Search Results for author:
Found 1 papers, 0 papers with code
Syntactic Ghost: An Imperceptible General-purpose Backdoor Attacks on Pre-trained Language Models
Specifically, the method hostilely manipulates poisoned samples with different predefined syntactic structures as stealth triggers and then implants the backdoor to pre-trained representation space without disturbing the primitive knowledge.
Cannot find the paper you are looking for? You can
Submit a new open access paper.
