Hallucination Detection in Large Language Models with Metamorphic Relations

Our summary

Introduces MetaQA, which uses metamorphic relations and prompt mutation to detect hallucinated or factually incorrect LLM outputs with no external resources, reporting gains over prior detectors across several models and datasets.

Why it matters

A worked application of metamorphic relations to a high-stakes failure mode — hallucination — showing the method's practical yield.

Cited by these methods

Related findings (2)

• Fabricated citations and referencesHigh

  Models invent plausible-looking but non-existent papers, authors, DOIs and URLs.

• Fabrication instead of admitting uncertaintyHigh

  Asked about something unknown or non-existent, models invent an answer rather than saying 'I don't know'.

Cite this