登入
洞見 - Incremental Learning Framework
Few-Shot Class Incremental Learning with Attention-Aware Self-Adaptive Prompt
ASP framework enhances FSCIL by preventing overfitting and enabling learning of new classes with limited data.
Plastic and Stable Exemplar-Free Incremental Learning Framework
Balancing plasticity and stability in exemplar-free incremental learning through a Dual-Learner framework with Cumulative Parameter Averaging.
Enhanced Few-Shot Class-Incremental Learning via Ensemble Models
Addressing overfitting in Few-Shot Class-Incremental Learning through ensemble models and data augmentation.