Retrieval-Augmented Generation for Knowledge-Intensive NLP Tasks

Open paper

Benchmark	Model	Rank	Results
fact-verification-on-fever	RAG	#5	Accuracy: 72.5
question-answering-on-webquestions	RAG	#5	EM: 45.2
question-answering-on-natural-questions	RAG	#22	EM: 44.5
question-answering-on-triviaqa	RAG	#33	EM: 56.1