الملخص
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.
الكلمات المفتاحية
الموضوع
بيانات النشر
- المجلة
- غير متاح
- وصول مفتوح
- وصول مفتوح أخضر
اقتبس هذه المقالة
APA 7
Mitra, S., Patel, R., Neupane, S., Mittal, S., Rahman, M. R., & Rahimi, S. (2026). Where Rules End and Judges Begin: Measuring the Judgment Boundary in Multi-Agent Systems Security. https://omanscience.com/ar/articles/where-rules-end-and-judges-begin-measuring-the-judgment-boundary-in-multi-agent-systems-security
MLA 9
Mitra, Shaswata, et al. "Where Rules End and Judges Begin: Measuring the Judgment Boundary in Multi-Agent Systems Security." https://omanscience.com/ar/articles/where-rules-end-and-judges-begin-measuring-the-judgment-boundary-in-multi-agent-systems-security.
شيكاغو (المؤلف–التاريخ)
Mitra, Shaswata, Raj Patel, Subash Neupane, Sudip Mittal, Md Rayhanur Rahman, and Shahram Rahimi. 2026. "Where Rules End and Judges Begin: Measuring the Judgment Boundary in Multi-Agent Systems Security." https://omanscience.com/ar/articles/where-rules-end-and-judges-begin-measuring-the-judgment-boundary-in-multi-agent-systems-security.
هارفارد
Mitra, S., Patel, R., Neupane, S., Mittal, S., Rahman, M. R. and Rahimi, S. (2026) 'Where Rules End and Judges Begin: Measuring the Judgment Boundary in Multi-Agent Systems Security', Available at: https://omanscience.com/ar/articles/where-rules-end-and-judges-begin-measuring-the-judgment-boundary-in-multi-agent-systems-security.
فانكوفر
Mitra S, Patel R, Neupane S, Mittal S, Rahman MR, Rahimi S. Where Rules End and Judges Begin: Measuring the Judgment Boundary in Multi-Agent Systems Security. https://omanscience.com/ar/articles/where-rules-end-and-judges-begin-measuring-the-judgment-boundary-in-multi-agent-systems-security
IEEE
S. Mitra, R. Patel, S. Neupane, S. Mittal, M. R. Rahman, and S. Rahimi, "Where Rules End and Judges Begin: Measuring the Judgment Boundary in Multi-Agent Systems Security," https://omanscience.com/ar/articles/where-rules-end-and-judges-begin-measuring-the-judgment-boundary-in-multi-agent-systems-security.