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

Existing multimodal tasks mostly target at the complete input modalitysetting, i.e., each modality is either complete or completely missing in bothtraining and test sets. However, the randomly missing situations have stillbeen underexplored. In this paper, we present a novel approach named MM-Alignto address the missing-modality inference problem. Concretely, we propose 1) analignment dynamics learning module based on the theory of optimal transport(OT) for indirect missing data imputation; 2) a denoising training algorithm tosimultaneously enhance the imputation results and backbone network performance.Compared with previous methods which devote to reconstructing the missinginputs, MM-Align learns to capture and imitate the alignment dynamics betweenmodality sequences. Results of comprehensive experiments on three datasetscovering two multimodal tasks empirically demonstrate that our method canperform more accurate and faster inference and relieve overfitting undervarious missing conditions.

MM-Align: Learning Optimal Transport-based Alignment Dynamics for Fast and Accurate Inference on Missing Modality Sequences