Instruction Tuning

Instruction Tuning is a model-level optimization method that refines a pre-trained language model through supervised fine-tuning on collections of human-annotated instructions and their corresponding outputs to improve its ability to generalize across downstream tasks.

156

Publications

15K

Citations

1.1K

Authors

123

Institutions

Publications per year

2017–2026

156

Authors

1.1K

  1. CL
    Chunyuan Li

    Microsoft Research (United Kingdom), United Kingdom

    7 publications · 2.3K citations

    7

    h-index

  2. QX
    Qianqian Xie

    University of Manchester, United Kingdom

    5 publications · 263 citations

    5

    h-index

  3. HW
    Hyung Won Chung

    Massachusetts Institute of Technology, United States

    5 publications · 4.8K citations

    5

    h-index

  4. SL
    Shayne Longpre

    Massachusetts Institute of Technology, United States

    4 publications · 1.3K citations

    4

    h-index

  5. MS
    Minjoon Seo

    Korea Advanced Institute of Science and Technology, South Korea

    4 publications · 88 citations

    4

    h-index

1–5 of 1.1K

Venues

  1. arXiv (Cornell University)

    Repository · 95 publications · 6K citations

    37

    h-index

  2. Proceedings of the AAAI Conference on Artificial Intelligence

    Conference · 5 publications · 108 citations

    5

    h-index

  3. Research Square

    Repository · 2 publications · 108 citations

    2

    h-index

  4. Journal of the American Medical Informatics Association

    JAMIA · 2 publications · 71 citations

    2

    h-index

  5. Bioinformatics

    1 publications · 92 citations

    1

    h-index

1–5