Multimodal Sentiment Analysis On Mosi

Metrics

Accuracy

Results

Performance results of various models on this benchmark

		Paper Title
MMML	90.35	Multimodal Multi-loss Fusion Network for Sentiment Analysis
SPECTRA	87.50	Speech-Text Dialog Pre-training for Spoken Dialog Understanding with Explicit Cross-Modal Alignment
UniMSE	86.9	UniMSE: Towards Unified Multimodal Sentiment Analysis and Emotion Recognition
CM-BERT	84.5%	Cross-Modal BERT for Text-Audio Sentiment Analysis
Proposed: B2 + B4 w/ multimodal fusion	83.91%	Gated Mechanism for Attention Based Multimodal Sentiment Analysis
MulT	83	Multimodal Transformer for Unaligned Multimodal Language Sequences
MMMU-BA	82.31%	Contextual Inter-modal Attention for Multi-modal Sentiment Analysis
Multilogue-Net	81.19	Multilogue-Net: A Context-Aware RNN for Multi-modal Emotion Detection and Sentiment Analysis in Conversation
bc-LSTM	80.3%	Context-Dependent Sentiment Analysis in User-Generated Videos
MARN	77.1%	Multi-attention Recurrent Network for Human Communication Comprehension
CHFusion	76.5%	Multimodal Sentiment Analysis using Hierarchical Fusion with Context Modeling

0 of 11 row(s) selected.