mlsanityData to 14 Oct 2026

Online Versatile Incremental Learning: Towards Class and Domain-Agnostic Adaptation at Any Time

arXiv 2609.36442 ↗First posted 29 Sep 2026cs.CV cs.AIcited by 0

On the map: Diffusion models → Diffusion and flow policies for robotics also fits: Image generation, text-to-image and editing

Abstract (arXiv)

Continual learning enables vision systems to adapt to ever-changing data distributions. Despite significant advances, existing approaches fail to capture continuous and concurrent shifts in classes and domains, a critical capability for real-world deployment. This work introduces Online VIL (Online Versatile Incremental Learning), a novel scenario where class concepts and visual domains evolve simultaneously online without explicit boundaries. To better adapt to the challenges of such dynamic environments that more closely resemble real-world conditions, we propose a novel framework TopFlow, Topology preservation with Flow matching representation that contains two complementary mechanisms: Domain-agnostic Flow Matching (DFM) and Global Topology Preservation (GTP). DFM guides the model to have domain-agnostic representations by integrating the geodesic flow kernel into contrastive learning. In contrast, GTP maintains the global structure of the feature space without explicitly storing past examples. Our extensive experiments demonstrate that TopFlow effectively addresses the limitations of existing methods within the Online VIL scenario, achieving state-of-the-art performance in challenging Online VIL. The proposed methods suggest potential directions for building continual learning systems in realistic dynamic environments. Our implementation code is available at https://github.com/KU-VGI/Online-VIL.

Key facts. No citing papers in the map yet (data through 14 Oct 2026). Placed in Diffusion and flow policies for robotics, one of 14 directions in Diffusion models. Builds on 26 of its 42 references that are in the map. First posted on arXiv 29 Sep 2026.

Builds on

26 of 42 references.

References · 26

  1. 2020An Image is Worth 16x16 Words: Transformers for Image Recognition at Scalecited by 30,186
  2. 2017Prototypical Networks for Few-shot Learningcited by 3,015
  3. 2016Learning without Forgettingcited by 2,496
  4. 2018Experience Replay for Continual Learningcited by 1,095
  5. 2021Learning to Prompt for Continual Learningcited by 944
  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. 2017CORe50: a New Dataset and Benchmark for Continuous Object Recognitioncited by 289
  9. 2021Rainbow Memory: Continual Learning with a Memory of Diverse Samplescited by 273
  10. 2020Few-Shot Class-Incremental Learningcited by 251
  11. 2022S-Prompts Learning with Pre-trained Transformers: An Occam's Razor for Domain Incremental Learningcited by 246
  12. 2023SLCA: Slow Learner with Classifier Alignment for Continual Learning on a Pre-trained Modelcited by 180
14 more
  1. 2021Supervised Contrastive Replay: Revisiting the Nearest Class Mean Classifier in Online Class-Incremental Continual Learningcited by 141
  2. 2023A Unified Continual Learning Framework with General Parameter-Efficient Tuningcited by 125
  3. 2019Efficient Parameter-free Clustering Using First Neighbor Relationscited by 97
  4. 2022The CLEAR Benchmark: Continual LEArning on Real-World Imagerycited by 93
  5. 2021Online Continual Learning on Class Incremental Blurry Task Configuration with Anytime Inferencecited by 67
  6. 2020Continual Adaptation of Visual Representations via Domain Randomization and Meta-learningcited by 55
  7. 2023Online Prototype Learning for Online Continual Learningcited by 53
  8. 2024Pre-trained Vision and Language Transformers Are Few-Shot Incremental Learnerscited by 36
  9. 2023Online Class Incremental Learning on Stochastic Blurry Task Boundary via Mask and Visual Prompt Tuningcited by 35
  10. 2023CBA: Improving Online Continual Learning via Continual Bias Adaptorcited by 16
  11. 2024Dual Consolidation for Pre-Trained Model-Based Domain-Incremental Learningcited by 16
  12. 2023Prediction Error-based Classification for Class-Incremental Learningcited by 14
  13. 2024Versatile Incremental Learning: Towards Class and Domain-Agnostic Incremental Learningcited by 9
  14. 2024Enhancing Online Continual Learning with Plug-and-Play State Space Model and Class-Conditional Mixture of Discretizationcited by 7

Built on by

Not cited yet.

Similar, not cited

Close in wording, no citation either way

  1. 2023CTP: Towards Vision-Language Continual Pretraining via Compatible Momentum Contrast and Topology Preservation
  2. 2024Advancing Cross-domain Discriminability in Continual Learning of Vision-Language Models
  3. 2025StPR: Spatiotemporal Preservation and Routing for Exemplar-Free Video Class-Incremental Learning
  4. 2021Always Be Dreaming: A New Approach for Data-Free Class-Incremental Learning
  5. 2023Learning without Forgetting for Vision-Language Models
  6. 2025Modular Embedding Recomposition for Incremental Learning
  7. 2026Black-Box Continual Learning for Vision-Language Models
  8. 2026Prototype Latent World Model Replay for Class-Incremental Learning