notesum.ai

Published at October 20

Faster-GCG: Efficient Discrete Optimization Jailbreak Attacks against Aligned Large Language Models

cs.LG
cs.AI
physics.chem-ph
q-bio.BM

Released Date: October 20, 2024

Authors: Xiao Li1, Zhuhong Li2, Qiongxiu Li3, Bingze Lee1, Jinghao Cui1, Xiaolin Hu1

Aff.: 1Tsinghua University; 2Duke University; 3Aalborg University

Arxiv: https://arxiv.org/abs/2410.15362v1