Showing cs.LGShow all
2 papers · 1 filter
cs.LG2024
Attack On Prompt: Backdoor Attack in Prompt-Based Continual Learning
Trang Nguyen, Anh Tran, Nhat Ho
Prompt-based approaches offer a cutting-edge solution to data privacy issues in continual learning, particularly in scenarios involving multiple data suppliers where long-term stor…
cs.LG2024
Neural Collapse for Cross-entropy Class-Imbalanced Learning with Unconstrained ReLU Feature Model
Hien Dang, Tho Tran, Tan Nguyen +1
The current paradigm of training deep neural networks for classification tasks includes minimizing the empirical risk that pushes the training loss value towards zero, even after t…