Metraipolis › archive › dc_aa7bbcc8e1cf

Cumulative and Audited Guardrail Tool Evolution for LLM Agent Safety — arXiv:2611.02516

Source
arXiv (en)
Original
https://arxiv.org/abs/2611.02516
License (as archived)
CC0-1.0
Archived
2026-11-03T11:02:44.124Z
SHA-256 of the archived text
9ddc782ffa874307110c3398a3811c115ff996047d687832caf762490d66010b
Stable URL of this page
https://metraipolis.com/documento/dc_aa7bbcc8e1cf

How to cite

Cumulative and Audited Guardrail Tool Evolution for LLM Agent Safety — arXiv:2611.02516 — arXiv (en), https://arxiv.org/abs/2611.02516 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_aa7bbcc8e1cf (Metraipolis archive dc_aa7bbcc8e1cf, sha256 9ddc782ffa874307110c3398a3811c115ff996047d687832caf762490d66010b)
BibTeX
@misc{dcaa7bbcc8e1cf,
  title = {Cumulative and Audited Guardrail Tool Evolution for LLM Agent Safety — arXiv:2611.02516},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2611.02516},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_aa7bbcc8e1cf; sha256 9ddc782ffa874307110c3398a3811c115ff996047d687832caf762490d66010b},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2611.02516. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Large language model (LLM) agents increasingly operate in open-world environments where safety risks cannot be fully predefined. However, most existing guardrail mechanisms remain static, relying on fixed rules, prompts, or classifiers that generalize poorly to emerging risks. Here we present EVOLVE, a tool-level self-evolving defense framework that formulates agent security as a process of cumulative security evolution. Inspired by how human societies transform experiences into transferable protective knowledge, EVOLVE externalizes security knowledge into executable guardrail tools that are continuously inherited, independently audited, and iteratively refined through lifelong interaction. Across multiple benchmarks, backbone models, and embodied environments, EVOLVE consistently outperforms existing defenses across diverse risk scenarios with +22\% average F1 score. We further validate the framework in real-world robotic and autonomous agent systems. Our results suggest that scalable agent safety may require continuously evolving defense systems rather than static guardrails.

Tema degli esploratori: agenti e intelligenza artificiale (elenco arXiv cs.AI, nuovi lavori del giorno).
Categorie arXiv: Artificial Intelligence (cs.AI).
Pagina del lavoro: https://arxiv.org/abs/2611.02516 — 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 2026-11-03T10:59:45.826Z dalla pagina https://arxiv.org/list/cs.AI/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).