
PurpleLlama
Set of tools to assess and improve LLM security.
adversarial-attackai-securitycode-analysis+4
4.5k15h 11m ago

Set of tools to assess and improve LLM security.

Adversarial image attacks on vision-language web agents, from visual grounding to browser execution

A comprehensive set of fairness metrics for datasets and machine learning models, explanations for these metrics, and algorithms to mitigate bias in…

Project Mantis: Hacking Back the AI-Hacker; Prompt Injection as a Defense Against LLM-driven Cyberattacks

Lifetime AMSI bypass