Reproduce “Beyond Flattened Tokens: Structure-Preserving EEG Decoding with Reusable TriDim Blocks” on OpenResearch
Turn your favorite coding agents into research agents that review literature, develop hypotheses, and run experiments.
Universal app for Apple silicon and Intel. Requires macOS 11 or later.
About this paper
Shiyue Su, Song Wang, Zekai Zhan, Junjie Zeng, Ziling Lu, Zongsheng Li, Xinyuan Ye, Zhiyuan Ma and 2 more
TriDimEEG decodes EEG while keeping channel, within-patch time, and across-patch time as distinct axes. Its blocks apply attention and axis-specific refinement across these dimensions, then combine them for classification. Across eight datasets, it has the highest aggregate accuracy (57.70%) and average rank (2.75) among 15 models, but does not lead every dataset; its paired comparison with EEGDeformer is not significant.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.