OpenAI’s SimpleQA to check language models’ factuality

OpenAI introduced a new benchmark named SimpleQA in a press release on Wednesday, which will help “measure the factuality of language models.””Current language models sometimes produce false outputs o…