mlsanityData to 14 Oct 2026

Learning to Prompt for Continual Learning

arXiv 2112.08654 ↗First posted 16 Dec 2021cs.LG cs.CVcited by 944

Not on the map. Why?

Cited by 97 in Multimodal LLMs / vision-language models, but placed by its own abstract, which didn’t fit.

Abstract (arXiv)

The mainstream paradigm behind continual learning has been to adapt the model parameters to non-stationary data distributions, where catastrophic forgetting is the central challenge. Typical methods rely on a rehearsal buffer or known task identity at test time to retrieve learned knowledge and address forgetting, while this work presents a new paradigm for continual learning that aims to train a more succinct memory system without accessing task identity at test time. Our method learns to dynamically prompt (L2P) a pre-trained model to learn tasks sequentially under different task transitions. In our proposed framework, prompts are small learnable parameters, which are maintained in a memory space. The objective is to optimize prompts to instruct the model prediction and explicitly manage task-invariant and task-specific knowledge while maintaining model plasticity. We conduct comprehensive experiments under popular image classification benchmarks with different challenging continual learning settings, where L2P consistently outperforms prior state-of-the-art methods. Surprisingly, L2P achieves competitive results against rehearsal-based methods even without a rehearsal buffer and is directly applicable to challenging task-agnostic continual learning. Source code is available at https://github.com/google-research/l2p.

Key facts. Cited by 944 papers in the map (data through 14 Oct 2026). Builds on 48 of its 68 references that are in the map. First posted on arXiv 16 Dec 2021.

Builds on

48 of 68 references.

References · 48

  1. 2017Attention Is All You Needcited by 60,361
  2. 2014Adam: A Method for Stochastic Optimizationcited by 37,240
  3. 2020An Image is Worth 16x16 Words: Transformers for Image Recognition at Scalecited by 30,186
  4. 2019Exploring the Limits of Transfer Learning with a Unified Text-to-Text Transformercited by 14,599
  5. 2021LoRA: Low-Rank Adaptation of Large Language Modelscited by 14,380
  6. 2017Towards Deep Learning Models Resistant to Adversarial Attackscited by 6,415
  7. 2016Overcoming catastrophic forgetting in neural networkscited by 4,787
  8. 2017Neural Discrete Representation Learningcited by 4,712
  9. 2017Fashion-MNIST: a Novel Image Dataset for Benchmarking Machine Learning Algorithmscited by 4,349
  10. 2021Prefix-Tuning: Optimizing Continuous Prompts for Generationcited by 3,915
  11. 2021The Power of Scale for Parameter-Efficient Prompt Tuningcited by 3,795
  12. 2016Learning without Forgettingcited by 2,496
36 more
  1. 2016iCaRL: Incremental Classifier and Representation Learningcited by 2,333
  2. 2021Pre-train, Prompt, and Predict: A Systematic Survey of Prompting Methods in Natural Language Processingcited by 2,234
  3. 2019A Survey on Bias and Fairness in Machine Learningcited by 1,917
  4. 2017Gradient Episodic Memory for Continual Learningcited by 1,898
  5. 2016Progressive Neural Networkscited by 1,361
  6. 2018Continual Lifelong Learning with Neural Networks: A Reviewcited by 1,330
  7. 2017Memory Aware Synapses: Learning what (not) to forgetcited by 1,105
  8. 2018Efficient Lifelong Learning with A-GEMcited by 1,058
  9. 2020Dark Experience for General Continual Learning: a Strong, Simple Baselinecited by 911
  10. 2017PackNet: Adding Multiple Tasks to a Single Network by Iterative Pruningcited by 898
  11. 2020AdapterFusion: Non-Destructive Task Composition for Transfer Learningcited by 788
  12. 2018Riemannian Walk for Incremental Learning: Understanding Forgetting and Intransigencecited by 742
  13. 2019Large Scale Incremental Learningcited by 732
  14. 2017Lifelong Learning with Dynamically Expandable Networkscited by 694
  15. 2018Overcoming catastrophic forgetting with hard attention to the taskcited by 681
  16. 2019Three scenarios for continual learningcited by 614
  17. 2019A Survey of Deep Learning Techniques for Autonomous Drivingcited by 403
  18. 2021DER: Dynamically Expandable Representation for Class Incremental Learningcited by 355
  19. 2020K-Adapter: Infusing Knowledge into Pre-Trained Models with Adapterscited by 319
  20. 2019Learn to Grow: A Continual Structure Learning Framework for Overcoming Catastrophic Forgettingcited by 302
  21. 2017CORe50: a New Dataset and Benchmark for Continuous Object Recognitioncited by 289
  22. 2021Online Continual Learning in Image Classification: An Empirical Surveycited by 245
  23. 2021Co2L: Contrastive Continual Learningcited by 244
  24. 2020Supermasks in Superpositioncited by 227
  25. 2020Using Hindsight to Anchor Past Knowledge in Continual Learningcited by 161
  26. 2018Towards Robust Evaluations of Continual Learningcited by 148
  27. 2019Continual Unsupervised Representation Learningcited by 147
  28. 2020Adversarial Continual Learningcited by 142
  29. 2018Memory Efficient Experience Replay for Streaming Learningcited by 128
  30. 2021Nested Hierarchical Transformer: Towards Accurate, Data-Efficient and Interpretable Visual Understandingcited by 97
  31. 2020Efficient Continual Learning with Modular Networks and Task-Driven Priorscited by 88
  32. 2018Task Agnostic Continual Learning Using Online Variational Bayescited by 61
  33. 2020Generalized Variational Continual Learningcited by 56
  34. 2020Learn-Prune-Share for Lifelong Learningcited by 32
  35. 2021Deep Bayesian Unsupervised Lifelong Learningcited by 21
  36. 2021Encoders and Ensembles for Task-Free Continual Learningcited by 17

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2022: 51; 2023: 166; 2024: 270; 2025: 262; 2026: 195.

944 papers.

Sort:

By area:

  1. 2022Visual Prompt Tuningcited by 1,681
  2. 2022Few-Shot Parameter-Efficient Fine-Tuning is Better and Cheaper than In-Context Learningcited by 896
  3. 2023A Comprehensive Survey of Continual Learning: Theory, Method and Applicationcited by 872
  4. 2022MaPLe: Multi-modal Prompt Learningcited by 687
  5. 2024Parameter-Efficient Fine-Tuning for Large Models: A Comprehensive Surveycited by 656
  6. 2022DualPrompt: Complementary Prompting for Rehearsal-free Continual Learningcited by 642
  7. 2022CODA-Prompt: COntinual Decomposed Attention-based Prompting for Rehearsal-Free Continual Learningcited by 450
  8. 2022Test-Time Prompt Tuning for Zero-Shot Generalization in Vision-Language Modelscited by 382
  9. 2023Orthogonal Subspace Learning for Language Model Continual Learningcited by 312
  10. 2025X-VLA: Soft-Prompted Transformer as Scalable Cross-Embodiment Vision-Language-Action Modelcited by 303
  11. 2024A Survey on Knowledge Distillation of Large Language Modelscited by 284
  12. 2023Self-regulating Prompts: Foundational Model Adaptation without Forgettingcited by 272
38 more
  1. 2022Aging with GRACE: Lifelong Model Editing with Discrete Key-Value Adaptorscited by 271
  2. 2022S-Prompts Learning with Pre-trained Transformers: An Occam's Razor for Domain Incremental Learningcited by 246
  3. 2024Continual Learning of Large Language Models: A Comprehensive Surveycited by 234
  4. 2023Class-Incremental Learning: A Surveycited by 209
  5. 2024InfLoRA: Interference-Free Low-Rank Adaptation for Continual Learningcited by 198
  6. 2023Revisiting Class-Incremental Learning with Pre-Trained Models: Generalizability and Adaptivity are All You Needcited by 184
  7. 2023Conditional Image-to-Video Generation with Latent Flow Diffusion Modelscited by 181
  8. 2023RSPrompter: Learning to Prompt for Remote Sensing Instance Segmentation based on Visual Foundation Modelcited by 181
  9. 2023SLCA: Slow Learner with Classifier Alignment for Continual Learning on a Pre-trained Modelcited by 180
  10. 2023TEMPO: Prompt-based Generative Pre-trained Transformer for Time Series Forecastingcited by 174
  11. 2023One-Dimensional Adapter to Rule Them All: Concepts, Diffusion Models and Erasing Applicationscited by 169
  12. 2023RanPAC: Random Projections and Pre-trained Models for Continual Learningcited by 164
  13. 2024Continual Learning with Pre-Trained Models: A Surveycited by 154
  14. 2023Continual Diffusion: Continual Customization of Text-to-Image Diffusion with C-LoRAcited by 141
  15. 2023A Unified Continual Learning Framework with General Parameter-Efficient Tuningcited by 125
  16. 2023Multimodal Prompting with Missing Modalities for Visual Recognitioncited by 121
  17. 2023Hierarchical Decomposition of Prompt-Based Continual Learning: Rethinking Obscured Sub-optimalitycited by 121
  18. 2024Expandable Subspace Ensemble for Pre-Trained Model-Based Class-Incremental Learningcited by 117
  19. 2023PromptIR: Prompting for All-in-One Blind Image Restorationcited by 114
  20. 2022Decorate the Newcomers: Visual Domain Prompt for Continual Test Time Adaptationcited by 98
  21. 2022Visual Prompt Tuning for Test-time Domain Adaptationcited by 97
  22. 2023Fine-tuning can cripple your foundation model; preserving features may be the solutioncited by 96
  23. 2023FeCAM: Exploiting the Heterogeneity of Class Distributions in Exemplar-Free Continual Learningcited by 87
  24. 2024UniST: A Prompt-Empowered Universal Model for Urban Spatio-Temporal Predictioncited by 87
  25. 2022ATTEMPT: Parameter-Efficient Multi-task Tuning via Attentional Mixtures of Soft Promptscited by 85
  26. 2022Continual Learning of Natural Language Processing Tasks: A Surveycited by 83
  27. 2023Exploring the Benefits of Training Expert Language Models over Instruction Tuningcited by 80
  28. 2023Learning without Forgetting for Vision-Language Modelscited by 79
  29. 2024Continual Learning and Catastrophic Forgettingcited by 77
  30. 2022P2P: Tuning Pre-trained Image Models for Point Cloud Analysis with Point-to-Pixel Promptingcited by 74
  31. 2024Towards Modular LLMs by Building and Reusing a Library of LoRAscited by 72
  32. 2023HIPTrack: Visual Tracking with Historical Promptscited by 71
  33. 2022LPT: Long-tailed Prompt Tuning for Image Classificationcited by 69
  34. 2023A Comprehensive Survey of Forgetting in Deep Learning Beyond Continual Learningcited by 69
  35. 2024MagMax: Leveraging Model Merging for Seamless Continual Learningcited by 67
  36. 2022CLIP model is an Efficient Continual Learnercited by 65
  37. 2023PILOT: A Pre-Trained Model-Based Continual Learning Toolboxcited by 64
  38. 2024Towards Lifelong Learning of Large Language Models: A Surveycited by 64

50 most cited of 944 All 944 (JSON)