Research map / Label-efficient and robust learning / Recognition and long-tail
Dual-Level Hypergraph Generation for Addressing Feature Scarcity in Whole-Slide Image Classification
Shuilian Yao, Qi Jia, Yu Liu, Pengshuo Zhang, Lili Sun, Weimin Wang, Yanmei Zhu, Bo Zhang, Xin Fan
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 prompt-guided hypergraph VAE (HGVAE) generation of minority-class hypergraphs (ablation) | hypergraph baseline · NIMM (overall) | F1-Score: 73.39 -> 76.51 | Improved |
| add anchor-diffusion node mixup (ablation) | hypergraph baseline · NIMM (overall) | F1-Score: 73.39 -> 75.29 | Improved |
| combine HGVAE and anchor-diffusion mixup (ablation) | hypergraph baseline · NIMM (overall) | F1-Score: 73.39 -> 77.19 | Improved |
| add both modules (ablation) | hypergraph baseline · NIMM ITC / micrometastasis | F1-Score: ITC 51.29 -> 57.13; Micro 49.95 -> 58.30 | Improved |
| add both modules (ablation) | hypergraph baseline · TCGA-NSCLC (2-shot) | F1-Score: 74.59 -> 80.96 | Improved |
| use class-agnostic vanilla VAE generation instead of none (ablation) | hypergraph baseline · NIMM / TCGA 2-shot | F1-Score: NIMM overall F1 73.59 -> 68.26 (Table 4 baseline) | Hurt |
| add class conditioning to the VAE (ablation) | VAE generator · NIMM / TCGA | AUC and F1: NIMM overall F1 68.26 -> 73.98 | Improved |
| replace class embeddings with learnable class text prompts (ablation) | Conditional VAE · NIMM | F1-Score (ITC): +2.2% ITC F1 (52.42 -> 54.62) | Improved |
| add topology-based sub-hypergraph generation (ablation) | Conditional VAE with prompts · NIMM micrometastasis | F1-Score: +2.72% micrometastasis F1 | Improved |
| use the full multi-hypergraph (feature plus topology) instead of feature-only or topology-only sub-hypergraph (ablation) | Dual-HGNet · NIMM | F1-Score | Improved |
| compare against TransMIL at 8-shot | Dual-HGNet · TCGA-RCC (8-shot) | AUC: 97.85 vs 97.97 | No clear effect |
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 0 of 1 on most shared results.
- TransMIL — TransMIL: Transformer based Correlated Multiple Instance Learning for Whole Slide Image Classification (NeurIPS 2021) · better on 1 of 2 shared results
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
- MAPLE: Multi-scale Attribute-enhanced Prompt Learning for Few-shot Whole Slide Image Classification (NeurIPS 2025 · similarity 0.86)
- Dynamic Entity-Masked Graph Diffusion Model for Histopathology Image Representation Learning (AAAI 2025 · similarity 0.85)
- Cell Graph Transformer for Nuclei Classification (AAAI 2024 · similarity 0.85)
- Sequential Attention-based Sampling for Histopathological Analysis (NeurIPS 2025 · similarity 0.85)
- Dynamic Graph Representation with Knowledge-aware Attention for Histopathology Whole Slide Image Analysis (CVPR 2024 · similarity 0.85)
- FOCUS: Knowledge-enhanced Adaptive Visual Compression for Few-shot Whole Slide Image Classification (CVPR 2025 · 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.
