MPNet aggregates multi-view Riemannian nodes from EEG rhythms via a new manifold node pooling layer to achieve state-of-the-art accuracy with up to 10x faster runtime than comparable models on public datasets.
Deep learning with convolutional neural networks for eeg decoding and visualization,
5 Pith papers cite this work. Polarity classification is still indexing.
5
Pith papers citing it
years
2026 5verdicts
UNVERDICTED 5representative citing papers
StaFlowNet improves MI-EEG decoding by separating and coordinating global state vectors with temporal flow features via a dual-branch design and state-modulated flow module, outperforming prior methods on three public datasets.
MB2L achieves 80.5% top-1 and 97.6% top-5 accuracy on zero-shot EEG-to-image retrieval by using biomimetic modules and bidirectional contrastive learning to align neural and visual features.
CNN-attention model decodes EEG to hand kinematics with within-subject PCCs above 0.98 on two axes, improved to 0.93 overall by a motion-state FSM copilot that drops under 20% of points, enabling simulated Franka Panda control.
EEG-MFTNet enhances EEGNet with multi-scale temporal convolutions and transformer fusion to achieve 58.9% average accuracy in subject-dependent cross-session motor imagery classification on the SHU dataset while keeping low computational cost.
citing papers explorer
-
MPNet: A Robust and Efficient Manifold Pooling Network for Multi-Rhythm EEG Signal Decoding
MPNet aggregates multi-view Riemannian nodes from EEG rhythms via a new manifold node pooling layer to achieve state-of-the-art accuracy with up to 10x faster runtime than comparable models on public datasets.
-
State-Flow Coordinated Representation for MI-EEG Decoding
StaFlowNet improves MI-EEG decoding by separating and coordinating global state vectors with temporal flow features via a dual-branch design and state-modulated flow module, outperforming prior methods on three public datasets.
-
Multi-Level Bidirectional Biomimetic Learning for EEG-Based Visual Decoding
MB2L achieves 80.5% top-1 and 97.6% top-5 accuracy on zero-shot EEG-to-image retrieval by using biomimetic modules and bidirectional contrastive learning to align neural and visual features.
-
Copilot-Assisted Second-Thought Framework for Brain-to-Robot Hand Motion Decoding
CNN-attention model decodes EEG to hand kinematics with within-subject PCCs above 0.98 on two axes, improved to 0.93 overall by a motion-state FSM copilot that drops under 20% of points, enabling simulated Franka Panda control.
-
EEG-MFTNet: An Enhanced EEGNet Architecture with Multi-Scale Temporal Convolutions and Transformer Fusion for Cross-Session Motor Imagery Decoding
EEG-MFTNet enhances EEGNet with multi-scale temporal convolutions and transformer fusion to achieve 58.9% average accuracy in subject-dependent cross-session motor imagery classification on the SHU dataset while keeping low computational cost.