← Все новости

Introducing SimpleQA

A factuality benchmark called SimpleQA that measures the ability for language models to answer short, fact-seeking questions.
Читать оригинал на OpenAI News →