1 paper
Zlatan Feric, Amir Taherin, Yanzhi Wang +1
Retrieval-augmented generation (RAG) improves language-model responses by grounding generation in external passages, which comes with overhead: retrieved context lengthens the prom…