Skip to main content

# Hierarchical Attention Mechanism

2 related articles

Subscribe to updates

arXiv Nov 11, 2026
A neuroimaging model that talks back: Neuro-VL beats GPT-5.4 on 16 of 17 task-group combinations and holds its ground under misleading prompts
Neuro-VL is a vision-language model trained on nearly one million routine clinical neuroimaging studies that can localize, describe, diagnose and report on complete examinations and discuss its findings in natural language. On NeuroBench, built from internal studies and 24 international public datasets, it outperformed GPT-5.4 and Gemma 4 on 16 of 17 task-group combinations and the neuroradiology-specialized NeuroVFM on all 17, leading by 15.9, 16.9 and 23.0 percentage points when averaged equally across combinations. A further-tuned longitudinal version compares prior and current studies, and the model resisted misleading diagnostic or localization prompts more reliably than GPT-5.4.
Neuro-VL is a vision-language model trained on nearly one million routine clinical neuroimaging studies that can localize, describe, diagnose and report on complete examinations and discuss its findings in natural language. On NeuroBench, built from internal studies and 24 international public datasets, it outperformed GPT-5.4 and Gemma 4 on 16 of 17 task-group combinations and the neuroradiology-specialized NeuroVFM on all 17, leading by 15.9, 16.9 and 23.0 percentage points when averaged equally across combinations. A further-tuned longitudinal version compares prior and current studies, and the model resisted misleading diagnostic or localization prompts more reliably than GPT-5.4.
Michigan Medicine
Neuro-VL NeuroBench Hierarchical Attention Mechanism Radiology Report Generation Longitudinal Disease Monitoring
arXiv Sep 25, 2026
BrainHO replaces fixed brain atlases with learnable subgraphs, reaching 69.68% accuracy on ABIDE from PCC input alone while localizing cross-network disease subgraphs
The work proposes Brain Hierarchical Organization Learning (BrainHO), which uses learnable subgraph and graph tokens to aggregate brain regions bottom-up via hierarchical attention driven by node feature affinity, combined with a subgraph orthogonality constraint and a hierarchical consistency constraint; on ABIDE (N=1009, 516 ASD/493 healthy controls) and REST-meta-MDD (N=2380, 1276 MDD/1104 healthy controls) it attains the highest accuracy (69.68% and 64.71%) and sensitivity (73.11% and 67.43%) using only the static PCC connectivity matrix, while visualizing disease-related subgraphs that partly overlap predefined networks such as the SMN and DAN and partly span multiple predefined networks.
The work proposes Brain Hierarchical Organization Learning (BrainHO), which uses learnable subgraph and graph tokens to aggregate brain regions bottom-up via hierarchical attention driven by node feature affinity, combined with a subgraph orthogonality constraint and a hierarchical consistency constraint; on ABIDE (N=1009, 516 ASD/493 healthy controls) and REST-meta-MDD (N=2380, 1276 MDD/1104 healthy controls) it attains the highest accuracy (69.68% and 64.71%) and sensitivity (73.11% and 67.43%) using only the static PCC connectivity matrix, while visualizing disease-related subgraphs that partly overlap predefined networks such as the SMN and DAN and partly span multiple predefined networks.
University of Alberta Shandong Normal University Northeastern University
Jingfeng Tang（First author）Peng Cao（Corresponding author）
BrainHO Hierarchical Attention Mechanism Brain Disorder Diagnosis Brain Network Analysis Functional Magnetic Resonance Imaging

You’re all caught up