Model deepening denotes the process of transforming flat representations into multi-level models. A flat representation is characterized by at most two levels of classifications. Examples for flat representations include class-based object-oriented software code, database schemata with respective data values, conventional conceptual models created e.g. with UML. A multi-level model may contain an unbounded number of classification levels. Multi-level models have been argued to improve the reusability of models and integration among models.1 Though intended as generic contributions, much of the current work is tailored towards the multi-level-modeling language FMMLx2, that is supported by the XModelerML3.
The tools developed as part of this organization are intended as research instruments to investigate means to support automated model deepening. ModelDeepener serves as the flagship repository of the organization that ultimately should combine various analysis approaches for automated model deepening. The following publications reference repositories of this organization:
- Maier and Kadziolka (2026)4 investigates potentials of LLMs, using ChatGPT, to support automated model deepening. The research has unearthed the need to switch between representations of FMMLx, leading to the SimpleFMMLxValidator
- Maier (2026)5 outlines a model-deepening analysis approach based on instance data, instead of relying on type-level heuristics. The analysis approach has been implemented in ModelDeepener
Footnotes
-
Frank U (2022) Multi-Level Modeling: Cornerstones of a Rationale. Software and Systems Modeling 21:451–480 ↩
-
Frank U (2014) Multilevel Modeling: Toward a New Paradigm of Conceptual Modeling and Information Systems Design. Business and Information Systems Engineering 6(6):319–337 ↩
-
Maier P, Töpel D (2025) XModelerML v3: Integrating Executable UML with a Multi-Level Language Engineering, Modeling, and Execution Environment. ACM/IEEE 28th International Conference on Model Driven Engineering Languages and Systems Companion (MODELS-C) ↩
-
Maier P, Kadziolka V (2026) Model Deepening with Large Language Models: Insights from Exploratory Studies with ChatGPT. Advances in Conceptual Modeling: - ER 2025 Workshops, CMLS, FCM, LLM4Modeling, OntoCom, and QUAMES, Poitiers, France, October 20–23, 2025, Proceedings, pp 119–136 ↩
-
Maier P (2026) Property-Precedence Analysis for Model Deepening. Proceedings of the ACM/IEEE 29th International Conference on Model Driven Engineering Languages and Systems Companion (MODELS-C). Accepted for Publication ↩
