The TruthfulQA benchmark evaluates large language models' susceptibility to factual hallucinations by presenting questions designed to provoke common misconceptions.

Authors

• Person: Not available Organization: Frontiers
  Survey and analysis of hallucinations in large language models

Sources

• Survey and analysis of hallucinations in large language models www.frontiersin.org Frontiers via serper

Referenced by nodes (1)

• TruthfulQA concept