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

Existing metrics for evaluating the quality of automatically generatedquestions such as BLEU, ROUGE, BERTScore, and BLEURT compare the reference andpredicted questions, providing a high score when there is a considerablelexical overlap or semantic similarity between the candidate and the referencequestions. This approach has two major shortcomings. First, we need expensivehuman-provided reference questions. Second, it penalises valid questions thatmay not have high lexical or semantic similarity to the reference questions. Inthis paper, we propose a new metric, RQUGE, based on the answerability of thecandidate question given the context. The metric consists of aquestion-answering and a span scorer modules, using pre-trained models fromexisting literature, thus it can be used without any further training. Wedemonstrate that RQUGE has a higher correlation with human judgment withoutrelying on the reference question. Additionally, RQUGE is shown to be morerobust to several adversarial corruptions. Furthermore, we illustrate that wecan significantly improve the performance of QA models on out-of-domaindatasets by fine-tuning on synthetic data generated by a question generationmodel and re-ranked by RQUGE.

RQUGE: Reference-Free Metric for Evaluating Question Generation by Answering the Question