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

Understanding what constitutes safe text is an important issue in naturallanguage processing and can often prevent the deployment of models deemedharmful and unsafe. One such type of safety that has been scarcely studied iscommonsense physical safety, i.e. text that is not explicitly violent andrequires additional commonsense knowledge to comprehend that it leads tophysical harm. We create the first benchmark dataset, SafeText, comprisingreal-life scenarios with paired safe and physically unsafe pieces of advice. Weutilize SafeText to empirically study commonsense physical safety acrossvarious models designed for text generation and commonsense reasoning tasks. Wefind that state-of-the-art large language models are susceptible to thegeneration of unsafe text and have difficulty rejecting unsafe advice. As aresult, we argue for further studies of safety and the assessment ofcommonsense physical safety in models before release.

SafeText: A Benchmark for Exploring Physical Safety in Language Models