Two Best Paper Awards for Nicolas Hubert from the BIRD team at the ESWC 2024 conference.

Congratulations to Nicolas Hubert, PhD student in the Loria’s BIRD team and in the ERPI Laboratory, for receiving two Best Paper Awards at ESWC 2024 (Extended Semantic Web Conference), the leading international conference in the field of the Semantic Web!

• Best Student Research Paper Award for the article “Treat Different Negatives Differently: Enriching Loss Functions with Domain and Range Constraints for Link Prediction“, Nicolas Hubert, Pierre Monnin, Armelle Brun, Davy Monticolo.

This paper proposes to enrich loss functions for link prediction in knowledge graphs by taking into account the semantic validity of negative triples. In this paper, Nicolas Hubert shows that negative triples that comply with relation signatures (domains and ranges) are high-quality negatives, and that they experimentally enhance the semantic and predictive capabilities of knowledge graph embedding models.

• Best Resource Paper Award for the article “PyGraft: Configurable Generation of Synthetic Schemas and Knowledge Graphs at Your Fingertips“, Nicolas Hubert, Pierre Monnin, Mathieu d’Aquin, Davy Monticolo, Armelle Brun.

PyGraft is an open-source Python-based library for generating fully configurable, synthetic schemas and knowledge graphs. The library was developed by Nicolas Hubert and gathered much interest in both industrial and academic communities. PyGraft paves the way for novel studies and experiments, especially in the areas of synthetic data generation, model benchmarking, and neuro-symbolic AI.

• • PyGraft library is available on GitHub.