AI Paper Atlas

Research map / Label-efficient and robust learning / Self-supervised and contrastive

Omni-fMRI: A Universal Atlas-Free fMRI Foundation Model

Mo Wang, Wenhao Ye, Junfeng Xia, Junxiang Zhang, Xuanye Pan, Minghao Xu, Haotian Deng, Hongkai Wen, Quanying Liu

ICML 2026 · Label-efficient and robust learning › Self-supervised and contrastive · approach: Self-supervised neuroimaging representations · 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 intensity variance instead of Shannon entropy for patch complexity (ablation)Omni-fMRI · HCPaccuracy: 78.43 -> 85.18Improved
use intensity variance instead of Laplacian response for patch complexity (ablation)Omni-fMRI · ADNI-ADaccuracy: 68.40 -> 75.69Improved
use intensity variance instead of MSE complexity (ablation)Omni-fMRI · HCPaccuracy: 78.37 -> 85.18Improved
omit per-patch normalization in MAE pretraining (ablation)Omni-fMRI · ADNI-ADaccuracy: 80.21 -> 84.72Improved
omit per-patch normalization in MAE pretraining (ablation)Omni-fMRI · HCPaccuracy: 86.61 -> 87.05Improved
raise patch subdivision threshold from 0.25 to 0.3 (ablation)Omni-fMRI · ABCDaccuracy: 67.29 -> 64.58Hurt
raise patch subdivision threshold from 0.25 to 0.3 (ablation)Omni-fMRI · HCPaccuracy: 85.10 -> 85.10No clear effect
lower threshold from 0.25 to 0.1 (more tokens) (ablation)Omni-fMRI · HCPaccuracy: 85.10 -> 85.77No clear effect
use no dynamic patching (threshold 0, 14K tokens) (ablation)Omni-fMRI · ABCD/HCPmemoryHurt
scale model size and pretraining data (ablation)Omni-fMRI · ADNI (AD) and 3 other tasksaccuracyImproved
replace NeuroSTORM with Omni-fMRI under linear probingOmni-fMRI · ADNI (AD)accuracy drop from fine-tuning: 3% versus 13%Improved
use Omni-fMRI instead of NeuroSTORMOmni-fMRI · NSD 100-way retrievalTop-1 accuracy (%): 3.00 -> 6.93Improved

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