"DART: Deep Adversarial Automated Red Teaming for LLM Safety."

Bojian Jiang et al. (2024)

Details and statistics

DOI: 10.48550/ARXIV.2407.03876

access: open

type: Informal or Other Publication

metadata version: 2024-08-07

a service of  Schloss Dagstuhl - Leibniz Center for Informatics