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 Tools annual

Global: Zendy Free Plan

Global: Zendy Tools monthly

Global: Zendy Plus monthly

Humans understand language by extracting information (meaning) fromsentences, combining it with existing commonsense knowledge, and thenperforming reasoning to draw conclusions. While large language models (LLMs)such as GPT-3 and ChatGPT are able to leverage patterns in the text to solve avariety of NLP tasks, they fall short in problems that require reasoning. Theyalso cannot reliably explain the answers generated for a given question. Inorder to emulate humans better, we propose STAR, a framework that combines LLMswith Answer Set Programming (ASP). We show how LLMs can be used to effectivelyextract knowledge -- represented as predicates -- from language. Goal-directedASP is then employed to reliably reason over this knowledge. We apply the STARframework to three different NLU tasks requiring reasoning: qualitativereasoning, mathematical reasoning, and goal-directed conversation. Ourexperiments reveal that STAR is able to bridge the gap of reasoning in NLUtasks, leading to significant performance improvements, especially for smallerLLMs, i.e., LLMs with a smaller number of parameters. NLU applicationsdeveloped using the STAR framework are also explainable: along with thepredicates generated, a justification in the form of a proof tree can beproduced for a given output.

Reliable Natural Language Understanding with Large Language Models and Answer Set Programming