DiSec: Mitigating Backdoors in Pre-trained Language Models via Disentanglement of Adversarial Weights for Secure Fine-Tuning.

Sunanda Das, Qinghua Li

Browse the full ACL paper archive.