{"ID":2835332,"CreatedAt":"2026-06-01T04:54:23.091178241Z","UpdatedAt":"2026-06-01T04:54:23.091178241Z","DeletedAt":null,"paper_url":"https://arxiv.org/abs/2511.22870","arxiv_id":"2511.22870","title":"Scalable Diffusion Transformer for Conditional 4D fMRI Synthesis","abstract":"Generating whole-brain 4D fMRI sequences conditioned on cognitive tasks remains challenging due to the high-dimensional, heterogeneous BOLD dynamics across subjects/acquisitions and the lack of neuroscience-grounded validation. We introduce the first diffusion transformer for voxelwise 4D fMRI conditional generation, combining 3D VQ-GAN latent compression with a CNN-Transformer backbone and strong task conditioning via AdaLN-Zero and cross-attention. On HCP task fMRI, our model reproduces task-evoked activation maps, preserves the inter-task representational structure observed in real data (RSA), achieves perfect condition specificity, and aligns ROI time-courses with canonical hemodynamic responses. Performance improves predictably with scale, reaching task-evoked map correlation of 0.83 and RSA of 0.98, consistently surpassing a U-Net baseline on all metrics. By coupling latent diffusion with a scalable backbone and strong conditioning, this work establishes a practical path to conditional 4D fMRI synthesis, paving the way for future applications such as virtual experiments, cross-site harmonization, and principled augmentation for downstream neuroimaging models.","short_abstract":"Generating whole-brain 4D fMRI sequences conditioned on cognitive tasks remains challenging due to the high-dimensional, heterogeneous BOLD dynamics across subjects/acquisitions and the lack of neuroscience-grounded validation. We introduce the first diffusion transformer for voxelwise 4D fMRI conditional generation, c...","url_abs":"https://arxiv.org/abs/2511.22870","url_pdf":"https://arxiv.org/pdf/2511.22870v1","authors":"[\"Jungwoo Seo\",\"David Keetae Park\",\"Shinjae Yoo\",\"Jiook Cha\"]","published":"2025-11-28T04:18:11Z","proceeding":"cs.CV","tasks":"[\"cs.CV\",\"q-bio.NC\"]","methods":"[\"Diffusion Model\",\"Transformer\",\"Generative Adversarial Network\",\"Convolutional Neural Network\"]","has_code":false}