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 Tools annual

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 Tools monthly

Global: Zendy Plus monthly

Presents the access of premium content as premium feature

Premium Content

Global: Zendy Plus annual

Global: Zendy Free Plan

Due to the broad range of social media platforms and their user groups, therequirements of abusive language detection systems are varied andever-changing. Already a large set of annotated corpora with differentproperties and label sets were created, such as hate or misogyny detection, butthe form and targets of abusive speech are constantly changing. Since, theannotation of new corpora is expensive, in this work we leverage datasets wealready have, covering a wide range of tasks related to abusive languagedetection, in order to build models cheaply for a new target label set and/orlanguage, using only a few training examples of the target domain. We propose atwo-step approach: first we train our model in a multitask fashion. We thencarry out few-shot adaptation to the target requirements. Our experiments showthat by leveraging already existing datasets and only a few-shots of the targettask the performance of models can be improved not only monolingually butacross languages as well. Our analysis also shows that our models acquire ageneral understanding of abusive language, since they improve the prediction oflabels which are present only in the target dataset. We also analyze thetrade-off between specializing the already existing datasets to a given targetsetup for best performance and its negative effects on model adaptability.

How to Solve Few-Shot Abusive Content Detection Using the Data We Actually Have