TL;DR
CHSIM, a unified MSA framework for robust text–image sentiment modeling that combines three complementary mechanisms: prediction-consistency regularization under stochastic latent masking, hierarchical cross-modal alignment, and global bidirectional sequence modeling, is proposed.
A Modality Dropout strategy is first introduced at the input stage to alleviate over-reliance on a sin-gle modality and improve robustness and the proposed Hierarchical Global-Local Interaction and Refinement framework for Multimodal Sentiment Analysis (HGLIR) is proposed.
An Adaptive Multimodal Fusion with Hierarchical Semantic Disentanglement with Hierarchical Semantic Disentanglement is proposed, which achieves robust multimodal representation learning through uncertainty-aware dynamic calibration, alongside semantic disentanglement and refinement.
The results show that ADRTW remains competitive with static averaging in controlled settings while preserving context-sensitive local variation in large-scale discourse analysis, and the proposed framework provides a transparent and reusable basis for examining emotional dynamics in social media discourse.
FUSE-Net is proposed, a Factorized and Unified Semantic Enhancement framework that decomposes each modality into shared, specific, and noise subspaces and applies contrastive learning, an information-gain constraint, and duality constraints for structured regularization to preserve task-relevant semantics during factor...
DualScope is proposed, a novel model that combines a global-local fusion strategy with bidirectional image-text generation for semantically consistent data augmentation and introduces both label contrastive learning and data contrastive learning to align heterogeneous modalities and enhance model robustness.
MGSI first encodes audio and visual streams at short-, medium-, and long-range temporal scales, preserving both local variations and global affective trends, and applies polarity- and intensity-aware enhancement to better handle ambiguous and near-neutral samples.
We use cookies to run the site and, with your consent, for analytics and to show ads. See our Cookie Policy.
