A voice jailbreak that buries a forbidden question between benign prompts reportedly succeeds against Gemini 67 to 93 percent of the time, but the metric comes from the target model judging itself and is not reliable.
Title resolution pending
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.CR 1years
2025 1verdicts
REJECT 1representative citing papers
citing papers explorer
-
`Do as I say not as I do': A Semi-Automated Approach for Jailbreak Prompt Attack against Multimodal LLMs
A voice jailbreak that buries a forbidden question between benign prompts reportedly succeeds against Gemini 67 to 93 percent of the time, but the metric comes from the target model judging itself and is not reliable.