AI Makes Human-Like Reasoning Mistakes

EurekAlert!

Large language models (LMs) can complete abstract reasoning tasks, but they are susceptible to many of the same types of mistakes made by humans. Andrew Lampinen, Ishita...