ECHO让脑电模型通过序列建模实现动态适应，无需重训练即可跨任务泛化。
ECHO: Toward Contextual Seq2Seq Paradigms in Large EEG Models
- 采用解码器中心架构，将脑电建模转为序列到序列学习
- 在多任务场景下优于现有单任务大模型，性能提升显著
- 支持上下文学习，可零参数适配新任务，适合跨领域应用
脑电图（EEG）应用广泛，亟需能在多种任务和数据集间有效泛化的模型。大型脑电模型（LEMs）通过在大规模无标签数据上预训练编码器架构来提取通用表征，但其解码器能力不足，限制了所学特征的充分利用。为此，我们提出ECHO，一种新型解码器中心的LEMs范式，将脑电建模重新定义为序列到序列学习。ECHO在序列空间中捕捉信号、标签与任务间的层次关系，并引入离散支持样本构建上下文线索，使模型具备上下文学习能力，可在不更新参数的情况下动态适应异构任务。在多个数据集上的大量实验表明，即使使用基础模型组件，ECHO在多任务设置中始终优于最先进的单任务LEMs，展现出更强的泛化性和适应性。
原文摘要 · Abstract (English)
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.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。