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

Despite significant advancements in multi-label text classification, theability of existing models to generalize to novel and seldom-encounteredcomplex concepts, which are compositions of elementary ones, remainsunderexplored. This research addresses this gap. By creating unique data splitsacross three benchmarks, we assess the compositional generalization ability ofexisting multi-label text classification models. Our results show that thesemodels often fail to generalize to compositional concepts encounteredinfrequently during training, leading to inferior performance on tests withthese new combinations. To address this, we introduce a data augmentationmethod that leverages two innovative text generation models designed to enhancethe classification models' capacity for compositional generalization. Ourexperiments show that this data augmentation approach significantly improvesthe compositional generalization capabilities of classification models on ourbenchmarks, with both generation models surpassing other text generationbaselines.

Compositional Generalization for Multi-label Text Classification: A Data-Augmentation Approach