Eliciting CLIP’s intrinsic attribute knowledge through a dual-cache guided mechanism for class-incremental learning
Authors:
- 29 June 2026
- journal article
- Published by Elsevier in Knowledge-Based Systems
- Vol. 350
Abstract
No abstract availableFunding Information
- National Natural Science Foundation of China
This publication has 31 references indexed in Scilit:
- MoTE: Mixture of task-specific experts for pre-trained model-based Class-incremental learningKnowledge-Based Systems, 2025
- Online task-free continual learning via discrepancy mechanismKnowledge-Based Systems, 2025
- Enhance the old representations’ adaptability dynamically for exemplar-free continual learningNeurocomputing, 2025
- Compositional diversity in visual concept learningCognition, 2024
- Learning to Prompt for Vision-Language ModelsInternational Journal of Computer Vision, 2022
- Transformer Interpretability Beyond Attention Visualization2021
- Prototype Augmentation and Self-Supervision for Incremental Learning2021
- DER: Dynamically Expandable Representation for Class Incremental Learning2021
- iTAML: An Incremental Task-Agnostic Meta-learning Approach2020
- iCaRL: Incremental Classifier and Representation Learning2017