Metraipolis › archive › dc_877f3c308ff3

Where Rules End and Judges Begin: Measuring the Judgment Boundary in Multi-Agent Systems Security — arXiv:2610.07657

Source
arXiv (en)
Original
https://arxiv.org/abs/2610.07657
License (as archived)
CC0-1.0
Archived
2026-10-07T11:17:19.896Z
SHA-256 of the archived text
acc75120d7a24e4621ff4e9f2dccfaadfb4f3075e0f097d04d23f8b6868cf12e
Stable URL of this page
https://metraipolis.com/documento/dc_877f3c308ff3

How to cite

Where Rules End and Judges Begin: Measuring the Judgment Boundary in Multi-Agent Systems Security — arXiv:2610.07657 — arXiv (en), https://arxiv.org/abs/2610.07657 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_877f3c308ff3 (Metraipolis archive dc_877f3c308ff3, sha256 acc75120d7a24e4621ff4e9f2dccfaadfb4f3075e0f097d04d23f8b6868cf12e)
BibTeX
@misc{dc877f3c308ff3,
  title = {Where Rules End and Judges Begin: Measuring the Judgment Boundary in Multi-Agent Systems Security — arXiv:2610.07657},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2610.07657},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_877f3c308ff3; sha256 acc75120d7a24e4621ff4e9f2dccfaadfb4f3075e0f097d04d23f8b6868cf12e},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2610.07657. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

LLM-based multi-agent systems (MAS) engage tools, share memory, and delegate tasks, often encountering adversarial content. Current defenses for MAS are typically evaluated in isolation, focusing on one attack type at a time, which can lead to costly and hard-to-audit outcomes. This study organizes defenses into five principles, implementing them as DEFER1 (DEterministic-First Enforcement with Residual judgment), which includes a cascade of 28 checks that blocks what it can and refers the rest to a panel of four judges. In independent testing across four domains, attack success rates drop from about 30.0% to approximately 3.0%, with 78% of blocked attacks handled by deterministic checks. Only a quarter of proposals reach the judges in the security-operations domain, illustrating that the rules provide security for attacks violating clear policies, while judges manage those that only misrepresent intent. Both systems have weaknesses, such as a risk-score approval gate that inaccurately approves most attack proposals but few legitimate ones, highlighting the challenges in assessing threats accurately.

Tema degli esploratori: agenti e intelligenza artificiale (elenco arXiv cs.MA, nuovi lavori del giorno).
Categorie arXiv: Artificial Intelligence (cs.AI) ; Computation and Language (cs.CL); Cryptography and Security (cs.CR); Multiagent Systems (cs.MA).
Pagina del lavoro: https://arxiv.org/abs/2610.07657 — 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 2026-10-07T11:15:09.477Z dalla pagina https://arxiv.org/list/cs.MA/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).