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

Natural language inference (NLI) requires models to learn and applycommonsense knowledge. These reasoning abilities are particularly important forexplainable NLI systems that generate a natural language explanation inaddition to their label prediction. The integration of external knowledge hasbeen shown to improve NLI systems, here we investigate whether it can alsoimprove their explanation capabilities. For this, we investigate differentsources of external knowledge and evaluate the performance of our models onin-domain data as well as on special transfer datasets that are designed toassess fine-grained reasoning capabilities. We find that different sources ofknowledge have a different effect on reasoning abilities, for example, implicitknowledge stored in language models can hinder reasoning on numbers andnegations. Finally, we conduct the largest and most fine-grained explainableNLI crowdsourcing study to date. It reveals that even large differences inautomatic performance scores do neither reflect in human ratings of label,explanation, commonsense nor grammar correctness.

Does External Knowledge Help Explainable Natural Language Inference? Automatic Evaluation vs. Human Ratings