Retrieval & Knowledge Integration (2019-2023)

Methods for augmenting LLMs with external knowledge and retrieval.

Retrieval-Augmented Generation

• Retrieval-Augmented Generation for Knowledge-Intensive NLP Tasks - RAG foundations
• Dense Passage Retrieval for Open-Domain Question Answering - DPR retriever
• When Not to Trust Language Models: Investigating Effectiveness of Parametric & Non-Parametric Memories - When to use retrieval

Vector Search & Embeddings

• Sentence-BERT: Sentence Embeddings using Siamese BERT-Networks - Efficient sentence embeddings
• Dense Retrieval for Open-Domain Question Answering - Dense retriever concepts

---

Key Insight: RAG enables LLMs to access and use external knowledge sources, improving accuracy on knowledge-intensive tasks.