all AI news
Prospects for inconsistency detection using large language models and sheaves. (arXiv:2401.16713v1 [cs.CY])
cs.CL updates on arXiv.org arxiv.org
We demonstrate that large language models can produce reasonable numerical
ratings of the logical consistency of claims. We also outline a mathematical
approach based on sheaf theory for lifting such ratings to hypertexts such as
laws, jurisprudence, and social media and evaluating their consistency
globally. This approach is a promising avenue to increasing consistency in and
of government, as well as to combating mis- and disinformation and related
ills.
arxiv cs.cy detection language language models large language large language models laws media numerical prospects ratings social social media theory