1991–, Computer scientist
Patrick Lewis is a British computer scientist whose 2020 paper at Facebook AI Research Retrieval-Augmented Generation for Knowledge-Intensive NLP Tasks (with Ethan Perez, Aleksandra Piktus, Fabio Petroni, Vladimir Karpukhin, Naman Goyal, Heinrich Küttler, Mike Lewis, Wen-tau Yih, Tim Rocktäschel, Sebastian Riedel and Douwe Kiela) introduced RAG, augmenting a generative model with a retrieval system that fetches relevant documents from a corpus and conditions generation on them.
RAG addresses two key limitations of pure parametric language models: limited training-time knowledge (which becomes stale and cannot be updated without retraining), and the inability to cite specific sources for factual claims. RAG and its many descendants (LLM-with-vector-search, retrieval-augmented chatbots, source-grounded generation) are now standard in production LLM systems.
Video
Works cited in this book:
- Retrieval-Augmented Generation for Knowledge-Intensive NLP Tasks (2020) (with Ethan Perez, Aleksandra Piktus, Fabio Petroni, Vladimir Karpukhin, Naman Goyal, Heinrich Küttler, Mike Lewis, Wen-tau Yih, Tim Rocktäschel, Sebastian Riedel, Douwe Kiela)
Discussed in:
- Chapter 15: Modern AI, Modern AI