Pretraining a region-aware spatiotemporal EEG encoder with JEPA generative and masked dynamic contrastive losses on FACED yields higher cross-subject accuracy than SSL baselines when fine-tuned on SEED-IV/V/VII.
GNN-based multi-source domain prototype representation for cross-subject EEG emotion recognition,
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.LG 1years
2026 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
Masked Generative-Contrastive Representation Learning for Cross-Dataset EEG-Based Emotion Recognition
Pretraining a region-aware spatiotemporal EEG encoder with JEPA generative and masked dynamic contrastive losses on FACED yields higher cross-subject accuracy than SSL baselines when fine-tuned on SEED-IV/V/VII.