Work

Language models cannot reliably distinguish belief from knowledge and fact

Year: 2025

Type: article

Source: Nature Machine Intelligence

DOI: 10.1038/s42256-025-01113-8

Authors: Mirac Suzgun, Tayfun Gur, Federico Bianchi, Daniel E. Ho, Thomas Icard, Dan Jurafsky, James Zou

Institutions: Stanford University, Duke University

Language: English

Topic: Logic, Reasoning, and Knowledge

Subfield: Artificial Intelligence

Field: Computer Science

Domain: Physical Sciences

Open Access status: green

Funders: Microsoft Research, Stanford Law School

OpenAlex ID: W4415816110

FWCI: 14.16

Cites: 20

Cited by: 55

Related to: 10