Research map / Label-efficient and robust learning / Knowledge distillation
Momentum Memory for Knowledge Distillation in Computational Pathology
Yongxin Guo, Hao Lu, Onur C. Koyun, Zhengjie Zhu, Muhammet F. Demir, Metin N. Gurcan
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 momentum-updated memory instead of fixed memory (ablation) | MoMKD · HER2 | AUC (%): 75.2 -> 79.6 | Improved |
| use momentum-updated memory instead of fixed memory (ablation) | MoMKD · In-house ODX | AUC (%): 73.5 -> 79.4 | Improved |
| use momentum-updated memory instead of fixed memory (ablation) | MoMKD · TCGA-BRCA ODX | AUC (%): 81.9 -> 82.3 | No clear effect |
| add omics reconstruction task (ablation) | MoMKD · TCGA-BRCA HER2 | AUC (%): 78.0 -> 79.6 | Improved |
| add omics alignment alongside WSI alignment (ablation) | MoMKD · TCGA-BRCA HER2 | AUC (%): 75.2 -> 79.6 | Improved |
| add WSI-only memory alignment (no omics alignment) (ablation) | WSI baseline · TCGA-BRCA HER2 | AUC (%): 73.9 -> 75.2 | Improved |
| add omics alignment only (no WSI alignment) (ablation) | WSI baseline · TCGA-BRCA HER2 | AUC (%): 73.9 -> 75.7 | Improved |
| use MoMKD instead of best WSI-only MIL (WIKG) | WSI-only MIL · TCGA-BRCA HER2 | AUC (%): 75.5 -> 79.6 | Improved |
| use MoMKD instead of multimodal KD baseline MKD | MKD · TCGA-BRCA HER2 | ACC (%): 77.2 -> 77.9 | No clear effect |
| evaluate MoMKD on external in-house cohort | MoMKD · In-house ODX | AUC (%): 76.5 (TDC) -> 79.4 | Improved |
Closest papers by meaning
- Memory-Efficient Prompt Tuning for Incremental Histopathology Classification (AAAI 2024 · similarity 0.87)
- Quilt-LLaVA: Visual Instruction Tuning by Extracting Localized Narratives from Open-Source Histopathology Videos (CVPR 2024 · similarity 0.86)
- xMIL: Insightful Explanations for Multiple Instance Learning in Histopathology (NeurIPS 2024 · similarity 0.86)
- Sequential Attention-based Sampling for Histopathological Analysis (NeurIPS 2025 · similarity 0.85)
- Dynamic Entity-Masked Graph Diffusion Model for Histopathology Image Representation Learning (AAAI 2025 · similarity 0.85)
- Modeling Dense Multimodal Interactions Between Biological Pathways and Histology for Survival Prediction (CVPR 2024 · similarity 0.85)
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.
