Incorporating Q&A Nuggets into Retrieval-Augmented Generation
Abstract
We present NagSys, a Nugget-Augmented Generation System that preserves explicit citation provenance by constructing a bank of Q&A nuggets from retrieved documents and uses them to guide extraction, selection, and report generation. Reasoning on nuggets avoids repeated information through clear and interpretable Q&A semantics—rather than opaque cluster abstractions—while maintaining citation provenance throughout the entire generation process. Evaluated on the TREC NeuCLIR 2024 collection, our NagSys system substantially outperforms Ginger, a recent nugget-based RAG system, in nugget recall, density, and citation grounding.
Type
Publication
ECIR
