发表机构
School of Information and Control Engineering, China University of Mining and Technology; School of Automation and Intelligent Sensing, Shanghai Jiao Tong University; Shanghai Innovation Institute（中国矿业大学信息与控制工程学院; 上海交通大学自动化与智能感知学院; 上海创新研究院）
机构由 AI 辅助整理，请以论文原文为准。AI 中文总结
本研究提出EEG-VID预训练框架，通过指数移动平均目标编码器与弱任务引导实现EEG解码，在多数据集及场景任务中提升准确率，为EEG解码与辅助目标选择提供可迁移预训练策略。
AI 中文摘要
我们提出了EEG-VID，这是一种用于处理会话与受试者偏移的EEG解码任务的任务引导型潜在预测预训练框架。EEG-VID通过指数移动平均目标编码器与弱任务引导，从近期历史数据预测未来潜在EEG状态，随后进行监督微调。在VIG-48与BCI Competition IV-2a/IV-2b数据集上，第1阶段在42组匹配的骨干网络-数据集-协议对比中，有41组提升了平均准确率，其中包括全部12种留一受试者设置，最大提升幅度达16.22个百分点。在48区域跨日VIG-48任务中，EEG-VID达到6.52%的Top-1准确率与30.50%的Top-5准确率。在一项包含6名受试者的离线机器人场景研究中，经受试者特定校准后，候选约束型目标选择的准确率达40.24%，而随机概率水平为25%。这些结果表明，任务引导型潜在预测是一种可迁移的预训练策略，适用于EEG解码与场景约束型辅助目标选择。
英文摘要
We propose EEG-VID, a task-guided latent predictive pretraining framework for EEG decoding under session and subject shifts. EEG-VID predicts future latent EEG states from recent history using an exponential-moving-average target encoder and weak task guidance, followed by supervised fine-tuning. Across VIG-48 and BCI Competition IV-2a/IV-2b, Stage 1 improves mean accuracy in 41 of 42 matched backbone-dataset-protocol comparisons, including all 12 leave-one-subject-out settings, with a maximum gain of 16.22 percentage points. On the 48-region cross-day VIG-48 task, EEG-VID achieves 6.52% Top-1 and 30.50% Top-5 accuracy. In a separate six-participant offline robot-scene study, candidate-constrained target selection reaches 40.24% versus a 25% chance level after subject-specific calibration. These results support task-guided latent prediction as a transferable pretraining strategy for EEG decoding and scene-constrained assistive target selection.
