A per-timestep conditioned diffusion transformer generates realistic fMRI dynamics for unseen cognitive tasks by injecting compositional language and optional spatial priors in-context.
13 A Dataset Details We combine four datasets that differ in subject count, task diversity, and recording structure
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.LG 1years
2026 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
Flow Matching with In-Context Priors for Out-of-Distribution Brain Dynamics
A per-timestep conditioned diffusion transformer generates realistic fMRI dynamics for unseen cognitive tasks by injecting compositional language and optional spatial priors in-context.