., subtyping, grading and staging. Aiming towards slide-level representations, we propose Slide-Level Prototypical Distillation (SLPD) to explore intra- and inter-slide semantic structures for context modeling on WSIs. Specifically, we iteratively perform intra-slide clustering for the regions (4096x4096 patches) within each WSI to yield the prototypes and encourage the region representations to be closer to the assigned prototypes. By representing each slide with its prototypes, we further select similar slides by the set distance of prototypes and assign the regions by cross-slide prototypes for distillation. SLPD achieves state-of-the-art results on multiple slide-level benchmarks and demonstrates that representation learning of semantic structures of slides can make a suitable proxy task for WSI analysis. Code will be available at https://github.com/Carboxy/SLPD.
Human review
Discussion
Know something about this paper? Leave a comment, or highlight any passage above to comment on it.