Dense Passage Retrieval for Open-Domain Question Answering

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.

PDF Abstract EMNLP 2020 PDF EMNLP 2020 Abstract

Results from the Paper

Task Dataset Model Metric Name Metric Value Global Rank Result Benchmark
Question Answering NaturalQA DPR EM 41.5 # 1
Question Answering Natural Questions DPR EM 41.5 # 15
Passage Retrieval Natural Questions DPR Precision@20 79.4 # 7
Precision@100 86 # 8
Question Answering SQuAD DPR EM 24.1 # 2
Question Answering TriviaQA DPR EM 56.8 # 21
Question Answering WebQuestions DPR EM 42.4 # 6


No methods listed for this paper. Add relevant methods here