Metraipolis › archive › dc_fe933eef5e12
No One Architecture Fits All: A Cross-Environment Evaluation of Hierarchical Red Team Agents — arXiv:2610.00557
- Source
- arXiv (en)
- Original
- https://arxiv.org/abs/2610.00557
- License (as archived)
- CC0-1.0
- Archived
- 2026-10-06T12:09:57.148Z
- SHA-256 of the archived text
0d1ec4f706f59442aee586f129164d8cbd61e26b446413a793329134adee991a- Stable URL of this page
https://metraipolis.com/documento/dc_fe933eef5e12
How to cite
No One Architecture Fits All: A Cross-Environment Evaluation of Hierarchical Red Team Agents — arXiv:2610.00557 — arXiv (en), https://arxiv.org/abs/2610.00557 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_fe933eef5e12 (Metraipolis archive dc_fe933eef5e12, sha256 0d1ec4f706f59442aee586f129164d8cbd61e26b446413a793329134adee991a)
BibTeX
@misc{dcfe933eef5e12,
title = {No One Architecture Fits All: A Cross-Environment Evaluation of Hierarchical Red Team Agents — arXiv:2610.00557},
howpublished = {arXiv (en)},
url = {https://arxiv.org/abs/2610.00557},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_fe933eef5e12; sha256 0d1ec4f706f59442aee586f129164d8cbd61e26b446413a793329134adee991a},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2610.00557. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
Autonomous red team agents increasingly stress-test AI-enabled cyber defenses by planning strategy and executing multistage attacks. Reinforcement learning (RL) and large language models (LLMs) offer complementary mechanisms for the planning and execution such agents require, and prior work has combined them in hybrid hierarchies. Yet a given architecture is typically developed and evaluated within a single environment, leaving open whether an observed advantage reflects a generally stronger decision mechanism or merely alignment with a particular setting. We address this gap with a controlled cross-environment comparison of two homogeneous hierarchical red team architectures: an RL planner with an RL executor (RL+RL) and an LLM planner with an LLM executor (LLM+LLM). We evaluate both against expert autonomous defenders in CybORG CAGE-4 and in Cyberwheel at two network scales, across 18 configurations under one unified disruption metric. We find a pronounced environment-dependent inversion. RL+RL wins the compact, densely rewarded CAGE-4 (78.5% disruption success versus 18.0% for the strongest LLM configuration) and the 100-host Cyberwheel network (81.0% versus 50.5%), while a pretrained cybersecurity LLM agent wins the larger, escalation-gated 1010-host Cyberwheel network (55.0% versus 0.0% for RL). A kill-chain analysis explains the inversion through architecture-specific bottlenecks that aggregate success rates this http URL the 1010-host Cyberwheel network, RL discovers and compromises hosts but stalls at privilege escalation, whereas in CAGE-4, LLM agents obtain privileged access but rarely convert it into operational impact. These results indicate that conclusions drawn in a single environment may not generalize, and that hybrid planner-executor designs should be motivated by specific failure modes rather than the assumption that one architecture is universally preferable. Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-10-02 in cs.AI). Categorie arXiv: Cryptography and Security (cs.CR) ; Artificial Intelligence (cs.AI). Pagina del lavoro: https://arxiv.org/abs/2610.00557 — 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 2026-10-06T10:21:28.080Z dalla pagina https://arxiv.org/catchup/cs.AI/2026-10-02?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).