Related work

The foundational work on continual learning, 1980 to 2026: methods, theory, benchmarks, surveys, and the neuroscience of memory. Updated weekly; some carry our notes.

27 papers of 8,653Sort Recent · Most cited
  1. 2026
    Lifelong Representations: A Survey on Continual Self-Supervised Learning for Vision ModelsSergi Masip, Alicja Dobrzeniecka, Jonathan Swinnen … Tinne TuytelaarsarXiv
    PDF ↗
  2. 2026
    Online Continual Learning with Dynamic Label HierarchiesXinrui Wang, Shao-Yuan Li, Bartłomiej Twardowski … Songcan ChenICML
    PDF ↗
  3. 2025
    Addressing The Devastating Effects Of Single-Task Data Poisoning In Exemplar-Free Continual LearningStanisław Pawlak, Bartłomiej Twardowski, T. P. Trzcinski, Joost van de WeijerarXiv · Warsaw University of Technology · Universitat Autònoma de Barcelona · +1
    PDF ↗
  4. 2025
    Query Drift Compensation: Enabling Compatibility in Continual Learning of Retrieval Embedding ModelsDipam Goswami, Liying Wang, Bartłomiej Twardowski, Joost van de WeijerarXiv
    PDF ↗
  5. 2024
    Category Adaptation Meets Projected Distillation in Generalized Continual Category DiscoveryGrzegorz Rypeść, Daniel Marczak, Sebastian Cygert … Bartłomiej TwardowskiECCV · Warsaw University of Technology · Gdańsk University of Technology · +1
    PDF ↗
  6. 2024
    Task-recency bias strikes back: Adapting covariances in Exemplar-Free Class Incremental LearningGrzegorz Rypeść, Sebastian Cygert, T. P. Trzcinski, Bartłomiej TwardowskiNeurIPS
    PDF ↗
  7. 2024
    Exemplar-free Continual Representation Learning via Learnable Drift CompensationAlex Gomez-Villa, Dipam Goswami, Kai Wang … Joost van de WeijerECCV
    PDF ↗
  8. 2024
    MagMax: Leveraging Model Merging for Seamless Continual LearningDaniel Marczak, Bartłomiej Twardowski, T. P. Trzcinski, Sebastian CygertECCV
    PDF ↗
  9. 2024PDF ↗
  10. 2024
    Resurrecting Old Classes with New Data for Exemplar-Free Continual LearningDipam Goswami, Albin Soutif--Cormerais, Yuyang Liu … Joost van de WeijerCVPR · University of Chinese Academy of Sciences
    PDF ↗
  11. 2025
    Improving Continual Learning Performance and Efficiency with Auxiliary ClassifiersFilip Szatkowski, Zheng, Yaoyue, Fei Yang … Joost van de WeijerICML
    PDF ↗
  12. 2024
    GUIDE: Guidance-based Incremental Learning with Diffusion ModelsBartosz Cywiński, Kamil Rafał Deja, T. P. Trzcinski … Łukasz KucińskiEuropean Conference on Artificial Intelligence
    PDF ↗
  13. 2024
    Divide and not forget: Ensemble of selectively trained experts in Continual LearningGrzegorz Rypeść, Sebastian Cygert, Valeriya Khan … Bartłomiej TwardowskiICLR · Warsaw University of Technology · Integrated Detector Electronics AS (Norway) · +6
    PDF ↗
  14. 2024
    Plasticity-Optimized Complementary Networks for Unsupervised Continual LearningAlex Gomez-Villa, Bartłomiej Twardowski, Kai Wang, Joost van de WeijerWACV · Universitat Autònoma de Barcelona · Barcelona Supercomputing Center · +1
    PDF ↗
  15. 2024
    Revisiting Supervision for Continual Representation LearningDaniel Marczak, Sebastian Cygert, T. P. Trzcinski, Bartłomiej TwardowskiECCV
    PDF ↗
  16. 2023PDF ↗
  17. 2023
    Bayesian Flow Networks in Continual LearningMateusz Pyla, Kamil Rafał Deja, Bartłomiej Twardowski, T. P. TrzcinskiarXiv
    PDF ↗
  18. 2023
    Adapt Your Teacher: Improving Knowledge Distillation for Exemplar-free Continual LearningFilip Szatkowski, Mateusz Pyla, Marcin Przewięźlikowski … T. P. TrzcinskiICCV · Warsaw University of Technology · Integrated Detector Electronics AS (Norway) · +5
    PDF ↗
  19. 2023
    Looking through the past: better knowledge retention for generative replay in continual learningValeriya Khan, Sebastian Cygert, Bartłomiej Twardowski, T. P. TrzcinskiICCV · Integrated Detector Electronics AS (Norway) · Corporación Universitaria de Colombia Ideas · +4
    PDF ↗
  20. 2023
    ICICLE: Interpretable Class Incremental Continual LearningDawid Rymarczyk, Joost van de Weijer, Bartosz Zieliński, Bartłomiej TwardowskiICCV · Jagiellonian University · Universitat Autònoma de Barcelona · +4
    PDF ↗
  21. 2023
    FeCAM: Exploiting the Heterogeneity of Class Distributions in Exemplar-Free Continual LearningDipam Goswami, Yuyang Liu, Bartłomiej Twardowski, Joost van de WeijerNeurIPS
    PDF ↗
  22. 2020
    Class-Incremental Learning: Survey and Performance Evaluation on Image ClassificationMarc Masana, Xialei Liu, Bartłomiej Twardowski … Joost van de WeijerTPAMI · Computer Vision Center
    PDF ↗
  23. 2022
    Continually Learning Self-Supervised Representations with Projected Functional RegularizationAlex Gomez-Villa, Bartłomiej Twardowski, Yu Lu … Joost van de WeijerCVPR · Universitat Autònoma de Barcelona · Tianjin University of Technology · +1
    PDF ↗
  24. 2021
    On the importance of cross-task features for class-incremental learningAlbin Soutif--Cormerais, Marc Masana, Joost van de Weijer, Bartłomiej TwardowskiarXiv
    PDF ↗
  25. 2020
    RATT: Recurrent Attention to Transient Tasks for Continual Image CaptioningRiccardo Del Chiaro, Bartłomiej Twardowski, Andrew D. Bagdanov, Joost van de WeijerNeurIPS
    PDF ↗
  26. 2020
    On Class Orderings for Incremental LearningMarc Masana, Bartłomiej Twardowski, Joost van de WeijerarXiv
    PDF ↗
  27. 2020
    Semantic Drift Compensation for Class-Incremental LearningLu Yu, Bartłomiej Twardowski, Xialei Liu … Joost van de WeijerCVPR · Universitat Autònoma de Barcelona · Northwestern Polytechnical University · +1
    PDF ↗
About this index

We keep this list because we read the field and wanted one place to see it. It covers work on continual learning itself, in the core areas of machine learning, and leaves out papers that apply it inside another field, such as medical imaging or fault diagnosis. By default it shows the papers we have a reason to trust: published at a venue like NeurIPS, ICML, ICLR, CVPR or TPAMI, or written by someone who has published there, or cited a few hundred times. The rest are one click away under “All papers”. It is seeded from the community lists kept by ContinualAI and by Xialei Liu, then filled out from OpenAlex, and every week a script looks for new papers on OpenAlex and arXiv. A model reads each candidate and decides whether it belongs; a person reviews the additions before they go live. Authors and affiliations come from OpenAlex, so a recent preprint can lack its institutions for a week or two.

Missing something, or filed under the wrong venue? Write to hello@unify.ai with the arXiv id or DOI.