Sign up to save your podcasts
Or
Share JBDistill Generates Its Own Jailbreaks - 81.8% Attack Rate
Share to email
Share to Facebook
Share to X
Share to email
Share to Facebook
Share to X
Or
JBDistill Generates Its Own Jailbreaks - 81.8% Attack Rate
Johns Hopkins and Microsoft's JBDistill achieves 81.8% attack success rate across 13 LLMs by auto-generating fresh adversarial prompts on demand.
View all episodes
By Awesome AgentsJohns Hopkins and Microsoft's JBDistill achieves 81.8% attack success rate across 13 LLMs by auto-generating fresh adversarial prompts on demand.