Low Rank Fusion based Transformers for Multimodal Sequences
This work addresses multimodal emotion recognition for applications like human-computer interaction, but it is incremental, building on existing fusion and transformer methods.
The paper tackles multimodal sentiment and emotion recognition by proposing a low-rank fusion-based transformer architecture that reduces model parameters and training time while achieving comparable performance to larger fusion-based models on CMU-MOSEI, CMU-MOSI, and IEMOCAP datasets.
Our senses individually work in a coordinated fashion to express our emotional intentions. In this work, we experiment with modeling modality-specific sensory signals to attend to our latent multimodal emotional intentions and vice versa expressed via low-rank multimodal fusion and multimodal transformers. The low-rank factorization of multimodal fusion amongst the modalities helps represent approximate multiplicative latent signal interactions. Motivated by the work of~\cite{tsai2019MULT} and~\cite{Liu_2018}, we present our transformer-based cross-fusion architecture without any over-parameterization of the model. The low-rank fusion helps represent the latent signal interactions while the modality-specific attention helps focus on relevant parts of the signal. We present two methods for the Multimodal Sentiment and Emotion Recognition results on CMU-MOSEI, CMU-MOSI, and IEMOCAP datasets and show that our models have lesser parameters, train faster and perform comparably to many larger fusion-based architectures.