Douwe KielaResearch Scientist Director · Google DeepMindResearcher↗3 shared research fields
Both at Google DeepMind

Researcher · Academic
Google DeepMindResearch Scientist
Riedel co-authored the 2020 paper on retrieval-augmented generation (RAG), a method that lets language models look up relevant documents before writing an answer.[1] His earlier work tested which facts models could recall from training alone[2]; he later helped develop a method for finding evidence spread across multiple documents[3].
Google DeepMindOct 2022-presentWork and education history is primarily focused on AI-relevant roles and may not be comprehensive.
RAG finds relevant documents to help a language model generate better-informed answers.
Lists Riedel as a co-author and describes how the models generate responses from retrieved Wikipedia passages.
Lists Riedel as a co-author and describes tests of factual knowledge learned during pretraining.
Lists Riedel as a co-author and describes retrieving evidence across passages for complex questions.
Based on public sources. If accuracy is important, verify details using the links or references.
For corrections, updates, or requests to remove information, contact us.
Reviewed August 8, 2026 · Checked for accuracy and new information.