EUVIMEDEUVIMED entdecken
Crossref

Response Attack: Exploiting Contextual Priming to Jailbreak Large Language Models

Miao Ziqi, Lijun Li, Yuan Xiong, Zhenhua Liu, Pengyu Zhu, Jing Shao

Proceedings of the AAAI Conference on Artificial Intelligence · 14.03.2026 · 40 (41) · 35284-35292

DOI: 10.1609/aaai.v40i41.40836 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Proceedings of the AAAI Conference on Artificial Intelligence, online erschienen am 14.03.2026. Verfasst von Miao Ziqi und 5 weiteren Autor:innen.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Contextual priming, where earlier stimuli covertly bias later judgments, offers an unexplored attack surface for large language models (LLMs). We uncover a contextual priming vulnerability in which the previous response in the dialogue can steer its subsequent behavior toward policy-violating content. While existing jailbreak attacks largely rely on single-turn or multi-turn prompt manipulations, or inject static in-context examples, these methods suffer from limited effectiveness, inefficiency, or semantic drift. We introduce Response Attack (RA), a novel framework that strategically leverages intermediate, mildly harmful responses as contextual primers within a dialogue. By reformulating harmful queries and injecting these intermediate responses before issuing a targeted trigger prompt, RA exploits a previously overlooked vulnerability in LLMs. Extensive experiments across eight state-of-the-art LLMs show that RA consistently achieves significantly higher attack success rates than nine leading jailbreak baselines. Our results demonstrate that the success of RA is directly attributable to the strategic use of intermediate responses, which induce models to generate more explicit and relevant harmful content while maintaining stealth, efficiency, and fidelity to the original query.

Zitieren

APA

Ziqi, M., Li, L., Xiong, Y., Liu, Z., Zhu, P., & Shao, J. (2026). Response Attack: Exploiting Contextual Priming to Jailbreak Large Language Models. Proceedings of the AAAI Conference on Artificial Intelligence, 40(41), 35284-35292. https://doi.org/10.1609/aaai.v40i41.40836

Vancouver

Ziqi M, Li L, Xiong Y, Liu Z, Zhu P, Shao J. Response Attack: Exploiting Contextual Priming to Jailbreak Large Language Models. Proceedings of the AAAI Conference on Artificial Intelligence. 2026;40(41):35284-35292. doi:10.1609/aaai.v40i41.40836

BibTeX
@article{Ziqi2026,
  author = {Ziqi, Miao and Li, Lijun and Xiong, Yuan and Liu, Zhenhua and Zhu, Pengyu and Shao, Jing},
  title = {Response Attack: Exploiting Contextual Priming to Jailbreak Large Language Models},
  journal = {Proceedings of the AAAI Conference on Artificial Intelligence},
  year = {2026},
  volume = {40},
  number = {41},
  pages = {35284--35292},
  doi = {10.1609/aaai.v40i41.40836}
}
Kostenlos starten →