Research map / Learning theory and algorithms / Clustering
Toward a Multi-View Brain Network Foundation Model: Cross-View Consistency Learning across Arbitrary Atlases
Jiaqing Xu, Jingying Ma, Xin Lin, Yuxiao Liu, Kai He, Qika Lin, Yiping Ke, Yang Li, Dinggang G. Shen, Mengling Feng
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 pretraining objectives (rec, cc, ent) (ablation) | MV-BrainFM · ABIDE (Schaefer100) | AUC: 69.72 -> 76.26 | Improved |
| add pretraining objectives (rec, cc, ent) (ablation) | MV-BrainFM · ADNI (Schaefer100) | AUC: 64.23 -> 69.40 | Improved |
| add atlas encoder (ablation) | MV-BrainFM · ABIDE (Schaefer100) | AUC: 74.38 -> 76.26 | Improved |
| add distance bias (ablation) | MV-BrainFM · ABIDE (Schaefer100) | AUC: 75.27 -> 76.26 | Improved |
| add reconstruction loss (ablation) | MV-BrainFM · ABIDE (Schaefer100) | AUC: 72.79 -> 76.26 | Improved |
| add clustering consistency loss (ablation) | MV-BrainFM · ABIDE (Schaefer100) | AUC: 73.22 -> 76.26 | Improved |
| add entropy loss (ablation) | MV-BrainFM · ADNI (Schaefer100) | AUC: 64.54 -> 69.40 | Improved |
| combine four atlases as multi-atlas input instead of single atlas | MV-BrainFM · ABIDE | AUC: 74.98 -> 77.93 | Improved |
| combine four atlases as multi-atlas input instead of single atlas | MV-BrainFM · ADNI | AUC: 70.02 -> 72.63 | Improved |
| increase hidden dimension from 128 to 1024 (0.95M to 31.35M params) (ablation) | MV-BrainFM · six datasets, Schaefer100 | AUC | Mixed |
| increase attention heads from 8 to 16 (ablation) | MV-BrainFM · six datasets, Schaefer100 | AUC | No clear effect |
| increase prototypes from 16 to 32 (ablation) | MV-BrainFM · six datasets, Schaefer100 | AUC | No clear effect |
Closest papers by meaning
- A Brain Graph Foundation Model: Pre-Training and Prompt-Tuning across Broad Atlases and Disorders (ICLR 2026 · similarity 0.90)
- DCA: Graph-Guided Deep Embedding Clustering for Brain Atlases (NeurIPS 2025 · similarity 0.86)
- Brain Harmony: A Multimodal Foundation Model Unifying Morphology and Function into 1D Tokens (NeurIPS 2025 · similarity 0.86)
- MoRE-Brain: Routed Mixture of Experts for Interpretable and Generalizable Cross-Subject fMRI Visual Decoding (NeurIPS 2025 · similarity 0.85)
- A Convolutional Neural Network Interpretable Framework for Human Ventral Visual Pathway Representation (AAAI 2024 · similarity 0.85)
- Beyond Pairwise Connections: Extracting High-Order Functional Brain Network Structures under Global Constraints (NeurIPS 2025 · 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.
