Understanding the causes of non-factual hallucinations in language models and proposing effective detection methods.
