GAI4FIS@ICDE'26

Workshop Scope

The Workshop on Graph AI for Future Intelligent Systems (GAI4FIS) serves as a premier forum for advancing research at the intersection of graph-based artificial intelligence and next-generation intelligent systems. As modern intelligent systems, spanning cloud-native platforms, IoT ecosystems, and autonomous networks evolve to become increasingly complex, dynamic, and interconnected, they demand advanced intelligence capabilities for managing dependencies, optimizing workflows, and enabling robust, context-aware decision-making.

Graph AI, encompassing graph neural networks, graph transformers, knowledge graph reasoning, and neuro-symbolic models, has emerged as a transformative force driving innovation in the design and operation of future intelligent systems. Beyond revolutionary developments such as Graph Foundation Models, self-supervised learning, and the integration of Large Language Models (LLMs) with knowledge graph reasoning, GAI4FIS Workshop expands its focus to explore how these innovations can be operationalized in real-world intelligent service ecosystems.

Importance and Timeliness

Graph AI is rapidly becoming a cornerstone technology in critical sectors such as healthcare, transportation, energy, and smart living, where understanding complex relationships, dynamic interactions, and making trustworthy real-time decisions is essential. The GAI4FIS Workshop seeks to spotlight innovative research that addresses challenges in scalability, trustworthiness, explainability, and seamless integration into intelligent and service-oriented environments.

The workshop focuses on areas including automated service composition, adaptive orchestration, predictive maintenance, context-aware personalization, and secure inter-service communication. By combining advanced graph learning techniques with the principles of modular, distributed, and service-oriented architectures, GAI4FIS aims to push the boundaries of what future intelligent systems can achieve.

Topics Covered

GAI4FIS main topics include (but are not limited to):

Foundations and Advances in Graph AI

• Self-supervised and unsupervised graph representation learning
• Large Language Models for Knowledge Graphs
• Graph Neural Networks (scalability, fairness, Pre-trained GNNs)
• Graph AI with LLMs (LLM-Augmented Knowledge Graphs, Graph-Based LLM Fine-Tuning)
• AI-generated graphs (privacy-preserving graphs, graph data augmentation)
• Federated graph learning
• Diffusion Models for Graphs
• EdgeAI (e.g., GraphGPT)
• Retrieval-Augmented Generation (RAG) using knowledge graphs
• Prompt engineering with structured knowledge

Explainability, Robustness, and Trust in Graph AI

• Privacy-preserving graph representation learning
• Fairness-aware graph representation learning algorithms
• Robustness to adversarial attacks on graphs

Graph AI Applications

• Social networks and recommender systems
• Edge and service intelligence
• Graph-based Agentic AI and decentralized intelligence
• Smart city dimensions (intelligent transportation, smart healthcare)
• Legal, financial, and industrial graph applications
• Bioinformatics and drug discovery with molecular graphs

Graph AI for Service-Oriented Computing

• Intelligent graph-based modeling of service systems
• Generative graph models for service design and management
• Few-shot and Zero-shot learning on service graphs
• Graph foundation models for intelligent service recommendation and composition
• Graph AI and language models for cross-domain service integration
• Graph-aware event-driven architecture for service management and evolution
• Quantum graph algorithms for service optimization

Important Dates

Submission Instructions

Manuscripts should follow the latest IEEE proceedings format and be submitted as PDF files. The paper length should not exceed eight (8) pages inclusive of everything (e.g., references).

Publication in Proceedings

The proceedings of the workshops will be published jointly alongside with ICDE proceedings. All papers should have the same format as conference research papers.

Workshop Co-Chairs

Program Committee Members

Invited Talks

To be announced

Contact Information

For any inquiries regarding the workshop, please contact the workshop co-chairs:

• Wissem Inoubli: wissem.inoubli@univ-artois.fr
• Haithem Mezni: haithem.mezni@fsjegj.rnu.tn
• Sabeur Aridhi: sabeur.aridhi@loria.fr