Late Interaction Models

Late Interaction Models is a class of neural information retrieval models that belong to the broader category of passage retrieval methods, distinguished by delayed computation of fine-grained interactions between query and document tokens until late in the encoding process, thereby balancing efficiency and contextual fidelity in relevance ranking.

14

Publications

2.2K

Citations

57

Authors

16

Institutions

Publications per year

2017–2026

11

Authors

57

  1. OK
    Omar Khattab

    Stanford University, United States

    4 publications · 1.5K citations

    4

    h-index

  2. SL
    Sheng-Chieh Lin

    University of Waterloo, Canada

    3 publications · 211 citations

    3

    h-index

  3. JL
    Jimmy Lin

    University of Waterloo, Canada

    3 publications · 211 citations

    3

    h-index

  4. MZ
    Matei Zaharia

    Stanford University, United States

    3 publications · 1.5K citations

    3

    h-index

  5. JY
    Jheng-Hong Yang

    University of Waterloo, Canada

    2 publications · 176 citations

    2

    h-index

1–5 of 57

Venues

  1. arXiv (Cornell University)

    Repository · 4 publications · 473 citations

    4

    h-index

  2. 1

    h-index

  3. 1

    h-index

  4. 1

    h-index

  5. Classical and Quantum Gravity

    1 publications · 116 citations

    1

    h-index

1–5