2 papers
cs.LG2026
Confine to the Task, Anchor the Rest: Concept-Guided Retention Regularizer for Parameter-Efficient Fine-Tuning
Ibna Kowsar, Ivan Medri, Fanyu Meng +1
Parameter-efficient fine-tuning (PEFT) has been widely used, yet the adapted model suffers catastrophic forgetting of pre-trained knowledge. While most existing work hardens the ad…
cs.CV2024
Diffusion Model Meets Non-Exemplar Class-Incremental Learning and Beyond
Jichuan Zhang, Yali Li, Xin Liu +1
Non-exemplar class-incremental learning (NECIL) is to resist catastrophic forgetting without saving old class samples. Prior methodologies generally employ simple rules to generate…