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/

Zendy Plus

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

Zendy Tools

Zendy Open

Large language models (LLMs) such as ChatGPT and Vicuna have shown remarkablecapacities in comprehending and producing language. However, their internalworkings remain a black box, and it is unclear whether LLMs and chatbots candevelop humanlike characteristics in language use. Cognitive scientists havedevised many experiments that probe, and have made great progress inexplaining, how people comprehend and produce language. We subjected ChatGPTand Vicuna to 12 of these experiments ranging from sounds to dialogue,preregistered and with 1000 runs (i.e., iterations) per experiment. ChatGPT andVicuna replicated the human pattern of language use in 10 and 7 out of the 12experiments, respectively. The models associated unfamiliar words withdifferent meanings depending on their forms, continued to access recentlyencountered meanings of ambiguous words, reused recent sentence structures,attributed causality as a function of verb semantics, and accessed differentmeanings and retrieved different words depending on an interlocutor's identity.In addition, ChatGPT, but not Vicuna, nonliterally interpreted implausiblesentences that were likely to have been corrupted by noise, drew reasonableinferences, and overlooked semantic fallacies in a sentence. Finally, unlikehumans, neither model preferred using shorter words to convey less informativecontent, nor did they use context to resolve syntactic ambiguities. We discusshow these convergences and divergences may result from the transformerarchitecture. Overall, these experiments demonstrate that LLMs such as ChatGPT(and Vicuna to a lesser extent) are humanlike in many aspects of human languageprocessing.

Do large language models resemble humans in language use?