Lightweight Semantic EEG Foundation Model for Frozen Cross-Disorder Transfer
Abstract
Large-scale EEG foundation models have demonstrated promising transferability across neurological disorders, but often require millions of parameters and substantial computational resources. In this paper, we present the Universal Semantic EEG Foundation Model (USE-FM), a lightweight EEG foundation model that learns transferable neural representations through self-supervised signal reconstruction on the Temple University Hospital EEG Corpus (TUEG). After pretraining, the encoder is frozen and evaluated on two clinically distinct downstream tasks, abnormal EEG detection (TUAB) and epileptic seizure recognition (TUEP), using a unified frozen-transfer protocol against recent EEG foundation models, including LUNA-Base and CBraMod. With only 1.46 million parameters, approximately one-fifth the size of existing models, USE-FM achieves competitive overall performance, including strong sensitivity and F1-score on TUEP (SEN 75.00 ± 14.14, F1 70.37 ± 4.01), while maintaining competitive performance on TUAB (AUC 85.24 ± 5.61). Beyond downstream classification, latent representation analysis using k-means clustering together with PCA and t-SNE demonstrates that USE-FM learns organized semantic EEG representations comparable to substantially larger foundation models. These results suggest that large-scale self-supervised pretraining enables lightweight architectures to learn transferable semantic EEG representations, providing a computationally efficient foundation for cross-disorder analysis and future clinical decision support in neurological disorders.
Related
- Agri-PIDNet-S: Lightweight semantic segmentation for field-structure and obstacle perception in seedling-stage cotton fields
- Do Vision CIL Methods Transfer to Protein? A Diagnostic Benchmark of Frozen Protein Encoders
- Probing EEG Foundation Model Attention Maps for Lightweight Interpretable Emotion Recognition
- Predictive Value of the Miscarriage Prediction Index for Early Miscarriage After Frozen Embryo Transfer
- Mangrove taxon mapping from Sentinel-1/2 using a compact one-dimensional network with probability calibration, predictive entropy, and frozen transfer evaluation
BibTeX
@misc{peng2026lightweight,
title = {{Lightweight Semantic EEG Foundation Model for Frozen Cross-Disorder Transfer}},
author = {Rita Huan-Ting Peng and Nhat Bui},
howpublished = {arXiv},
year = {2026},
month = oct,
doi = {10.48550/arxiv.2610.05503},
eprint = {2610.05503},
archivePrefix = {arXiv},
url = {https://arxiv.org/abs/2610.05503},
}
RIS
TY - UNPB
TI - Lightweight Semantic EEG Foundation Model for Frozen Cross-Disorder Transfer
AU - Rita Huan-Ting Peng
AU - Nhat Bui
PY - 2026
DA - 2026/10/04
PB - arXiv
DO - 10.48550/arxiv.2610.05503
UR - https://arxiv.org/abs/2610.05503
AB - Large-scale EEG foundation models have demonstrated promising transferability across neurological disorders, but often require millions of parameters and substantial computational resources. In this paper, we present the Universal Semantic EEG Foundation Model (USE-FM), a lightweight EEG foundation model that learns transferable neural representations through self-supervised signal reconstruction on the Temple University Hospital EEG Corpus (TUEG). After pretraining, the encoder is frozen and evaluated on two clinically distinct downstream tasks, abnormal EEG detection (TUAB) and epileptic seizure recognition (TUEP), using a unified frozen-transfer protocol against recent EEG foundation models, including LUNA-Base and CBraMod. With only 1.46 million parameters, approximately one-fifth the size of existing models, USE-FM achieves competitive overall performance, including strong sensitivity and F1-score on TUEP (SEN 75.00 ± 14.14, F1 70.37 ± 4.01), while maintaining competitive performance on TUAB (AUC 85.24 ± 5.61). Beyond downstream classification, latent representation analysis using k-means clustering together with PCA and t-SNE demonstrates that USE-FM learns organized semantic EEG representations comparable to substantially larger foundation models. These results suggest that large-scale self-supervised pretraining enables lightweight architectures to learn transferable semantic EEG representations, providing a computationally efficient foundation for cross-disorder analysis and future clinical decision support in neurological disorders.
ER -