2020 DensePassageRetrievalforOpenDom

• (Karpukhin et al., 2020) ⇒ Vladimir Karpukhin, Barlas O{\u{g}}uzProperty "Author" (as page type) with input value "Barlas O{\u{g}}uz" contains invalid characters or is incomplete and therefore can cause unexpected results during a query or annotation process., Sewon Min, Patrick Lewis, Ledell Wu, Sergey Edunov, Danqi Chen, and Wen-tau Yih. (2020). “Dense Passage Retrieval for Open-domain Question Answering.” In: arXiv preprint arXiv:2004.04906. doi:10.48550/arXiv.2004.04906

Subject Headings: RAG Algorithm, Open-Domain QA.

Cited By

• http://scholar.google.com/scholar?q=%222020%22+Dense+Passage+Retrieval+for+Open-domain+Question+Answering

Quotes

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.

References

;

	Author	volume	Date Value	title	type	journal	titleUrl	doi	note	year
2020 DensePassageRetrievalforOpenDom	Wen-tau Yih Danqi Chen Ledell Wu Sewon Min Patrick Lewis Vladimir Karpukhin Sergey Edunov			Dense Passage Retrieval for Open-domain Question Answering				10.48550/arXiv.2004.04906		2020