GRAAI 2026
Workshop on Graphs Across AI 2026

Table of Contents

• Preface
  Summary: There were 14 papers submitted for peer-review to this workshop. Out of these, 6 papers were accepted for this workshop.

Keynote

• Invited Talk: Designing Graph Neural Networks as Dynamical Systems 1–1
  Davide Bacciu

Regular Papers

• Building the Dual Graph of the Activation Regions in a Deep Neural Network: What It Means for Generalization 2–19
  Sabrina M. Drammis, Karthik Srinivasan, Nancy A. Lynch, Robert Ajemian
• Efficient Path-Coherent Retrieval via Branch Pruning in KG-RAG 20–29
  Christopher Buratti, Michele Marchetti, Federica Parlapiano, Giulia Quaglieri, Davide Traini, Domenico Ursino, Luca Virgili
• Graph-Structured Operator Learning with Conformal Calibration for Airport Network Forecasting 30–42
  Soufiane Momtaz
• Investigating Sparse Artificial Neural Network Topology through Similarity and Balance scores 43–50
  Amirhossein Douzandeh Zenoozi, Christos Aronis, Antonio Liotta, Lucia Cavallaro
• Augmenting the Unified Cybersecurity Knowledge Graph Using RAG for Enhanced Threat Detection 51–63
  Diego Sanchez, Erdogan Dogdu, Roya Choupani
• Graph Neural Networks for Multi-Entity Access Modeling in Zero Trust Architectures 64–74
  Filippo Bernabucci, Gianluca Bonifazi, Andrea Rossi, Luca Spalazzi