arXiv：cs.AI· Nikolaos Kekatos, Michael Ioannou, Panagiotis Katsaros, Alexios Lekidis, Theodoros Nestoridis, Tom Nianios, Dimitrios Nikou·· 6 小时前
LLM 辅助 ISR 集群的任务级运行时保障：面向验证感知消息层的三层组合验证框架
Mission-Level Runtime Assurance for LLM-Assisted ISR Swarms over a Verification-Aware Fabric
AI 导读
研究提出平台/小队/任务三层组合式运行时验证框架，将任务策略分解为单智能体与跨智能体切面，在验证感知消息层上聚合各平台判定结果。在模拟 ISR 任务中，一次间接提示词注入使真实 LLM 规划器将禁止的收集任务拆分到四个平台，所有单平台监控均未察觉，该框架却组合式检测出并给出完整溯源；注入故障时，尽力而为的中心监控会静默误报全域正常，验证感知消息层则不会。
正文
Abstract:Swarms of LLM-assisted autonomous robots are increasingly proposed for cooperative intelligence, surveillance, and reconnaissance (ISR) in contested environments. A growing class of their assurance failures arises not within any single platform but across the swarm: individually-compliant actions compose into a mission-level violation: a prohibited objective split across platforms to evade per-platform limits, or a collective budget quietly exceeded. Per-platform guardrails miss these by construction, and contested communications let the violation hide behind lost or delayed evidence. We present a three-tier (platform/squad/mission) compositional runtime-verification framework that decomposes a mission policy into per-agent and cross-agent aspects, aggregates per-platform verdicts over a verification-aware messaging fabric, and fuses them with an evidence-aware, two-axis (security x completeness) algebra whose provenance names the platforms that jointly triggered a violation. Because the fabric makes evidence loss and silence observable, unsupported negative verdicts are downgraded to an explicit unknown rather than reported as mission-wide all-clears. On a simulated ISR mission, an indirect prompt injection that causes real LLM planners to split a prohibited collection task across four platforms is invisible to every per-platform monitor yet detected compositionally with full provenance; under an injected fault campaign a best-effort central monitor emits silent false all-clears while the verification-aware fabric emits none.
| Subjects: | Cryptography and Security (cs.CR); Artificial Intelligence (cs.AI); Logic in Computer Science (cs.LO); Robotics (cs.RO) |
| Cite as: | arXiv:2607.23532 [cs.CR] |
| (or arXiv:2607.23532v3 [cs.CR] for this version) | |
| https://doi.org/10.48550/arXiv.2607.23532 arXiv-issued DOI via DataCite |
Submission history
From: Nikolaos Kekatos [view email]
[v1]
Sun, 26 Jul 2026 08:02:20 UTC (413 KB)
[v2]
Fri, 31 Jul 2026 04:50:20 UTC (413 KB)
[v3]
Sun, 11 Oct 2026 17:38:19 UTC (413 KB)
来源：arXiv：cs.AI · arxiv.org