Related Experiment Videos
SEDAT: A hybrid tokenizer for large EEG models
Muhammad Zulkifal Aziz1, Yue Zhuo2, Binwen Huang3
1School of Automation, Northwestern Polytechnical University, 1 Dongxiang Road, Chang'an District,, China, Xi'an, 710072, China.
Objective:
The fidelity of neural representations learned by large EEG foundation models depends on how raw brain signals are tokenized. Existing methods suffer from arbitrary temporal boundaries misaligned with neural state transitions, neglecting inter-channel spatial information, and fixed segmentation criteria that fail to generalize across heterogeneous EEG paradigms.
Approach:
This study proposes the SE-DAGAF Adaptive Tokenizer (SEDAT), a hybrid framework integrating squeeze and- excitation (SE)-based spatial aggregation, data-adaptive Gaussian average filtering (DAGAF)- based signal decomposition, instantaneous-frequency-guided adaptive segmentation, and Fourier domain resampling into a single computationally efficient pipeline. SEDAT is evaluated across 10 heterogeneous EEG datasets spanning motor imagery, mental imagery, P300, slow cortical potentials, sleep staging, and epilepsy paradigms, using four large foundation models: LaBraM, EEGFormer, EEGPT, and NeuroGPT. It is benchmarked against five competitive baselines: fixed length windowing, CTXSEG, LiPCoT, TFM-Tokenizer, and SiS.
Main Results:
SEDAT achieves classification improvements of up to 15.3% over fixed-length windowing and 1.2-4.6% over the second-best method, with all comparisons reaching p < 0.001 after Benjamini-Hochberg correction. Token quality analysis confirms substantially improved feature separability, with Silhouette scores of 0.81-0.85 versus 0.33-0.48 for rigid baselines.
Significance:
With O(CN + KN logN) complexity, SEDAT explores new applications for SE and DAGAF as tokenizers and provides a physiologically grounded and computationally practical tokenization solution for large-scale EEG foundation models.