1 paper
Alejandro Botas, Paul de Font-Reaulx, Luke Hewitt
Current AI models frequently exhibit epistemic sycophancy, endorsing claims to agree with a user. Existing evaluations typically measure this either by assessing what it takes to m…