ECHO: Toward Contextual Seq2Seq Paradigms in Large EEG Models
Fuente:
arXiv
Saved in:
| Main Authors: | , , , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2025
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866909809611112448 |
|---|---|
| author | Liu, Chenyu Deng, Yuqiu Liu, Tianyu Zhou, Jinan Zhou, Xinliang Jia, Ziyu Ding, Yi |
| author_facet | Liu, Chenyu Deng, Yuqiu Liu, Tianyu Zhou, Jinan Zhou, Xinliang Jia, Ziyu Ding, Yi |
| contents | Electroencephalography (EEG), with its broad range of applications, necessitates models that can generalize effectively across various tasks and datasets. Large EEG Models (LEMs) address this by pretraining encoder-centric architectures on large-scale unlabeled data to extract universal representations. While effective, these models lack decoders of comparable capacity, limiting the full utilization of the learned features. To address this issue, we introduce ECHO, a novel decoder-centric LEM paradigm that reformulates EEG modeling as sequence-to-sequence learning. ECHO captures layered relationships among signals, labels, and tasks within sequence space, while incorporating discrete support samples to construct contextual cues. This design equips ECHO with in-context learning, enabling dynamic adaptation to heterogeneous tasks without parameter updates. Extensive experiments across multiple datasets demonstrate that, even with basic model components, ECHO consistently outperforms state-of-the-art single-task LEMs in multi-task settings, showing superior generalization and adaptability. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2509_22556 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | ECHO: Toward Contextual Seq2Seq Paradigms in Large EEG Models Liu, Chenyu Deng, Yuqiu Liu, Tianyu Zhou, Jinan Zhou, Xinliang Jia, Ziyu Ding, Yi Machine Learning Signal Processing Electroencephalography (EEG), with its broad range of applications, necessitates models that can generalize effectively across various tasks and datasets. Large EEG Models (LEMs) address this by pretraining encoder-centric architectures on large-scale unlabeled data to extract universal representations. While effective, these models lack decoders of comparable capacity, limiting the full utilization of the learned features. To address this issue, we introduce ECHO, a novel decoder-centric LEM paradigm that reformulates EEG modeling as sequence-to-sequence learning. ECHO captures layered relationships among signals, labels, and tasks within sequence space, while incorporating discrete support samples to construct contextual cues. This design equips ECHO with in-context learning, enabling dynamic adaptation to heterogeneous tasks without parameter updates. Extensive experiments across multiple datasets demonstrate that, even with basic model components, ECHO consistently outperforms state-of-the-art single-task LEMs in multi-task settings, showing superior generalization and adaptability. |
| title | ECHO: Toward Contextual Seq2Seq Paradigms in Large EEG Models |
| topic | Machine Learning Signal Processing |
| url | https://arxiv.org/abs/2509.22556 |