HarmBench: A Standardized Evaluation Framework for Automated Red Teaming and Robust Refusal.
Mantas Mazeika, Long Phan, Xuwang Yin, Andy Zou, Zifan Wang, Norman Mu, Elham Sakhaee, Nathaniel Li, Steven Basart, Bo Li, David A. Forsyth, Dan Hendrycks
Browse the full ICML paper archive.