超越扁平化标记：使用可复用TriDim块进行结构保持的脑电解码
Beyond Flattened Tokens: Structure-Preserving EEG Decoding with Reusable TriDim Blocks
- Southern University of Science and Technology（南方科技大学）
- Omni-Intelligence
- Shenzhen Loop Area Institute（深圳河套学院）
- The Chinese University of Hong Kong, Shenzhen（香港中文大学（深圳））
- Tsinghua University（清华大学）
机构由 AI 辅助整理，请以论文原文为准。
AI总结：
针对EEG解码中结构信息丢失问题，提出可复用TriDim块保持通道、短期和长期时间轴，构建TriDimEEG解码器，在八数据集上取得最优性能，并提升基础模型下游准确率。
AI中文摘要：
有效的脑电（EEG）解码需要能够保持通道间组织、局部波形动态和长程时间上下文的表示。现有的EEG架构通常使用独立的专门模块来捕获这些结构，或将它们压缩成单一的标记序列，这使得在整个骨干网络中维持它们的不同角色并协调它们的交互变得困难。我们提出TriDim，一个可复用的块，它保持表示形状并明确三个EEG轴：通道、每个补丁内的样本位置以及整个记录中的补丁位置。这些轴分别对应于空间、短期时间和长期时间信息。每个TriDim块沿各个轴应用前馈变换，并通过跨轴注意力协调它们之间的信息交换。通过堆叠TriDim块和多级三轴读出，我们构建了TriDimEEG，一个独立的EEG解码器。在涵盖临床诊断、睡眠分期、运动想象和情绪识别的八个数据集上进行的严格跨受试者评估中，TriDimEEG在十五个评估模型中实现了最佳整体性能，平均准确率比第二好的模型相对提高4.3%。将三个EEG基础模型中的Transformer块替换为TriDim块，在下游准确率上平均相对提高7.4%，同时参数数量减少17.0%至47.3%。这些结果确立了TriDim作为有效且可复用的构建块，以及TriDimEEG作为强大的独立EEG解码器。TriDimEEG的代码和参数可在该https URL获取。
英文摘要：
Effective EEG decoding requires representations that preserve organization among channels, local waveform dynamics, and long-range temporal context. Existing EEG architectures often capture these structures using separate specialized modules or collapse them into a single token sequence, making it difficult to maintain their distinct roles and coordinate their interactions throughout the backbone. We propose TriDim, a reusable block that preserves the representation shape and keeps three EEG axes explicit: channel, sample position within each patch, and patch position across the recording. These axes correspond to spatial, short-term temporal, and long-term temporal information, respectively. Each TriDim block applies feed-forward transformations along individual axes and cross-axis attention to coordinate information exchange among them. By stacking TriDim blocks with a multi-level tri-axis readout, we construct TriDimEEG, a standalone EEG decoder. Under strict cross-subject evaluation on eight datasets spanning clinical diagnosis, sleep staging, motor imagery, and emotion recognition, TriDimEEG achieves the best overall performance among fifteen evaluated models, with a 4.3% relative improvement in average accuracy over the second-best model. Replacing Transformer blocks in three EEG foundation models with TriDim blocks yields an average relative improvement of 7.4% in downstream accuracy while reducing parameter counts by 17.0% to 47.3%. These results establish TriDim as an effective and reusable building block and TriDimEEG as a strong standalone EEG decoder. Code and parameters of TriDimEEG are available at https://github.com/ncclab-sustech/TriDim_model.
