PromptCARE: Prompt Copyright Protection by Watermark Injection and Verification
Categories
5 个页面
论文
记录读论文后的学习笔记
Backdoor Defense via Deconfounded Representation Learning
Graph Unlearning
APMSA: Adversarial Perturbation Against Model Stealing Attacks
Feature Inference Attack on Model Predictions in Vertical Federated Learning
