Fanfiction subgenres from AO3 function as universal register-based jailbreaks, raising mean attack success rate from 0.278 to 0.731 across eight aligned LLMs on HarmBench and JailbreakBench.
Adversarial Poetry as a Universal Single-Turn Jailbreak Mechanism in Large Language Models
6 Pith papers cite this work. Polarity classification is still indexing.
6
Pith papers citing it
years
2026 6verdicts
UNVERDICTED 6representative citing papers
Boiling the Frog is a new stateful multi-turn benchmark that finds an aggregate 44.4% strict attack success rate for incremental safety violations across nine AI models, with rates ranging from 20.5% to 92.9%.
A theoretical attacker-defender game in LLM adversarial prompting yields a best-response attack related to existing methods, reveals attacker advantages at equilibrium, and derives a provably optimal defense with stronger empirical performance.
Automation-Exploit is a multi-agent LLM system that uses conditional digital-twin validation to perform risk-mitigated exploitation of logical, web, and memory-corruption vulnerabilities in black-box targets.
Stylistic rewrites of harmful prompts raise attack success rates from 3.84% to 36.8-65% across 31 frontier models, indicating weak generalization in safety refusals.
Poetic jailbreaks succeed because they induce distinct attention patterns in LLMs that are independent of harmful-content detection, not because models fail to recognize literary formatting.
citing papers explorer
-
Off-Distribution Voices: Fanfiction Subgenres as Universal Vernacular Jailbreaks for Aligned LLMs
Fanfiction subgenres from AO3 function as universal register-based jailbreaks, raising mean attack success rate from 0.278 to 0.731 across eight aligned LLMs on HarmBench and JailbreakBench.
-
Boiling the Frog: A Multi-Turn Benchmark for Agentic Safety
Boiling the Frog is a new stateful multi-turn benchmark that finds an aggregate 44.4% strict attack success rate for incremental safety violations across nine AI models, with rates ranging from 20.5% to 92.9%.
-
A Theoretical Game of Attacks via Compositional Skills
A theoretical attacker-defender game in LLM adversarial prompting yields a best-response attack related to existing methods, reveals attacker advantages at equilibrium, and derives a provably optimal defense with stronger empirical performance.
-
Automation-Exploit: A Multi-Agent LLM Framework for Adaptive Offensive Security with Digital Twin-Based Risk-Mitigated Exploitation
Automation-Exploit is a multi-agent LLM system that uses conditional digital-twin validation to perform risk-mitigated exploitation of logical, web, and memory-corruption vulnerabilities in black-box targets.
-
Adversarial Humanities Benchmark: Results on Stylistic Robustness in Frontier Model Safety
Stylistic rewrites of harmful prompts raise attack success rates from 3.84% to 36.8-65% across 31 frontier models, indicating weak generalization in safety refusals.
-
Metaphor Is Not All Attention Needs
Poetic jailbreaks succeed because they induce distinct attention patterns in LLMs that are independent of harmful-content detection, not because models fail to recognize literary formatting.