Confidence-Guided Protocol IR for LLM-Aided Security Protocol Modeling

2026-09-29

AI Summary

This study addresses the insufficient semantic accuracy and limited expert trust encountered when large language models generate formal models of security protocols. To overcome these challenges, this work proposes a human-AI collaborative modeling framework. The method introduces an auditable protocol intermediate representation (IR) as a semantic checkpoint to explicitly capture essential elements, including participants, message flows, and cryptographic operations. Furthermore, it designs a model-confidence-based interactive interface that highlights uncertain fields, thereby guiding experts to precisely review core semantic decisions. This project achieves a reliable translation from natural language specifications into Tamarin verification models, significantly reducing expert validation overhead while enhancing both the efficiency and credibility of formal security protocol modeling.

Abstract

Large language models offer a promising interface for translating natural-language protocol descriptions into formal security models, but their outputs remain difficult to trust without expert validation. In this paper, we present a human-in-the-loop framework for generating Tamarin-verifiable formal models of security protocols. Our key observation is that the main correctness bottleneck is the semantic accuracy rather than the syntactic validity of the intermediate protocol representation. To address this problem, we introduce a protocol intermediate representation (IR) that serves as a human-auditable semantic checkpoint between natural-language parsing and formal model generation. The IR explicitly captures protocol participants, message flows, value provenance, cryptographic operations, proof targets, and compromise assumptions. We further design an interactive interface that highlights uncertain fields and guides users to inspect the most critical semantic decisions based on model confidence before model generation. Rather than replacing formal-methods experts, our approach uses LLMs to produce auditable semantic drafts while leveraging verification tools to check the resulting formal models. Code and verification artifacts are available at https://github.com/laplace1002/TamarinAgent.git.

Problem

• Large Language Models
• Security Protocol Modeling
• Formal Verification
• Intermediate Representation
• Semantic Accuracy

Innovation

• Protocol Intermediate Representation
• Confidence-Guided Interaction
• Human-in-the-Loop
• Security Protocol Modeling
• Tamarin Verification