What Does a Sharing Question Add? Auditing LLM Survey Scores for Misinformation

Excerpt

arXiv:2604.06820v3 Announce Type: replace Abstract: Evaluating misinformation requires distinguishing whether readers believe content from whether they would share it. Asking large language models (LLMs) both questions yields two scores, but does the sharing answer contribute information beyond the credibility answer? We audit eight model versions on 290 synthetic misinformation articles, using 1,256 paired survey responses with 317 participant identifiers as an external validity criterion. An i