Language models cannot reliably distinguish belief from knowledge and fact
Author: Suzgun, Mirac; Gur, Tayfun; Bianchi, Federico; Ho, Daniel E.; Icard, Thomas; Jurafsky, Dan; Zou, James Description: As language models (LMs) increasingly infiltrate into high-stakes domains such as law, medicine, journalism and science, their ability to distinguish belief from knowledge, and fact from fiction, becomes imperative. Failure to make such distinctions can mislead diagnoses, distort judicial judgments and amplify misinformation. Here we evaluate 24 cutting-edge LMs using a new KaBLE benchmark of 13,000 questions across 13 epistemic tasks. Our findings reveal crucial limitations. In particular, all models tested systematically fail…
See more and a link to full text