Documents

Actions Registration Registration Fees Deadlines and Policies Submit Abstract Submit Paper Guidelines Preparing your Presentation Templates Glossary AI Tools Author's Login Reviewer's Login Ethics of Review Information Conference Details Important Dates Technical Program Social Event Call for Papers Program Committee Event Chairs Keynote Lectures Best Paper Awards Satellite Events Workshops Special Sessions Tutorials Demos Panels Travel and Accommodation Conference Venue About the Region Reaching the City Visa Information Hotel Reservation Partners Academic Partners Industrial Partners Institutional Partners Media Partners Partner Events Publication Partners Previous Conferences Websites Abstracts Invited Speakers Awards Books Published

Keynote Lectures

Reasoning at Higher-levels of Abstraction: AI Meets Model-Driven Engineering
Lola Burgueño, University of Malaga, Spain, Spain

Can AI Make the Rules? A Pipeline for Generating and Graph Transformation Systems
Reiko Heckel, University of Leicester, United Kingdom, United Kingdom

The Rise of AI-Native Telcos: Reimagining Networks, Control, and Value in the Age of Agents
Ali Tizghadam, TELUS Fellow / Chief Automation Architect / Senior Researcher and Lecturer at University of Toronto, Canada, Canada

Reasoning at Higher-levels of Abstraction: AI Meets Model-Driven Engineering

Lola Burgueño
University of Malaga, Spain

Brief Bio
Lola Burgueño is an Associate Professor at the University of Malaga (UMA), Spain. She is a member of the Atenea Research Group and part of the Institute of Technology and Software Engineering (ITIS).
Her research interests focus mainly on the fields of Software Engineering (SE) and of Model-Driven Software Engineering (MDE). She has made contributions to the application of artificial intelligence techniques to improve software development processes and tools; uncertainty management during the software design phase; model-based software testing, and the design of algorithms and tools for improving the performance of model transformations, among others.
She is an active member of the software engineering community. She has been PC (co-)chair of ECMFA'21, SLE'22, JISBD'25 and ICWE'26. She is the General Chair of MODELS'26. She is part of Program Committees and has served as co-chair and organizer in prestigious international conferences such as ASE, ICSE, MODELS, SLE, STAF, ICMT, ECMFA, ICSOC, etc. She regularly gives invited talks and lectures.
She is part of the Editorial Board of the Journal on Software and Systems Modeling (SoSyM), she is a member of the Steering Committee of SLE, MODELS and JISBD, and she is part of the WG on Diversity & Inclusion of Informatics Europe.

Abstract
As Artificial Intelligence (AI) continues to transform software engineering, there is increasing evidence that using AI in combination with Model-Driven Engineering (MDE) practices improves the reasoning capabilities of AI at higher levels of abstraction. This keynote explores how AI can enhance and automate the pillars of MDE - ranging from specific MDE tasks such as applying AI to model transformations to using AI for the very task of (meta-)modeling itself.
The talk first examines the inference of model transformations, where AI models learn the transformation logic directly from examples, eliminating the need for manually defined rules. Next, AI-assisted model autocompletion is introduced, which enables intelligent suggestions to be provided during modeling. The talk also touches on the advances in AI-based generation of model instances, which are crucial for testing, simulation, and design exploration. To conclude, an emerging framework for assessing LLM capabilities in generating domain-specific language (DSL) code is presented, together with the ongoing efforts to create benchmarks for evaluating the effectiveness of LLMs for performing modeling tasks.
As a whole, these contributions articulate a vision of AI-augmented MDE, as well as highlight the current challenges and future research directions needed to achieve robust, explainable, and trustworthy integration of AI within modeling practices.

Can AI Make the Rules? A Pipeline for Generating and Graph Transformation Systems

Reiko Heckel
University of Leicester, United Kingdom

Brief Bio
Reiko studied Computer Science at the Technical Universities of Dresden and Berlin and received his PhD (Dr.-Ing.) from the TU Berlin in 1998. He held academic positions at Paderborn and Dortmund before coming to the University of Leicester, where he is a Professor of Software Engineering.
Reiko is known for his work on model-based development, reengineering and testing, model transformations and the semantics of modelling languages. He published extensively on graph transformation, its concurrency theory, timed and stochastic variants, modularity and refinement.
Reiko is chair of the Steering Committees of the International Conference on Graph Transformation (ICGT) and the European Conference on Modelling Foundations and Applications (ECMFA), president of the European Association for Software Science and Technology (EASST) and a member of the IFIP WG 1.3 - Foundations of System Specification.

Abstract
We introduce a pipeline that turns informal requirements into executable graph-transformation rules by orchestrating multiple uses of generative AI. The process begins with LLM-generated natural-language rules, curated for correctness, which are then transformed into a JSON representation of graph transformation systems. A Python toolchain converts the rules into UML visualisations and subsequently into the GXL format used by the GROOVE graph rewriting engine for execution and debugging.
To understand the practical implications of such AI-assisted modelling, we conduct an experimental evaluation across four distinct development approaches : “vibe-coded” Python, an LLM-generated Python backend guided by detailed tax rule descriptions, direct use of the GPT API, and a fully model-based solution generated via our graph-transformation pipeline. Our findings expose strong contrasts in reliability, transparency, and maintainability—and show where structured modelling beats unconstrained code generation.

The Rise of AI-Native Telcos: Reimagining Networks, Control, and Value in the Age of Agents

Ali Tizghadam
TELUS Fellow / Chief Automation Architect / Senior Researcher and Lecturer at University of Toronto, Canada

Brief Bio
Ali Tizghadam is a TELUS Fellow and the Chief Automation Architect within the CTO team, spearheading TELUS’s transformation toward self-driving networks using the TM Forum Autonomous Networks (AN) framework. With over 25 years of experience in the telecom industry, Ali brings a unique blend of deep technical expertise and seasoned leadership, having served as VP of Innovation, Director of R&D, startup founder, and technical executive. His passion for AI and control systems began with his M.A.Sc thesis on stabilising an inverted pendulum using neural networks and fuzzy logic, and continued through his PhD work inventing a new metric to measure and control network robustness. Throughout his career, he has applied these innovations across transportation and telecommunications, leading major collaborations between academia, government, and industry. Ali also designed and led the development of TINAA, TELUS’s operating system for autonomous network operation, and continues to drive the architecture for intent-driven, closed-loop automation. He remains active in open source initiatives and standards bodies — including CAMARA (Telco APIs), IETF, MEF, IEEE, and ACM — with numerous contributions and publications, while maintaining strong academic ties through teaching and research in Autonomous Networks.

Abstract
Telecommunication networks are at a defining crossroads. Traditional telcos were built around a small set of long-lived, deterministic services — voice, Internet, and mobility — delivered to stable, human customers. With the advent of SDN, virtualization, and cloud-native design, networks began to serve a more dynamic ecosystem of services and partners, demanding agility, programmability, and intent-based control. Now, a new paradigm is emerging: networks must serve not only humans but also AI agents — digital entities that sense, decide, and act at machine speed. This shift requires a radical rethinking of how telcos operate, architect, and even define value. This keynote explores the evolution from models to intents to agents, outlining how Autonomous Networks (as defined by TM Forum) provide the structural foundation for this transformation. It discusses what it truly means for a telco to become AI-native — embedding intelligence throughout the network fabric, enabling self-learning assurance, and reimagining the business model around intent, trust, and adaptability. In an AI-native world, autonomy isn’t optional — it’s the only path to survival and relevance for telcos.