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

Most evaluations of attribution methods focus on the English language. Inthis work, we present a multilingual approach for evaluating attributionmethods for the Natural Language Inference (NLI) task in terms of faithfulnessand plausibility. First, we introduce a novel cross-lingual strategy to measurefaithfulness based on word alignments, which eliminates the drawbacks oferasure-based evaluations.We then perform a comprehensive evaluation ofattribution methods, considering different output mechanisms and aggregationmethods. Finally, we augment the XNLI dataset with highlight-basedexplanations, providing a multilingual NLI dataset with highlights, to supportfuture exNLP studies. Our results show that attribution methods performing bestfor plausibility and faithfulness are different.

A Multilingual Perspective Towards the Evaluation of Attribution Methods in Natural Language Inference