Research

Paper

AI LLM March 16, 2026

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

Authors

Andrea Bombarda, Silvia Bonfanti, Angelo Gargantini, Nico Pellegrinelli

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.

Metadata

arXiv ID: 2603.15375
Provider: ARXIV
Primary Category: cs.SE
Published: 2026-03-16
Fetched: 2026-03-17 06:02

Related papers

Raw Data (Debug)
{
  "raw_xml": "<entry>\n    <id>http://arxiv.org/abs/2603.15375v1</id>\n    <title>Formalizing and validating properties in Asmeta with Large Language Models (Extended Abstract)</title>\n    <updated>2026-03-16T14:50:39Z</updated>\n    <link href='https://arxiv.org/abs/2603.15375v1' rel='alternate' type='text/html'/>\n    <link href='https://arxiv.org/pdf/2603.15375v1' rel='related' title='pdf' type='application/pdf'/>\n    <summary>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.</summary>\n    <category scheme='http://arxiv.org/schemas/atom' term='cs.SE'/>\n    <published>2026-03-16T14:50:39Z</published>\n    <arxiv:primary_category term='cs.SE'/>\n    <author>\n      <name>Andrea Bombarda</name>\n    </author>\n    <author>\n      <name>Silvia Bonfanti</name>\n    </author>\n    <author>\n      <name>Angelo Gargantini</name>\n    </author>\n    <author>\n      <name>Nico Pellegrinelli</name>\n    </author>\n  </entry>"
}