Metraipolis › archive › dc_83db0c1735ef

GuardrailAgent: An Agentic Framework for Jailbreak Defense — arXiv:2611.09734

Source
arXiv (en)
Original
https://arxiv.org/abs/2611.09734
License (as archived)
CC0-1.0
Archived
2026-11-09T22:40:35.591Z
SHA-256 of the archived text
72beb6e42a3cf74aaf9097efed8a04fae7abfb6f9ed98521268c237bf576fea8
Stable URL of this page
https://metraipolis.com/documento/dc_83db0c1735ef

How to cite

GuardrailAgent: An Agentic Framework for Jailbreak Defense — arXiv:2611.09734 — arXiv (en), https://arxiv.org/abs/2611.09734 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_83db0c1735ef (Metraipolis archive dc_83db0c1735ef, sha256 72beb6e42a3cf74aaf9097efed8a04fae7abfb6f9ed98521268c237bf576fea8)
BibTeX
@misc{dc83db0c1735ef,
  title = {GuardrailAgent: An Agentic Framework for Jailbreak Defense — arXiv:2611.09734},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2611.09734},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_83db0c1735ef; sha256 72beb6e42a3cf74aaf9097efed8a04fae7abfb6f9ed98521268c237bf576fea8},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2611.09734. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Despite undergoing safety training to align with human values, large language models (LLMs) remain vulnerable to jailbreak attacks that circumvent built-in safeguards and elicit malicious behaviors given by the adversary. Existing defenses are mostly static and fail to adapt to new or unknown attack strategies, providing limited resilience. To address these, we propose a multi-agent defense framework, GuardrailAgent, that offers adaptive, end-to-end protection. The system autonomously gathers and summarizes attack characteristics into a continuously updated knowledge base that supports both input- and output-side detection. A feedback module integrates misclassified cases for iterative improvement at minimal cost. Tested on over ten representative jailbreaks, the framework achieved a 93.1% input-side detection rate, 93.9% overall detection, and a 95.8% F1 score, while rapidly adapting to unseen attacks. These results demonstrate GuardrailAgent's outstanding capability to maintain LLM safety in the face of evolving jailbreak tactics.

Tema degli esploratori: fiducia digitale e crittografia (elenco arXiv cs.CR, nuovi lavori del giorno).
Categorie arXiv: Cryptography and Security (cs.CR).
Pagina del lavoro: https://arxiv.org/abs/2611.09734 — autori e testo completo sono lì (qui omessi i nomi: nessun dato personale nel deposito).
Licenza: titolo e abstract sono metadati arXiv distribuiti in CC0 1.0; il testo completo ha la licenza scelta dagli autori, indicata sulla pagina.
Letto il 2026-11-09T22:33:17.038Z dalla pagina https://arxiv.org/list/cs.CR/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).