Artificial contexts generated by LLMs improve accuracy in medical question answering, surpassing traditional retrieval methods.
