Adapter-based Continual Learning

Adapter-Based Continual Learning is a continual learning paradigm that employs lightweight, modular neural network adapters inserted into a pre-trained backbone model, where each adapter is selectively activated or updated to accommodate new tasks while preserving previously learned knowledge through parameter isolation and reuse.

4

Publications

190

Citations

19

Authors

6

Institutions

Publications per year

2017–2026

4

Authors

19

  1. GZ
    Giovanni Zappella

    University of Milan, Italy

    2 publications · 28 citations

    2

    h-index

  2. BE
    Beyza Ermiş

    Boğaziçi University, Türkiye

    2 publications · 28 citations

    2

    h-index

  3. MW
    Martin Wistuba

    University of Hildesheim, Germany

    2 publications · 28 citations

    2

    h-index

  4. JZ
    Jian Zhang

    University of Technology Sydney, Australia

    1 publications · 75 citations

    1

    h-index

  5. CA
    Cédric Archambeau

    University College London, United Kingdom

    1 publications · 16 citations

    1

    h-index

1–5 of 19

Venues

  1. 1

    h-index

  2. arXiv (Cornell University)

    Repository · 1 publications · 12 citations

    1

    h-index

1–2