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

Exploiting label hierarchies has become a promising approach to tackling thezero-shot multi-label text classification (ZS-MTC) problem. Conventionalmethods aim to learn a matching model between text and labels, using a graphencoder to incorporate label hierarchies to obtain effective labelrepresentations \cite{rios2018few}. More recently, pretrained models like BERT\cite{devlin2018bert} have been used to convert classification tasks into atextual entailment task \cite{yin-etal-2019-benchmarking}. This approach isnaturally suitable for the ZS-MTC task. However, pretrained models areunderexplored in the existing work because they do not generate individualvector representations for text or labels, making it unintuitive to combinethem with conventional graph encoding methods. In this paper, we explore toimprove pretrained models with label hierarchies on the ZS-MTC task. We proposea Reinforced Label Hierarchy Reasoning (RLHR) approach to encourageinterdependence among labels in the hierarchies during training. Meanwhile, toovercome the weakness of flat predictions, we design a rollback algorithm thatcan remove logical errors from predictions during inference. Experimentalresults on three real-life datasets show that our approach achieves betterperformance and outperforms previous non-pretrained methods on the ZS-MTC task.

Improving Pretrained Models for Zero-shot Multi-label Text Classification through Reinforced Label Hierarchy Reasoning