NEXUS: Structured Runtime Safety for Tool-Using LLM Agents
Researchers have introduced NEXUS, a structured runtime safety monitor for tool-using LLM agents, which applies a formal intervention policy to ensure safe execution of high-impact actions. NEXUS combines deterministic safety rules, argument-level inspection, and a calibrated logistic-regression risk score for graded escalation. The system has achieved high accuracy and low latency in various benchmarks, making it a promising solution for runtime safety monitoring.
- ▪NEXUS is a structured-plan safety monitor that applies a formal intervention policy to select among four actions: allow, block, request confirmation, or request revision.
- ▪The system combines deterministic safety rules, argument-level inspection, and a calibrated logistic-regression risk score for graded escalation.
- ▪NEXUS has achieved an F1 score of 0.949 and a 4-class intervention accuracy of 0.6406 on a 128-instance synthetic benchmark.
Opening excerpt (first ~120 words) tap to expand
Computer Science > Artificial Intelligence arXiv:2607.19356 (cs) [Submitted on 25 May 2026] Title:NEXUS: Structured Runtime Safety for Tool-Using LLM Agents Authors:Elias Hossain, Md Mehedi Hasan Nipu, Tasfia Nuzhat Ornee, Rajib Rana, Niloofar Yousefi View a PDF of the paper titled NEXUS: Structured Runtime Safety for Tool-Using LLM Agents, by Elias Hossain and 4 other authors View PDF HTML (experimental) Abstract:Tool-using LLM agents increasingly execute high-impact actions, making runtime safety monitoring essential. We present NEXUS (Neural EXecution Utility and Safety), a structured-plan safety monitor that applies a formal intervention policy to select among four actions: allow, block, request confirmation, or request revision.
…
Excerpt limited to ~120 words for fair-use compliance. The full article is at arXiv.org.
Discussion
0 commentsMore from arXiv.org
