"Improved Few-Shot Jailbreaking Can Circumvent Aligned Language Models and ..."

Xiaosen Zheng et al. (2024)

Details and statistics

DOI: 10.48550/ARXIV.2406.01288

access: open

type: Informal or Other Publication

metadata version: 2024-07-04

a service of  Schloss Dagstuhl - Leibniz Center for Informatics