Sparse Retrieval

Sparse Retrieval is a retrieval paradigm within information retrieval that operates on high-dimensional, sparsely populated representations—typically term-based vector spaces—where matching is determined by exact or weighted term overlap, distinguishing it from dense retrieval approaches that rely on continuous, low-dimensional embeddings to capture semantic similarity.

79

Publications

4.8K

Citations

271

Authors

99

Institutions

Publications per year

2017–2026

55

Variants

Sparse Information Retrieval

Authors

271

  1. JL
    Jimmy Lin

    University of Waterloo, Canada

    9 publications · 533 citations

    9

    h-index

  2. XM
    Xueguang Ma

    University of Waterloo, Canada

    6 publications · 475 citations

    6

    h-index

  3. JM
    Joel Mackenzie

    The University of Melbourne, Australia

    4 publications · 107 citations

    4

    h-index

  4. CX
    Chenyan Xiong

    Carnegie Mellon University, United States

    4 publications · 636 citations

    4

    h-index

  5. ZD
    Zhicheng Dou

    Renmin University of China, China

    3 publications · 118 citations

    3

    h-index

1–5 of 271

Venues

  1. arXiv (Cornell University)

    Repository · 9 publications · 604 citations

    9

    h-index

  2. 3

    h-index

  3. 3

    h-index

  4. Neural Networks

    2 publications · 167 citations

    2

    h-index

  5. Information Processing & Management

    2 publications · 23 citations

    2

    h-index

1–5