Abstract:
Few-Shot Class Incremental Learning (FSCIL) is a task that requires a model to learn new classes incrementally without forgetting when only a few samples for each class are given. FSCIL encounters two significant challenges: catastrophic forgetting and overfitting, and these challenges have driven prior studies to primarily rely on shallow models, such as ResNet-18. Even though their limited capacity can mitigate both forgetting and overfitting issues, it leads to inadequate knowledge transfer during few-shot incremental sessions. In this paper, we argue that To this end, we propose a novel FSCIL framework called , e-traned sion and anguage transformers with prompting functions and knowld distillation. Our framework effectively addresses the challenges of catastrophic forgetting and overfitting in large models through new pre-trained knowledge tuning (PKT) and two losses: entropy-based divergence loss and semantic knowledge distillation loss.Experimental results show that the proposed PriViLege significantly outperforms the existing state-of-the-art methods with a large margin, \textit{e.g.}, % in CUB200, % in CIFAR-100, and % in miniImageNet. The code will be publicly released after the review.
Chat is not available.