HauntAttack embeds harmful instructions into reasoning-question conditions and reports a 70% average attack success rate across 11 large reasoning models, outperforming prior jailbreak baselines.
Google DeepMind
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.CR 1years
2025 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
HauntAttack: When Attack Follows Reasoning as a Shadow
HauntAttack embeds harmful instructions into reasoning-question conditions and reports a 70% average attack success rate across 11 large reasoning models, outperforming prior jailbreak baselines.