Natural Language Access Control (NLAC): From Help Desk Requests to Structured Policies

2026-06-04

AI Summary

This work addresses the challenge of error-prone and expert-dependent configuration of access control policies in large-scale complex networks, where efficient and accurate automatic translation from natural language to formal policies remains lacking. The authors propose NLAC, a novel architecture that integrates large language models with a subgraph extraction mechanism to automatically translate user-specified natural language requests into structured access control policies. By leveraging embedding-based similarity retrieval to identify relevant network components, NLAC constructs compact subgraphs tailored to large-scale scenarios and introduces NLACBench, a new benchmark for evaluation. Experimental results demonstrate that the approach achieves 96.9% accuracy on small networks and, with subgraph-based policy generation, improves accuracy to 98.7% on large networks while significantly reducing inference time, hardware requirements, and operational costs. Further performance gains are attained through multi-model ensembling.

Abstract

Configuring network access control policies in large, complex networks is error-prone and requires significant expert effort. LLMs offer a promising interface for expressing such policies in natural language, but their capability for translating user requests into access policies, and the system architectures best suited to leverage LLMs, remain underexplored. We present an architecture for natural-language access control (NLAC) that uses LLMs to translate user requests into access policies, and introduce NLACBench, a benchmark for evaluating LLM-based intent translation systems in large-scale networks. Our evaluation across multiple state-of-the-art models shows that top-performing LLMs achieve up to 96.9% accuracy in small-network settings, but performance degrades substantially (below 20% for some models) as network size increases. To address this limitation, we identify relevant network components via embedding similarity and construct compact subgraphs that are passed to the LLM. This approach enables scaling to larger networks with up to 98.7% accuracy, while simultaneously reducing inference time, hardware requirements, and operating costs to a constant resource budget. Finally, a case study indicates that top-performing models exhibit largely complementary error patterns, suggesting that intent translation accuracy may be further improved through multi-LLM architectures.

Problem

• Natural Language Access Control
• LLM-based intent translation
• network access control policies
• large-scale networks
• NLAC

Innovation

• Natural Language Access Control
• LLM-based Policy Translation
• Network Subgraph Pruning
• NLACBench
• Multi-LLM Architecture