notesum.ai

Published at November 28

PEFT-as-an-Attack! Jailbreaking Language Models during Federated Parameter-Efficient Fine-Tuning

cs.CR
cs.AI

Released Date: November 28, 2024

Authors: Shenghui Li1, Edith C. -H. Ngai, Fanghua Ye2, Thiemo Voigt3

Aff.: 1Uppsala University, Uppsala, Sweden; 2Tencent Inc., Shenzhen, China; 3Research Institutes of Sweden, Stockholm, Sweden

Arxiv: http://arxiv.org/pdf/2411.19335v1