Lorenzo Emer, PhD student, wins the "Best Student Paper Award" at NetSci 2026 in Boston

Lorenzo Emer, a PhD student in the National PhD Programme in Artificial Intelligence, has won the "Best Student Paper Award" at the International Conference NetSci 2026 in Boston with the paper "The hidden structure of innovation networks", presented in the satellite session Networks in Science of Science.

The paper's authors, alongside Emer, include his supervisors Andrea Mina and Andrea Vandin, together with Tiziano Squartini, Anna Gallo and Mattia Marzi from IMT.

The research, the result of a collaboration between Scuola Superiore Sant'Anna and IMT School for Advanced Studies Lucca, examines how innovation emerges from complex patterns of collaboration among inventors, firms and institutions. Drawing on patent data from three strategic sectors — artificial intelligence, biotechnology and semiconductors — the paper analyses co-invention and co-ownership networks to identify the mesoscopic structures around which inventive activity is organised.

The proposed method is based on the minimisation of the Bayesian Information Criterion within the Stochastic Block Model, a more refined approach compared to traditional modularity maximisation methods. The results show that inventor networks tend to be denser and more interconnected than those among organisations — consistent with the presence of small, recurring teams embedded within broader institutional hierarchies — while inter-organisational networks reveal sharper hierarchical structures, with a few central firms coordinating more peripheral ones.

The study also demonstrates that these structures are directly linked to the capacity to generate innovative impact: analysis of patent citations reveals a marked concentration of technological influence within a small number of clusters, suggesting that traditional methods may fail to fully capture the dynamics governing the diffusion of innovation across different technological domains.