Retrieval-augmented generation
Combining two memories rather than enlarging one. A pre-trained sequence-to-sequence model supplies the parametric memory; a dense vector index,
Official source: Retrieval-Augmented Generation for Knowledge-Intensive NLP Tasks →
reached through a neural retriever, supplies the non-parametric one. The founding work fine-tunes both together for generation, which is the distinction worth keeping: retrieval feeds generation here, instead of extracting an answer span the way earlier retrieval systems did.
Sources 1 source on record · Automated review 3 angles
Sources
What this page is based on — every source is verified and links out so you can check it yourself.
- Verified / official
- Provisional
- Out of date