/ EN
2026-09-01 00:00 China Papers Foundations & Methods Translated from EN

SEDAT: Hybrid Tokenizer Lifts EEG Foundation Model Accuracy by up to 15.3%

Summary Researchers at Northwestern Polytechnical University propose SEDAT, a hybrid tokenizer for large EEG foundation models that folds squeeze-and-excitation spatial aggregation, data-adaptive Gaussian average filtering, instantaneous-frequency-guided segmentation and Fourier-domain resampling into a single efficient pipeline. Evaluated on 10 heterogeneous EEG datasets with four foundation models, LaBraM, EEGFormer, EEGPT and NeuroGPT, SEDAT improved classification by up to 15.3% over fixed-length windowing and by 1.2-4.6% over the next-best tokenizer, the authors report. The study was published in the Journal of Neural Engineering on September 1, 2026.
Why it matters EEG foundation models are becoming the new substrate for non-invasive BCI decoding, and tokenization is the upstream step most papers ignore. A systematic comparison across 10 datasets and four models gives teams pretraining on EEG a reusable reference point rather than another single-benchmark claim.

BCIwiki (bciwiki.com) — Changing how raw brain signals are cut into tokens can lift the classification accuracy of large EEG foundation models by up to 15.3%, according to a study from Northwestern Polytechnical University published in the Journal of Neural Engineering on September 1, 2026.

How well a large EEG foundation model represents neural activity depends on how the raw signal is tokenized. Existing methods, the authors say, impose arbitrary temporal boundaries that are misaligned with neural state transitions, neglect spatial information across channels, and use fixed segmentation criteria that fail to generalize across heterogeneous EEG paradigms. Their SE-DAGAF Adaptive Tokenizer (SEDAT) integrates four components into a single computationally efficient pipeline: squeeze-and-excitation based spatial aggregation, signal decomposition by data-adaptive Gaussian average filtering, instantaneous-frequency-guided adaptive segmentation, and Fourier-domain resampling.

The evaluation spans 10 heterogeneous EEG datasets covering motor imagery, mental imagery, P300, slow cortical potentials, sleep staging and epilepsy, run through four foundation models, LaBraM, EEGFormer, EEGPT and NeuroGPT, and benchmarked against five baselines: fixed-length windowing, CTXSEG, LiPCoT, TFM-Tokenizer and SiS. SEDAT improved classification by up to 15.3% over fixed-length windowing and by 1.2-4.6% over the next-best method, with all comparisons reaching statistical significance after correction, the researchers report. Token quality analysis showed Silhouette scores of 0.81-0.85 versus 0.33-0.48 for rigid baselines. The authors describe SEDAT as a physiologically grounded and computationally practical tokenization solution for large-scale EEG foundation models.

Compiled by BCIwiki from public sources

Sources · 1

Northwestern Polytechnical University timeline

2026-08 Review Maps Deep Learning for Motor Imagery BCIs 2026-06 Northwestern Polytechnical University Registers BCI Stroke Rehabilitation Trial With 4 Arms All entries →
Read original ↗
Suggest a correction Revisions · none / EN
© 2026 BCIwiki.com Digest Topics Tips Subscribe Revisions About