Research map / Architectures and efficiency / Attention and transformers
MAtt: A Manifold Attention Network for EEG Decoding
Yue-Ting Pan, Jing-Lun Chou, Chun-Shu Wei
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| use MAtt instead of SCCNet | EEG decoding · BCIC-IV-2a | accuracy: 71.95±5.05 -> 74.71±5.01 | Improved |
| use MAtt instead of SCCNet | EEG decoding · MAMEM-SSVEP-II | accuracy: 62.11±7.70 -> 65.50±8.20 | Improved |
| use MAtt instead of EEG-TCNet | EEG decoding · BCI-ERN | AUC: 77.05±2.46 -> 76.01±2.28 | Hurt |
| add MA to FE (ablation) | MAtt · BCIC-IV-2a | accuracy: 26.08±0.70 -> 74.71±5.01 | Improved |
| add MA to FE (ablation) | MAtt · MAMEM-SSVEP-II | accuracy: 20.18±1.11 -> 65.50±8.20 | Improved |
| add MA to FE (ablation) | MAtt · BCI-ERN | AUC: 73.40±2.27 -> 76.01±2.28 | Improved |
| add FE to MA (ablation) | MAtt · BCIC-IV-2a | accuracy: 60.73±5.80 -> 74.71±5.01 | Improved |
| add FE to MA (ablation) | MAtt · MAMEM-SSVEP-II | accuracy: 30.51±2.57 -> 65.50±8.20 | Improved |
| add FE to MA (ablation) | MAtt · BCI-ERN | AUC: 59.47±3.56 -> 76.01±2.28 | Improved |
| use manifold attention instead of self-attention (ablation) | FE+MA · BCIC-IV-2a | accuracy: 49.19±2.72 -> 74.71±5.01 | Improved |
| use manifold attention instead of self-attention (ablation) | FE+MA · MAMEM-SSVEP-II | accuracy: 22.91±2.00 -> 65.50±8.20 | Improved |
| use manifold attention instead of self-attention (ablation) | FE+MA · BCI-ERN | AUC: 63.77±1.71 -> 76.01±2.28 | Improved |
Closest papers by meaning
- CSBrain: A Cross-scale Spatiotemporal Brain Foundation Model for EEG Decoding (NeurIPS 2025 · similarity 0.86)
- Decoding Natural Images from EEG for Object Recognition (ICLR 2024 · similarity 0.86)
- SPDIM: Source-Free Unsupervised Conditional and Label Shift Adaptation in EEG (ICLR 2025 · similarity 0.85)
- Neural decoding from stereotactic EEG: accounting for electrode variability across subjects (NeurIPS 2024 · similarity 0.85)
- MindAlign: Bridging EEG, Vision, and Language for Zero-Shot Visual Decoding (NeurIPS 2026 · similarity 0.85)
- LAtte: Hyperbolic Lorentz Attention for Joint-Subject EEG Classification (NeurIPS 2026 · similarity 0.84)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.
