K. Pinel-Sauvagnat

RAGnRoll: Learning to Iteratively Retrieve and Generate Attributable Answer Snippets

This work proposes RAGnRoll, a language model for attributed answer generation within a multi-round Retrieval-Augmented Generation (RAG) framework that leverages the iterative nature of multi-round RAG to train an LLM to incrementally build answers guided by subqueries.