English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

In multimodal sentiment analysis (MSA), the performance of a model highlydepends on the quality of synthesized embeddings. These embeddings aregenerated from the upstream process called multimodal fusion, which aims toextract and combine the input unimodal raw data to produce a richer multimodalrepresentation. Previous work either back-propagates the task loss ormanipulates the geometric property of feature spaces to produce favorablefusion results, which neglects the preservation of critical task-relatedinformation that flows from input to the fusion results. In this work, wepropose a framework named MultiModal InfoMax (MMIM), which hierarchicallymaximizes the Mutual Information (MI) in unimodal input pairs (inter-modality)and between multimodal fusion result and unimodal input in order to maintaintask-related information through multimodal fusion. The framework is jointlytrained with the main task (MSA) to improve the performance of the downstreamMSA task. To address the intractable issue of MI bounds, we further formulate aset of computationally simple parametric and non-parametric methods toapproximate their truth value. Experimental results on the two widely useddatasets demonstrate the efficacy of our approach. The implementation of thiswork is publicly available athttps://github.com/declare-lab/Multimodal-Infomax.

Improving Multimodal Fusion with Hierarchical Mutual Information Maximization for Multimodal Sentiment Analysis