Bridging Natural Language and Formal Specification--Automated Translation of Software Requirements to LTL via Hierarchical Semantics Decomposition Using LLMs

2025-12-19

AI Summary

Automated translation of natural-language requirements into formal Linear Temporal Logic (LTL) specifications for safety-critical domains—such as aerospace—remains hindered by semantic ambiguity, logical complexity, and limited generalizability of existing rule-based or learning-based approaches. This paper introduces Req2LTL, a novel framework featuring OnionL: a hierarchical intermediate representation that synergistically combines LLM-driven (GPT-4o) semantic decomposition with a domain-customized rule engine to enable deterministic, semantics-preserving LTL synthesis—ensuring both syntactic correctness and semantic fidelity. As a modular, end-to-end verifiable pipeline, Req2LTL achieves 100% syntactic correctness and 88.4% semantic accuracy on a real-world aerospace requirements benchmark, substantially outperforming prior methods. The framework establishes a new, industrially viable paradigm for formal verification—delivering high reliability, full interpretability, and robust automation.

Technology Category

• Natural Language Processing: Safety and Robustness
• Knowledge Representation and Reasoning: Geometric, Spatial, and Temporal Reasoning
• Machine Learning: Large Multimodal Models (LMMs)

Application Category

• Semantics and Knowledge: Data modeling to support human-machine intelligence, including LLMs agents, intelligent system behavior, explanations, and user-friendly interactions
• Economics, Online Markets and Human Computation: Cost models of using LLMs in production systems
• Search and Retrieval-Augmented AI: Search Tool Learning with LLM: Teaching LLMs to invoke search and make use of retrieved information

Abstract

Automating the translation of natural language (NL) software requirements into formal specifications remains a critical challenge in scaling formal verification practices to industrial settings, particularly in safety-critical domains. Existing approaches, both rule-based and learning-based, face significant limitations. While large language models (LLMs) like GPT-4o demonstrate proficiency in semantic extraction, they still encounter difficulties in addressing the complexity, ambiguity, and logical depth of real-world industrial requirements. In this paper, we propose Req2LTL, a modular framework that bridges NL and Linear Temporal Logic (LTL) through a hierarchical intermediate representation called OnionL. Req2LTL leverages LLMs for semantic decomposition and combines them with deterministic rule-based synthesis to ensure both syntactic validity and semantic fidelity. Our comprehensive evaluation demonstrates that Req2LTL achieves 88.4% semantic accuracy and 100% syntactic correctness on real-world aerospace requirements, significantly outperforming existing methods.

Problem

• Automates translation of natural language requirements to formal LTL specifications
• Addresses complexity and ambiguity in industrial software requirements
• Ensures syntactic validity and semantic fidelity via hierarchical decomposition

Innovation

• Hierarchical decomposition using LLMs for semantic extraction
• Modular framework combining LLMs with rule-based synthesis
• Ensures syntactic validity and semantic fidelity in translation