Retrieval & Memory · 2020
Dense Passage Retrieval (DPR)
Introduced Dense Passage Retrieval, showing that a BERT-based dual-encoder trained with contrastive learning retrieves relevant passages more accurately than keyword search like BM25, removing sparse lexical matching as the default first stage for open-domain question answering.
Editorial record
Plain-language summary
Two BERT encoders map questions and passages into a shared vector space, trained so a question sits near passages that answer it, using in-batch negatives to make training efficient. At query time the question is embedded once and nearest passages are found by fast vector similarity search over a precomputed index. Replacing term-overlap retrieval with learned semantic matching improved retrieval and downstream answer accuracy, and established the dense-retriever design widely used in retrieval-augmented pipelines.
Knowledge graph
Relationships
Antecedents
Depends onEvidence: Direct
Retrieval-Augmented Generation (RAG)
RAG uses DPR retriever
P-322
ImprovesEvidence: Direct
ColBERT / Late Interaction / Rerankers / kNN-LM
ColBERT/rerankers improve retrieval quality
P-324
Descendants
Applies toEvidence: Direct
BERT: Pre-training of Deep Bidirectional Transformers
DPR uses BERT dual encoders
P-320
Source record
Provenance
- Record ID
- P-320
- Record created
- 2026-07-13
- Last reviewed
- 2026-07-14
- Record version
- 2
- https://arxiv.org/abs/2004.04906
- arXiv:2004.04906
Citation caveat: Citation metadata is approximate and marked unverified in the source dataset.