A latent world model of the brain, trained to predict the next latent fMRI state from past brain states and current movie stimuli, outperforms regression-based encoders in causal multi-step rollout on three naturalistic movie-fMRI benchmarks.
Brainvista: Modeling naturalistic brain dynamics as multimodal next-token prediction,
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
citation-role summary
contradiction 1
citation-polarity summary
fields
q-bio.NC 1years
2026 1verdicts
CONDITIONAL 1roles
contradiction 1polarities
contest 1representative citing papers
citing papers explorer
-
NeuroWorld: A Latent Brain World Model for Stimulus-Conditioned Human Brain Dynamics
A latent world model of the brain, trained to predict the next latent fMRI state from past brain states and current movie stimuli, outperforms regression-based encoders in causal multi-step rollout on three naturalistic movie-fMRI benchmarks.