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

While there are numerous benchmarks comparing the performance of modernlanguage models (LMs), end-task evaluations often conflate notions of *factualaccuracy* ("truth") and *reasoning ability* ("rationality", or "honesty" in thesense of correctly reporting implications of beliefs). Our goal is a datasetthat clearly distinguishes these two notions. Our approach is to leverage andextend a collection of human-annotated *entailment trees*, engineered toexpress both good and bad chains of reasoning, and using a mixture of true andfalse facts, in particular including counterfactual examples, to avoid beliefbias (also known as the "content effect"). The resulting dataset, called BaRDa,contains 3000 entailments (1787 valid, 1213 invalid), using 6681 true and 2319false statements. Testing on four GPT-series models,GPT3(curie)/GPT3(davinici)/3.5/4, we find factual accuracy (truth) scores of74.1/80.6/82.6/87.1 and reasoning accuracy scores of 63.1/78.0/71.8/79.2. Thisshows the clear progression of models towards improved factual accuracy andentailment reasoning, and the dataset provides a new benchmark that morecleanly separates and quantifies these two notions.

BaRDa: A Belief and Reasoning Dataset that Separates Factual Accuracy and Reasoning Ability