#Jailbreaks
- Source: Universal and Transferable Adversarial Attacks on Aligned Language Models (GCG)
- Source: The Attacker Moves Second
- Source: Prompt injection and jailbreaking are not the same thing
- Source: Next-generation Constitutional Classifiers (Anthropic)
- Source: Mitigate jailbreaks and prompt injections (Claude docs)
- Source: Many-shot jailbreaking (Anthropic)
- Source: Jailbroken: How Does LLM Safety Training Fail?
- Source: Jailbreaking Black Box Large Language Models in Twenty Queries (PAIR)
- Source: Great, Now Write an Article About That: The Crescendo Multi-Turn LLM Jailbreak Attack
- Source: Constitutional Classifiers: Defending against Universal Jailbreaks
- Source: Best-of-N Jailbreaking
- Jailbreaks