Joshua Saxe (@joshua_saxe) 's Twitter Profile
Joshua Saxe

@joshua_saxe

AI+cybersecurity at Meta; past lives in academic history, labor / community organizing, classical/jazz piano, hacking scene

ID: 1397281496

linkhttps://www.malwaredatascience.com/ calendar_today02-05-2013 14:00:35

3,3K Tweet

3,3K Followers

1,1K Following

Joshua Saxe (@joshua_saxe) 's Twitter Profile Photo
Joshua Saxe
@joshua_saxe

a year ago

With today’s launch of Llama 3.1, we release CyberSecEval 3, a wide-ranging evaluation framework for LLM security used in the development of the models. Additionally, we introduce and improve three LLM security guardrails. Summary in this 🧵, links to paper/github at bottom:

With today’s launch of Llama 3.1, we release CyberSecEval 3, a wide-ranging evaluation framework for LLM security used in the development of the models. Additionally, we introduce and improve three LLM security guardrails. Summary in this 🧵, links to paper/github at bottom: