English static mirror for SEO/GEO · AI-assisted translation · Read Chinese original

BrainTaskonomy: Learning How to Pretrain and What to Transfer in fMRI Foundation Models

Forum topic · 小凯 · 2026-09-11

Summary

BrainTaskonomy (arXiv:2609.10518) proposes organizing fMRI foundation model pretraining and downstream adaptation through measured learning relations, without modifying the backbone. During pretraining, a lightweight Brain-DiT proxy estimates difficulty and directed facilitation across ten fMRI domains, producing a priority-guided cumulative domain curriculum combined with high-to-low-noise timestep scheduling and joint consolidation. During adaptation, controlled first- and higher-order transfer analysis across fifteen tasks builds a directed taskonomy, from which budgeted integer programming (BIP) selects directly supervised source tasks and target-specific transfer routes. The joint priority-domain and high-to-low-timestep curriculum reduces v-NMSE, PSD-NMSE, and FC-MSE by 6.5%, 16.3%, and 10.5% respectively versus uniform sampling, with strong downstream performance on six in-domain and out-of-domain tasks. The taskonomy reveals asymmetric, target-dependent transfer, and sealed-test evaluation shows larger gains for BIP policies when higher-order route spaces are available.

BrainTaskonomy: Learning How to Pretrain and What to Transfer in fMRI Foundation Models

Field: Computer Vision / Neuroimaging Authors: Junfeng Xia, Wenhao Ye, Junxiang Zhang, Jiayu Zuo, Mo Wang, Quanying Liu Published: 2026-09-09 arXiv: 2609.10518

Abstract

fMRI foundation models increasingly aggregate heterogeneous data across brain states, cohorts, and acquisition settings, yet pretraining domains are commonly treated as a flat mixture and downstream tasks are adapted independently. This work studies whether measured learning relations can organize both stages without modifying the backbone.

Method

Pretraining stage

  • A lightweight Brain-DiT proxy estimates difficulty and directed facilitation across ten fMRI domains.
  • These estimates yield a priority-guided cumulative domain curriculum.
  • Combined with high-to-low-noise timestep scheduling and joint consolidation.
  • Adaptation stage

  • Controlled first- and higher-order transfer across fifteen tasks constructs a directed taskonomy.
  • Budgeted integer programming (BIP) selects directly supervised source tasks and target-specific transfer routes.
  • Results

  • The joint priority-domain and high-to-low-timestep curriculum reduces v-NMSE, PSD-NMSE, and FC-MSE by 6.5%, 16.3%, and 10.5%, respectively, relative to uniform sampling over both dimensions.
  • Strong downstream performance across six in-domain and out-of-domain tasks.
  • The taskonomy reveals asymmetric, target-dependent transfer.
  • Exploratory sealed-test evaluation shows larger descriptive gains for BIP policies when higher-order route spaces are available, compared to matched random controls.

Conclusion

Organizing fMRI pretraining and adaptation by measured learning relations outperforms treating domains and tasks as independent flat sets.

---

*Auto-collected on 2026-09-11.*

Tags

#fmri#foundation-models#pretraining#taskonomy#transfer-learning#curriculum-learning#neuroimaging#arxiv

This page is an English static mirror generated for search and AI citation. It may be a full translation or structured summary of the Chinese original. Canonical interactive discussion lives on the Chinese page: https://zhichai.net/topic/178634715