SUP-MCRL: Subject-aware Unified Pseudo-feature Coded Multimodal Contrastive Representation Learning for EEG Visual Decoding
Two new research papers submitted to arXiv on June 15, 2026, explore advanced methods for decoding electroencephalography (EEG) signals. The first paper introduces subject-specific encoders to improve cross-subject EEG decoding by addressing distribution shifts, showing promise in improving accuracy for most subjects. The second paper, SUP-MCRL, presents a unified framework for EEG visual decoding that integrates semantic awareness, subject robustness, and representation consistency to overcome fidelity degradation in neural visual decoding. AI
IMPACT Advances in subject-aware EEG decoding could improve the accuracy and robustness of brain-computer interfaces for various applications.
- Hugging Face
- arXiv
- DagsHub
- alphaXiv
- CORE Recommender
- ScienceCast
- Gotit.pub
- THINGS-EEG
- CatalyzeX Code Finder for Papers
- Influence Flower
- SUP-MCRL
- AttentionBaseNet
- Euclidean Alignment (EA)
- Unified EEG Enhancer (UEE)
- Semantic-entity Aware Visual Encoder (SAVE)
- Prototype-based Progressive Augmenter (PPA)