ZUNA: Flexible EEG Superresolution with Position-Aware Diffusion Autoencoders
Fuente:
arXiv
Salvato in:
| Autori principali: | , , , , |
|---|---|
| Natura: | Preprint |
| Pubblicazione: |
2026
|
| Soggetti: | |
| Accesso online: | |
| Tags: |
Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
|
| _version_ | 1866918349495074816 |
|---|---|
| author | Warner, Christopher Mago, Jonas Huml, JR Osman, Mohamed Millidge, Beren |
| author_facet | Warner, Christopher Mago, Jonas Huml, JR Osman, Mohamed Millidge, Beren |
| contents | We present \texttt{ZUNA}, a 380M-parameter masked diffusion autoencoder trained to perform masked channel infilling and superresolution for arbitrary electrode numbers and positions in EEG signals. The \texttt{ZUNA} architecture tokenizes multichannel EEG into short temporal windows and injects spatiotemporal structure via a 4D rotary positional encoding over (x,y,z,t), enabling inference on arbitrary channel subsets and positions. We train ZUNA on an aggregated and harmonized corpus spanning 208 public datasets containing approximately 2 million channel-hours using a combined reconstruction and heavy channel-dropout objective. We show that \texttt{ZUNA} substantially improves over ubiquitous spherical-spline interpolation methods, with the gap widening at higher dropout rates. Crucially, compared to other deep learning methods in this space, \texttt{ZUNA}'s performance \emph{generalizes} across datasets and channel positions allowing it to be applied directly to novel datasets and problems. Despite its generative capabilities, \texttt{ZUNA} remains computationally practical for deployment. We release Apache-2.0 weights and an MNE-compatible preprocessing/inference stack to encourage reproducible comparisons and downstream use in EEG analysis pipelines. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2602_18478 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | ZUNA: Flexible EEG Superresolution with Position-Aware Diffusion Autoencoders Warner, Christopher Mago, Jonas Huml, JR Osman, Mohamed Millidge, Beren Signal Processing Artificial Intelligence Machine Learning We present \texttt{ZUNA}, a 380M-parameter masked diffusion autoencoder trained to perform masked channel infilling and superresolution for arbitrary electrode numbers and positions in EEG signals. The \texttt{ZUNA} architecture tokenizes multichannel EEG into short temporal windows and injects spatiotemporal structure via a 4D rotary positional encoding over (x,y,z,t), enabling inference on arbitrary channel subsets and positions. We train ZUNA on an aggregated and harmonized corpus spanning 208 public datasets containing approximately 2 million channel-hours using a combined reconstruction and heavy channel-dropout objective. We show that \texttt{ZUNA} substantially improves over ubiquitous spherical-spline interpolation methods, with the gap widening at higher dropout rates. Crucially, compared to other deep learning methods in this space, \texttt{ZUNA}'s performance \emph{generalizes} across datasets and channel positions allowing it to be applied directly to novel datasets and problems. Despite its generative capabilities, \texttt{ZUNA} remains computationally practical for deployment. We release Apache-2.0 weights and an MNE-compatible preprocessing/inference stack to encourage reproducible comparisons and downstream use in EEG analysis pipelines. |
| title | ZUNA: Flexible EEG Superresolution with Position-Aware Diffusion Autoencoders |
| topic | Signal Processing Artificial Intelligence Machine Learning |
| url | https://arxiv.org/abs/2602.18478 |