AI Paper Atlas

Research map / Architectures and efficiency / Attention and transformers

Scaling Vision Transformers for Functional MRI with Flat Maps

Connor Lane, Mihir Tripathy, Leema Krishna Murali, Ratna Sagari Grandhi, Shamus Sim Zi Yang, Sam Gijsen, Debojyoti Das, Manish Ram, Utkarsh Kumar Singh, Cesar Kadir Torrico Villanueva, YUXIANG WEI, Will Beddow and 6 more

ICML 2026 · Architectures and efficiency › Attention and transformers · approach: Token merging and masked pretraining · 0 citations (OpenAlex)

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 changedWhereMeasuredEffect
use flat map representation instead of parcellationCortexMAE · HCP-A Sexprobe accuracy: 71.2 -> 87.4Improved
use flat map representation instead of parcellationCortexMAE · HCP-YA Task21probe accuracy: 97.5 -> 98.9Improved
use flat map representation instead of parcellationCortexMAE · NSD COCO24probe accuracy: 27.5 -> 31.0Improved
use volume representation instead of flat mapCortexMAE · HCP-A Ageprobe accuracy: 47.5 -> 53.4Improved
use flat map representation instead of parcellationCortexMAE · ABIDE, ADHD200, ADNI, PPMIprobe accuracy: 61.4 vs 62.0 on ABIDENo clear effect
add tube masking instead of uniform masking (ablation)CortexMAE flat · NSD COCO24 (ratio 90)accuracy: 29.8 -> 31.4Improved
add TR scaling augmentation (ablation)CortexMAE flat · NSD COCO24accuracy: 30.4 -> 31.7No clear effect
add strong random crop augmentation (ablation)CortexMAE flat · NSD COCO24accuracy: 30.4 -> 24.4Hurt
use PC norm with 2 components as target (ablation)CortexMAE flat · NSD COCO24accuracy: 29.5 -> 32.1No clear effect
use PC norm with 8 components as target (ablation)CortexMAE flat · NSD COCO24accuracy: 29.5 -> 24.6Hurt
reduce temporal patch size from 4 to 2 (ablation)CortexMAE flat · NSD COCO24accuracy: 29.6 -> 32.9Improved
add subcortical structures to parcellation (Schaefer-400 + Tian S3) (ablation)CortexMAE parcel · NSD COCO24accuracy: 27.5 -> 27.2No clear effect

Closest papers by meaning

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.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy