Dense Passage Retrieval for Open-Domain Question Answering

Vladimir Karpukhin, Barlas Oğuz, Sewon Min, Patrick Lewis, Ledell Wu +3 more
4/10/2020
cs.CL

Official code: facebookresearch/DPR (1,869 stars, last commit 2023-04-06).

Abstract

Open-domain question answering relies on efficient passage retrieval to select candidate contexts, where traditional sparse vector space models, such as TF-IDF or BM25, are the de facto method. In this work, we show that retrieval can be practically implemented using dense representations alone, where embeddings are learned from a small number of questions and passages by a simple dual-encoder framework. When evaluated on a wide range of open-domain QA datasets, our dense retriever outperforms a strong Lucene-BM25 system largely by 9%-19% absolute in terms of top-20 passage retrieval accuracy, and helps our end-to-end QA system establish new state-of-the-art on multiple open-domain QA benchmarks.

View on arXivView PDF

Code Implementations(1)

facebookresearch/DPROfficial100%

Dense Passage Retriever - is a set of tools and models for open domain Q&A task.

1,869310PythonMay 13, 20203 years agoNOASSERTION

Similar Papers

Learning Dense Representations of Phrases at Scale

86% match1 repo6062020

Contextualized Sparse Representations for Real-Time Open-Domain Question Answering

85% match1 repo492019

Beyond Prompting: An Efficient Embedding Framework for Open-Domain Question Answering

84% match1 repo2025

Can a Frozen Pretrained Language Model be used for Zero-shot Neural Retrieval on Entity-centric Questions?

83% match2023

Merging Generated and Retrieved Knowledge for Open-Domain QA

82% match1 repo212023

Expand, Rerank, and Retrieve: Query Reranking for Open-Domain Question Answering

82% match1 repo382023

Cite this paper

@article{karpukhin2020dense,
  title  = {Dense Passage Retrieval for Open-Domain Question Answering},
  author = {Vladimir Karpukhin and Barlas Oğuz and Sewon Min and Patrick Lewis and Ledell Wu and Sergey Edunov and Danqi Chen and Wen-tau Yih},
  year   = {2020},
  eprint = {2004.04906},
  archivePrefix = {arXiv},
  url    = {https://arxiv.org/abs/2004.04906}
}

Add a badge to your README

Shows whether official code is linked for this paper.

Code of Paper badge
[![Code of Paper](https://codeofpaper.com/badge/2004.04906.svg)](https://codeofpaper.com/paper/2004.04906)

Discussion