Research map / Vision-language and multimodal / Cross-modal learning
Anchor-driven Multi-modal Multi-scale Expert Selection for Survival Prediction
Tao Zhou, Ying Hu, Huazhu Fu, Yi Zhou, Xiao-Jun Wu, Haibin Ling
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 genomic to histology via concatenation (ablation) | ABMIL · BRCA, LUAD, UCEC avg | C-index: 0.620 -> 0.650 | Improved |
| use three scales instead of single 20x scale (ablation) | AM2ES · BRCA, LUAD, UCEC avg | C-index: 0.683 -> 0.714 | Improved |
| use dual scale (add 5x) to single scale (ablation) | AM2ES · BRCA, LUAD, UCEC avg | C-index: 0.683 -> 0.687 | Improved |
| use anchor fusion instead of cross-attention (ablation) | AM2ES · BRCA, LUAD, UCEC avg | C-index: 0.689 -> 0.714 | Improved |
| use anchor fusion instead of bilinear pooling (ablation) | AM2ES · BRCA, LUAD, UCEC avg | C-index: 0.682 -> 0.714 | Improved |
| add top-K selection instead of global average pooling (ablation) | AM2ES · BRCA, LUAD, UCEC avg | C-index: 0.686 -> 0.714 | Improved |
| add top-K selection instead of attention pooling (ablation) | AM2ES · BRCA, LUAD, UCEC avg | C-index: 0.691 -> 0.714 | Improved |
| add MoE routing instead of linear concatenation (ablation) | AM2ES · BRCA, LUAD, UCEC avg | C-index: 0.687 -> 0.714 | Improved |
| add rank-based alignment instead of no alignment (ablation) | AM2ES · BRCA, LUAD, UCEC avg | C-index: 0.688 -> 0.714 | Improved |
| add anchor regularization loss (ablation) | AM2ES · BRCA, LUAD, UCEC avg | C-index: 0.700 -> 0.714 | Improved |
| increase anchors K from 16 to 32 (ablation) | AM2ES · BRCA, LUAD, UCEC avg | C-index: 0.714 -> 0.700 | Hurt |
| increase retained instances K' from 4 to 16 (ablation) | AM2ES · BRCA, LUAD, UCEC avg | C-index: 0.714 -> 0.694 | Hurt |
Closest papers by meaning
- Modeling Dense Multimodal Interactions Between Biological Pathways and Histology for Survival Prediction (CVPR 2024 · similarity 0.88)
- CA-MLIF: Cross-Attention and Multimodal Low-Rank Interaction Fusion Framework for Tumor Prognostic Prediction (AAAI 2025 · similarity 0.88)
- Multimodal Prototyping for cancer survival prediction (ICML 2024 · similarity 0.88)
- Prototypical Information Bottlenecking and Disentangling for Multimodal Cancer Survival Prediction (ICLR 2024 · similarity 0.88)
- Robust Multimodal Survival Prediction with Conditional Latent Differentiation Variational AutoEncoder (CVPR 2025 · similarity 0.87)
- Interpretable Vision-Language Survival Analysis with Ordinal Inductive Bias for Computational Pathology (ICLR 2025 · similarity 0.87)
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.
