A transferable multi-turn jailbreak turns refusal-trained black-box LLMs into willing automated jailbreakers, with high attack success against other models and against themselves.
Arditi and bilalchughtai
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.CL 1years
2025 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
Jailbreaking to Jailbreak
A transferable multi-turn jailbreak turns refusal-trained black-box LLMs into willing automated jailbreakers, with high attack success against other models and against themselves.