Improving Communication of Changes in Model-Based Engineering with Model-Independent Change Descriptions

• URL: http://arxiv.org/abs/2607.21084v1
• Date: Thu, 23 Jul 2026 09:13:10 GMT
• Title: Improving Communication of Changes in Model-Based Engineering with Model-Independent Change Descriptions
• Abstract summary: In model-based engineering, inter-disciplinary teams collaborate through models, which change over time for purposes of system development.<n>Any change made by the engineer of one discipline will be difficult to understand by the engineers of other disciplines.<n>Model-independent change descriptions can be derived instead, which preserve semantics of the changes and do not require model-specific knowledge.
• Score: 1.241287366207879
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: In model-based engineering, inter-disciplinary teams collaborate through models, which change over time for purposes of system development, what makes the proper description of such changes crucial for engineers. However, any change made by the engineer of one discipline will be difficult to understand by the engineers of other disciplines. To overcome this limitation, model-independent change descriptions can be derived instead, which preserve semantics of the changes and do not require model-specific knowledge. The two opposing approaches here are to describe changes using either informal language or formal notions of change. While informal language lacks objectivity and standardisation, formal notions of change lack human interpretability, and thus offering no support for inter-disciplinary communication. In this paper, we propose functions to map formally specified changes, represented in the approach of delta modelling, to change descriptions in model-independent language. In an exhaustive mixed-methods evaluation, we bridge the gap between the theoretical and the practical representation of changes. We quantitatively assess technical feasibility, with an implementation framework, and technical applicability, along a case study; and we qualitatively assess plausibility, practical applicability, and extensibility, in a user study. Our work shows a promising starting point for automated, model-independent description of changes in model-based engineering projects.

Related papers

• Ontology-supported Design Parameter Management for Change Impact Analysis [65.57339762906196]
  OWL is used to represent information and the underlying data model can improve knowledge transfer among heterogeneous systems.<n>The approach is suitable for both software and hardware designs.
  arXiv  Detail & Related papers  (2026-08-22T13:25:31Z)
• A Six-Dimensional Taxonomy of Post-Training Adaptation Techniques with Applications in AI Governance [42.798415096963396]
  Post-training adaptation has become central to modern machine learning practice.<n>The literature remains fragmented across technique families, model classes, and deployment contexts.<n>This survey synthesizes the post-training adaptation literature and introduces a six-dimensional taxonomy.
  arXiv  Detail & Related papers  (2026-08-06T16:32:26Z)
• Model Science: getting serious about verification, explanation and control of AI systems [30.654411279722396]
  Model Science places the trained model at the core of analysis, aiming to interact, verify, explain, and control its behavior.<n>The proposed framework aims to guide the development of credible, safe, and human-aligned AI systems.
  arXiv  Detail & Related papers  (2025-08-27T16:50:17Z)
• Neural Network Reprogrammability: A Unified Theme on Model Reprogramming, Prompt Tuning, and Prompt Instruction [57.19302613163439]
  We introduce neural network reprogrammability as a unifying framework for model adaptation.<n>We present a taxonomy that categorizes such information manipulation approaches across four key dimensions.<n>We also analyze remaining technical challenges and ethical considerations.
  arXiv  Detail & Related papers  (2025-06-05T05:42:27Z)
• The Geometry of Prompting: Unveiling Distinct Mechanisms of Task Adaptation in Language Models [40.128112851978116]
  We study how different prompting methods affect the geometry of representations in language models.<n>Our analysis highlights the critical role of input distribution samples and label semantics in few-shot in-context learning.<n>Our work contributes to the theoretical understanding of large language models and lays the groundwork for developing more effective, representation-aware prompting strategies.
  arXiv  Detail & Related papers  (2025-02-11T23:09:50Z)
• Sparse autoencoders reveal selective remapping of visual concepts during adaptation [54.82630842681845]
  Adapting foundation models for specific purposes has become a standard approach to build machine learning systems.<n>We develop a new Sparse Autoencoder (SAE) for the CLIP vision transformer, named PatchSAE, to extract interpretable concepts.
  arXiv  Detail & Related papers  (2024-12-06T18:59:51Z)
• Generally-Occurring Model Change for Robust Counterfactual Explanations [1.3121410433987561]
  We study the robustness of counterfactual explanation generation algorithms to model changes. In this paper, we first generalize the concept of Naturally-Occurring Model Change. We also propose a more general concept of model parameter changes, Generally-Occurring Model Change.
  arXiv  Detail & Related papers  (2024-07-16T06:44:00Z)
• Adversarial Representation Engineering: A General Model Editing Framework for Large Language Models [7.41744853269583]
  We propose an Adversarial Representation Engineering (ARE) framework to provide a unified and interpretable approach for conceptual model editing. Experiments on multiple tasks demonstrate the effectiveness of ARE in various model editing scenarios.
  arXiv  Detail & Related papers  (2024-04-21T19:24:15Z)
• MergeNet: Knowledge Migration across Heterogeneous Models, Tasks, and Modalities [72.05167902805405]
  We present MergeNet, which learns to bridge the gap of parameter spaces of heterogeneous models.<n>The core mechanism of MergeNet lies in the parameter adapter, which operates by querying the source model's low-rank parameters.<n> MergeNet is learned alongside both models, allowing our framework to dynamically transfer and adapt knowledge relevant to the current stage.
  arXiv  Detail & Related papers  (2024-04-20T08:34:39Z)
• Explainability for Large Language Models: A Survey [59.67574757137078]
  Large language models (LLMs) have demonstrated impressive capabilities in natural language processing. This paper introduces a taxonomy of explainability techniques and provides a structured overview of methods for explaining Transformer-based language models.
  arXiv  Detail & Related papers  (2023-09-02T22:14:26Z)
• Foundation Models for Decision Making: Problems, Methods, and Opportunities [124.79381732197649]
  Foundation models pretrained on diverse data at scale have demonstrated extraordinary capabilities in a wide range of vision and language tasks. New paradigms are emerging for training foundation models to interact with other agents and perform long-term reasoning. Research at the intersection of foundation models and decision making holds tremendous promise for creating powerful new systems.
  arXiv  Detail & Related papers  (2023-03-07T18:44:07Z)
• Models we Can Trust: Toward a Systematic Discipline of (Agent-Based) Model Interpretation and Validation [0.0]
  We advocate the development of a discipline of interacting with and extracting information from models. We outline some directions for the development of a such a discipline.
  arXiv  Detail & Related papers  (2021-02-23T10:52:22Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.