Prompting-induced hallucinations in large language models often arise from ambiguous formulations or a lack of context, which causes the model to rely on probabilistic associations rather than grounded knowledge.

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 (2)

• Large Language Models concept
• prompt-induced hallucination concept