Research map / Vision-language and multimodal / Human and embodied modeling
Mind the State: Towards Unified, Context-Aware EEG-to-fMRI Synthesis
Yamin Li, Shiyu Wang, Chang Li, Ange Lou, Haatef Pourmotabbed, Sarah Goodale, Dario J. Englot, Daniel Moyer, Roza G Bayrak, Catie Chang
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 changed | Where | Measured | Effect |
|---|---|---|---|
| add context prompt tokens (dataset, age, sex, vigilance) (ablation) | UniEFS · Parkinson EEG linear probe | PD AUC: 0.79 -> 0.87 | Improved |
| add fMRI pretraining (ablation) | UniEFS · Parkinson EEG linear probe | PD AUC: 0.70 -> 0.87 | Improved |
| use mask ratio 0.5 instead of 0.25 (ablation) | f-MSM pretraining · Eyes-closed rest EEG-fMRI | full-brain TCorr: 0.352 -> 0.367 | Improved |
| use mask ratio 0.5 instead of 0.75 (ablation) | f-MSM pretraining · Eyes-closed rest EEG-fMRI | full-brain TCorr: 0.357 -> 0.367 | Improved |
| use patch size 1 instead of 2 (ablation) | f-MSM · Eyes-closed rest EEG-fMRI | full-brain TCorr: 0.347 -> 0.367 | Improved |
| use patch size 1 instead of 8 (ablation) | f-MSM · Eyes-closed rest EEG-fMRI | full-brain TCorr: 0.362 -> 0.367 | Improved |
| lengthen EEG window from 4 s to 16 s (ablation) | UniEFS · Eyes-closed rest EEG-fMRI | full-brain TCorr: 0.157 -> 0.367 | Improved |
| lengthen EEG window from 12 s to 16 s (ablation) | UniEFS · Eyes-closed rest EEG-fMRI | Conn MSE: 0.187 -> 0.233 | Mixed |
| multi-region single UniEFS vs region-specific NeuroBOLT | UniEFS · Eyes-closed rest EEG-fMRI | full-brain TCorr: 0.331 -> 0.367 | Improved |
| UniEFS vs Li et al. on functional connectivity | UniEFS · Eyes-closed rest EEG-fMRI | Conn Corr: 0.535 -> 0.527 | Hurt |
| UniEFS vs CBraMod encoder on cerebellum | UniEFS · Eyes-closed rest EEG-fMRI | cerebellum TCorr: 0.257 -> 0.247 | Hurt |
| add vigilance and dataset prompts for leave-one-dataset-out (ablation) | UniEFS · Leave-one-dataset-out | TCorr | Improved |
Closest papers by meaning
- NeuroBOLT: Resting-state EEG-to-fMRI Synthesis with Multi-dimensional Feature Mapping (NeurIPS 2024 · similarity 0.91)
- Brain-OF: An Omnifunctional Foundation Model for fMRI, EEG and MEG (NeurIPS 2026 · similarity 0.91)
- MindEye2: Shared-Subject Models Enable fMRI-To-Image With 1 Hour of Data (ICML 2024 · similarity 0.89)
- BrainLM: A foundation model for brain activity recordings (ICLR 2024 · similarity 0.88)
- Brain decoding: toward real-time reconstruction of visual perception (ICLR 2024 · similarity 0.88)
- CSBrain: A Cross-scale Spatiotemporal Brain Foundation Model for EEG Decoding (NeurIPS 2025 · similarity 0.88)
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.
