Formalizing and validating properties in Asmeta with Large Language Models (Extended Abstract)

TL;DR

This paper proposes integrating Large Language Models into the Asmeta framework to assist users in formalizing, explaining, and validating temporal logic properties, demonstrating feasibility through illustrative examples.

Contribution

It introduces a novel workflow that leverages LLMs within Asmeta for property formalization and validation, showcasing potential benefits and feasibility.

Findings

LLMs can effectively assist in formalizing temporal properties.

The integrated workflow supports explanation and validation tasks.

Proof of concept demonstrates practical feasibility.

Abstract

Writing temporal logic properties is often a challenging task for users of model-based development frameworks, particularly when translating informal requirements into formal specifications. In this paper, we explore the idea of integrating Large Language Models (LLMs) into the Asmeta framework to assist users during the definition, formalization, explanation, and validation of temporal properties. We present a workflow in which an LLM-based agent supports these activities by leveraging the Asmeta specification and the feedback produced by the model checker. This work serves as a proof of concept that illustrates the feasibility and potential benefits of such an integration through representative examples.