English

Real-Time Open-Domain Question Answering with Dense-Sparse Phrase Index

Computation and Language 2019-06-17 v2

Abstract

Existing open-domain question answering (QA) models are not suitable for real-time usage because they need to process several long documents on-demand for every input query. In this paper, we introduce the query-agnostic indexable representation of document phrases that can drastically speed up open-domain QA and also allows us to reach long-tail targets. In particular, our dense-sparse phrase encoding effectively captures syntactic, semantic, and lexical information of the phrases and eliminates the pipeline filtering of context documents. Leveraging optimization strategies, our model can be trained in a single 4-GPU server and serve entire Wikipedia (up to 60 billion phrases) under 2TB with CPUs only. Our experiments on SQuAD-Open show that our model is more accurate than DrQA (Chen et al., 2017) with 6000x reduced computational cost, which translates into at least 58x faster end-to-end inference benchmark on CPUs.

Keywords

Cite

@article{arxiv.1906.05807,
  title  = {Real-Time Open-Domain Question Answering with Dense-Sparse Phrase Index},
  author = {Minjoon Seo and Jinhyuk Lee and Tom Kwiatkowski and Ankur P. Parikh and Ali Farhadi and Hannaneh Hajishirzi},
  journal= {arXiv preprint arXiv:1906.05807},
  year   = {2019}
}

Comments

ACL 2019; Code & demo available at https://nlp.cs.washington.edu/denspi/ ; Added comparison to Weaver (Raison et al., 2018)

R2 v1 2026-06-23T09:53:01.763Z