Research map / Vision-language and multimodal / CLIP and zero-shot
Universal-to-Specific: Dynamic Knowledge-Guided Multiple Instance Learning for Few-Shot Whole Slide Image Classification
Junjian Li, Hulin Kuang, Jin Liu, Hailin Yue, Mengshen He, Jianxin 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 |
|---|---|---|---|
| add the structural consistency (SC) loss (ablation) | DyKo · CAMELYON16 (4-shot) | AUC: 0.750 -> 0.871 | Improved |
| add the WSI-adaptive knowledge instantiation (WAKI) module (ablation) | DyKo · CAMELYON16 (4-shot) | AUC: 0.773 -> 0.871 | Improved |
| add dual-stream cross-attention fusion instead of averaging stream scores (ablation) | DyKo · CAMELYON16 (4-shot) | AUC: 0.756 -> 0.871 | Improved |
| add the SC loss (ablation) | DyKo · CAMELYON16 (16-shot) | AUC: 0.954 -> 0.961 | Improved |
| add WAKI (ablation) | DyKo · CAMELYON16 (8-shot) | AUC: 0.925 -> 0.956 | Improved |
| set number of visual prototypes M=10 versus 5 or 15 (ablation) | DyKo · CAMELYON16 | AUC | Mixed |
| select top-10 concepts per prototype versus 5, 15, 20 (ablation) | DyKo · CAMELYON16 | AUC | Improved |
| use more advanced LLM (Claude-3.5-Sonnet, GPT-4.1) for class prompts (ablation) | DyKo · CAMELYON16 (4-shot) | AUC | Improved |
| use TITAN encoder instead of QuiltNet or CONCH (ablation) | DyKo · CAMELYON16 (4-shot) | AUC: 0.871 (TITAN, best) | Improved |
| use Quilt-1M instead of Pathology Outlines as knowledge base (ablation) | DyKo · CAMELYON16 | AUC | Improved |
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.
- ViLaMIL — ViLa-MIL: Dual-scale Vision-Language Multiple Instance Learning for Whole Slide Image Classification (CVPR 2024) · better on 1 of 1 shared result
Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.
Closest papers by meaning
- FOCUS: Knowledge-enhanced Adaptive Visual Compression for Few-shot Whole Slide Image Classification (CVPR 2025 · similarity 0.92)
- Few-Shot Learning from Gigapixel Images via Hierarchical Vision-Language Alignment and Modeling (NeurIPS 2025 · similarity 0.90)
- Generalizable Whole Slide Image Classification with Fine-Grained Visual-Semantic Interaction (CVPR 2024 · similarity 0.89)
- Dynamic Policy-Driven Adaptive Multi-Instance Learning for Whole Slide Image Classification (CVPR 2024 · similarity 0.89)
- PathVQ: Reforming Computational Pathology Foundation Model for Whole Slide Image Analysis via Vector Quantization (NeurIPS 2025 · similarity 0.89)
- Queryable Prototype Multiple Instance Learning with Vision-Language Models for Incremental Whole Slide Image Classification (AAAI 2025 · 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.
