GSIFN: A Graph-Structured and Interlaced-Masked Multimodal Transformer-based Fusion Network for Multimodal Sentiment Analysis

Fuente: arXiv
Saved in:
Bibliographic Details
Main Author: Jin, Yijie
Format: Preprint
Published: 2024
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866917854051303424
author Jin, Yijie
author_facet Jin, Yijie
contents Multimodal Sentiment Analysis (MSA) leverages multiple data modals to analyze human sentiment. Existing MSA models generally employ cutting-edge multimodal fusion and representation learning-based methods to promote MSA capability. However, there are two key challenges: (i) in existing multimodal fusion methods, the decoupling of modal combinations and tremendous parameter redundancy, lead to insufficient fusion performance and efficiency; (ii) a challenging trade-off exists between representation capability and computational overhead in unimodal feature extractors and encoders. Our proposed GSIFN incorporates two main components to solve these problems: (i) a graph-structured and interlaced-masked multimodal Transformer. It adopts the Interlaced Mask mechanism to construct robust multimodal graph embedding, achieve all-modal-in-one Transformer-based fusion, and greatly reduce the computational overhead; (ii) a self-supervised learning framework with low computational overhead and high performance, which utilizes a parallelized LSTM with matrix memory to enhance non-verbal modal features for unimodal label generation. Evaluated on the MSA datasets CMU-MOSI, CMU-MOSEI, and CH-SIMS, GSIFN demonstrates superior performance with significantly lower computational overhead compared with previous state-of-the-art models.
format Preprint
id arxiv_https___arxiv_org_abs_2408_14809
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle GSIFN: A Graph-Structured and Interlaced-Masked Multimodal Transformer-based Fusion Network for Multimodal Sentiment Analysis
Jin, Yijie
Computation and Language
Multimodal Sentiment Analysis (MSA) leverages multiple data modals to analyze human sentiment. Existing MSA models generally employ cutting-edge multimodal fusion and representation learning-based methods to promote MSA capability. However, there are two key challenges: (i) in existing multimodal fusion methods, the decoupling of modal combinations and tremendous parameter redundancy, lead to insufficient fusion performance and efficiency; (ii) a challenging trade-off exists between representation capability and computational overhead in unimodal feature extractors and encoders. Our proposed GSIFN incorporates two main components to solve these problems: (i) a graph-structured and interlaced-masked multimodal Transformer. It adopts the Interlaced Mask mechanism to construct robust multimodal graph embedding, achieve all-modal-in-one Transformer-based fusion, and greatly reduce the computational overhead; (ii) a self-supervised learning framework with low computational overhead and high performance, which utilizes a parallelized LSTM with matrix memory to enhance non-verbal modal features for unimodal label generation. Evaluated on the MSA datasets CMU-MOSI, CMU-MOSEI, and CH-SIMS, GSIFN demonstrates superior performance with significantly lower computational overhead compared with previous state-of-the-art models.
title GSIFN: A Graph-Structured and Interlaced-Masked Multimodal Transformer-based Fusion Network for Multimodal Sentiment Analysis
topic Computation and Language
url https://arxiv.org/abs/2408.14809