Metraipolis › archive › dc_2cd01defb85d

Beyond Flattened Tokens: Structure-Preserving EEG Decoding with Reusable TriDim Blocks — arXiv:2609.19842

Source: arXiv (en)
Original: https://arxiv.org/abs/2609.19842
License (as archived): CC0-1.0
Archived: 2026-10-10T04:08:45.606Z
SHA-256 of the archived text: 9d479ba0516bd0c41ba0e67a898d0548524f4c12c44fd719619811898d784aec
Stable URL of this page: https://metraipolis.com/documento/dc_2cd01defb85d

How to cite

Beyond Flattened Tokens: Structure-Preserving EEG Decoding with Reusable TriDim Blocks — arXiv:2609.19842 — arXiv (en), https://arxiv.org/abs/2609.19842 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_2cd01defb85d (Metraipolis archive dc_2cd01defb85d, sha256 9d479ba0516bd0c41ba0e67a898d0548524f4c12c44fd719619811898d784aec)

BibTeX

@misc{dc2cd01defb85d,
  title = {Beyond Flattened Tokens: Structure-Preserving EEG Decoding with Reusable TriDim Blocks — arXiv:2609.19842},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2609.19842},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_2cd01defb85d; sha256 9d479ba0516bd0c41ba0e67a898d0548524f4c12c44fd719619811898d784aec},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.19842. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

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 this https URL .

Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-09-18 in cs.LG).
Categorie arXiv: Machine Learning (cs.LG).
Pagina del lavoro: https://arxiv.org/abs/2609.19842 — autori e testo completo sono lì (qui omessi i nomi: nessun dato personale nel deposito).
Licenza: titolo e abstract sono metadati arXiv distribuiti in CC0 1.0; il testo completo ha la licenza scelta dagli autori, indicata sulla pagina.
Letto il 2026-10-10T03:48:18.291Z dalla pagina https://arxiv.org/catchup/cs.LG/2026-09-18?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).