A temporal attention-guided fusion module plus attention-modulated loss raises multimodal SNN accuracy to 77.55% on CREMA-D, 70.65% on AVE, and 97.5% on EAD.
mplug-owl2: Revolutionizing multi-modal large language model with modality collaboration
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
citation-role summary
background 1
citation-polarity summary
fields
cs.LG 1years
2025 1verdicts
CONDITIONAL 1roles
background 1polarities
support 1representative citing papers
citing papers explorer
-
Spiking Neural Networks with Temporal Attention-Guided Adaptive Fusion for imbalanced Multi-modal Learning
A temporal attention-guided fusion module plus attention-modulated loss raises multimodal SNN accuracy to 77.55% on CREMA-D, 70.65% on AVE, and 97.5% on EAD.