Get the latest tech news
Detecting hallucinations in large language models using semantic entropy
Hallucinations (confabulations) in large language model systems can be tackled by measuring uncertainty about the meanings of generated responses rather than the text itself to improve question-answering accuracy.
Answering unreliably or without the necessary information prevents adoption in diverse fields, with problems including fabrication of legal precedents 5 or untrue facts in news articles 6 and even posing a risk to human life in medical domains such as radiology 7. We distinguish this from cases in which a similar ‘symptom’ is caused by the following different mechanisms: when LLMs are consistently wrong as a result of being trained on erroneous data such as common misconceptions 13; when the LLM ‘lies’ in pursuit of a reward 14; or systematic failures of reasoning or generalization. We pursue this slightly indirect way of generating answers because we find that simply resampling each sentence creates variation unrelated to the uncertainty of the model about the factual claim, such as differences in paragraph structure.
Or read this on Hacker News
