Query pipeline optimization for cancer patient question answering systems
arXiv:2412.14751v2 Announce Type: replace
Abstract: Retrieval-augmented generation (RAG) mitigates hallucination in Large Language Models (LLMs) by using query pipelines to retrieve relevant external information and grounding responses in retrieved kn…