SceneSplit jailbreaks text-to-video models by fragmenting harmful narratives into benign scenes that collectively constrain output to unsafe videos, reporting average attack success rates above 68% on five commercial T2V systems.
Title resolution pending
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.CV 1years
2025 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
Jailbreaking on Text-to-Video Models via Scene Splitting Strategy
SceneSplit jailbreaks text-to-video models by fragmenting harmful narratives into benign scenes that collectively constrain output to unsafe videos, reporting average attack success rates above 68% on five commercial T2V systems.