mlsanityData to 7 Oct 2026

Task-Adaptive Saliency Guidance for Exemplar-free Class Incremental Learning

arXiv 2212.08251 ↗First posted 16 Dec 2022cs.CVcited by 8

Not on the map. Why?

Abstract (arXiv)

Exemplar-free Class Incremental Learning (EFCIL) aims to sequentially learn tasks with access only to data from the current one. EFCIL is of interest because it mitigates concerns about privacy and long-term storage of data, while at the same time alleviating the problem of catastrophic forgetting in incremental learning. In this work, we introduce task-adaptive saliency for EFCIL and propose a new framework, which we call Task-Adaptive Saliency Supervision (TASS), for mitigating the negative effects of saliency drift between different tasks. We first apply boundary-guided saliency to maintain task adaptivity and textitplasticity on model attention. Besides, we introduce task-agnostic low-level signals as auxiliary supervision to increase the textitstability of model attention. Finally, we introduce a module for injecting and recovering saliency noise to increase the robustness of saliency preservation. Our experiments demonstrate that our method can better preserve saliency maps across tasks and achieve state-of-the-art results on the CIFAR-100, Tiny-ImageNet, and ImageNet-Subset EFCIL benchmarks. Code is available at urlhttps://github.com/scok30/tass.

Key facts. Cited by 8 papers in the map (data through 7 Oct 2026). Builds on 35 of its 60 references that are in the map. First posted on arXiv 16 Dec 2022.

Builds on

35 of 60 references.

References · 35

  1. 2015Deep Residual Learning for Image Recognitioncited by 51,809
  2. 2016Grad-CAM: Visual Explanations from Deep Networks via Gradient-based Localizationcited by 6,690
  3. 2016Overcoming catastrophic forgetting in neural networkscited by 4,718
  4. 2016Learning without Forgettingcited by 2,465
  5. 2016iCaRL: Incremental Classifier and Representation Learningcited by 2,313
  6. 2017Gradient Episodic Memory for Continual Learningcited by 1,871
  7. 2018Efficient Lifelong Learning with A-GEMcited by 1,044
  8. 2017PackNet: Adding Multiple Tasks to a Single Network by Iterative Pruningcited by 882
  9. 2019Large Scale Incremental Learningcited by 727
  10. 2013An Empirical Investigation of Catastrophic Forgetting in Gradient-Based Neural Networkscited by 711
  11. 2018End-to-End Incremental Learningcited by 596
  12. 2018Learning to Learn without Forgetting by Maximizing Transfer and Minimizing Interferencecited by 511
23 more
  1. 2020Class-incremental learning: survey and performance evaluation on image classificationcited by 446
  2. 2019Dreaming to Distill: Data-free Knowledge Transfer via DeepInversioncited by 419
  3. 2020Semantic Drift Compensation for Class-Incremental Learningcited by 241
  4. 2019EGNet:Edge Guidance Network for Salient Object Detectioncited by 237
  5. 2019A Simple Pooling-Based Design for Real-Time Salient Object Detectioncited by 211
  6. 2019Data-Free Learning of Student Networkscited by 203
  7. 2014The Secrets of Salient Object Segmentationcited by 200
  8. 2020Modeling the Background for Incremental Learning in Semantic Segmentationcited by 189
  9. 2018Reinforced Continual Learningcited by 186
  10. 2019Class-incremental Learning via Deep Model Consolidationcited by 172
  11. 2021Always Be Dreaming: A New Approach for Data-Free Class-Incremental Learningcited by 130
  12. 2020Semantic Flow for Fast and Accurate Scene Parsingcited by 128
  13. 2018Rotate your Networks: Better Weight Consolidation and Less Catastrophic Forgettingcited by 127
  14. 2022FeTrIL: Feature Translation for Exemplar-Free Class-Incremental Learningcited by 120
  15. 2022Self-Sustaining Representation Expansion for Non-Exemplar Class-Incremental Learningcited by 118
  16. 2020A Comprehensive Study of Class Incremental Learning Algorithms for Visual Taskscited by 107
  17. 2022Overcoming Catastrophic Forgetting in Incremental Object Detection via Elastic Response Distillationcited by 72
  18. 2021Improving Deep Learning Interpretability by Saliency Guided Trainingcited by 68
  19. 2022Long-Tailed Class Incremental Learningcited by 36
  20. 2020Dynamic Feature Integration for Simultaneous Detection of Salient Object, Edge and Skeletoncited by 34
  21. 2020Remembering for the Right Reasons: Explanations Reduce Catastrophic Forgettingcited by 33
  22. 2020Self-Training for Class-Incremental Semantic Segmentationcited by 32
  23. 2020On Learning the Right Attention Point for Feature Enhancementcited by 1

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2025: 6; 2026: 2.

8 papers.

  1. 2025Language Guided Concept Bottleneck Models for Interpretable Continual Learningcited by 28
  2. 2025On the Discrimination and Consistency for Exemplar-Free Class Incremental Learningcited by 8
  3. 2025When Large Multimodal Models Confront Evolving Knowledge: Challenges and Explorationscited by 6
  4. 2025Exploring Open-world Continual Learning with Knowns-Unknowns Knowledge Transfercited by 2
  5. 2025Sculpting Margin Penalty: Intra-Task Adapter Merging and Classifier Calibration for Few-Shot Class-Incremental Learningcited by 0
  6. 2025KORE: Enhancing Knowledge Injection for Large Multimodal Models via Knowledge-Oriented Controlscited by 0
  7. 2026CI-CBM: Class-Incremental Concept Bottleneck Model for Interpretable Continual Learningcited by 0
  8. 2026Revisiting Prototype Rehearsal for Exemplar-Free Continual Learning: Manifold-Aware Boundary Sampling with Adaptive Class-Balanced Losscited by 0