Abstract
AbstractMost experts agree that large language models (LLMs), such as those used by Copilot and ChatGPT, are expected to revolutionize the way in which software is developed. Many papers are currently devoted to analyzing the potential advantages and limitations of these generative AI models for writing code. However, the analysis of the current state of LLMs with respect to software modeling has received little attention. In this paper, we investigate the current capabilities of ChatGPT to perform modeling tasks and to assist modelers, while also trying to identify its main shortcomings. Our findings show that, in contrast to code generation, the performance of the current version of ChatGPT for software modeling is limited, with various syntactic and semantic deficiencies, lack of consistency in responses and scalability issues. We also outline our views on how we perceive the role that LLMs can play in the software modeling discipline in the short term, and how the modeling community can help to improve the current capabilities of ChatGPT and the coming LLMs for software modeling.
Publisher
Springer Science and Business Media LLC
Subject
Modeling and Simulation,Software
Cited by
16 articles.
订阅此论文施引文献
订阅此论文施引文献,注册后可以免费订阅5篇论文的施引文献,订阅后可以查看论文全部施引文献
1. Generative AI And Software Variability - A Research Vision;Proceedings of the 18th International Working Conference on Variability Modelling of Software-Intensive Systems;2024-02-07
2. Lessons learned from applying model-driven engineering in 5 domains: The success story of the MontiGem generator framework;Science of Computer Programming;2024-01
3. Adaptation of Enterprise Modeling Methods for Large Language Models;Lecture Notes in Business Information Processing;2023-11-25
4. An Assessment of ChatGPT on Log Data;AI-generated Content;2023-11-02
5. Quo Vadis modeling?;Software and Systems Modeling;2023-10-10