Prof. Dr. Harald Sack

As neuro-symbolic systems gain momentum by combining the learning capabilities of neural networks with the expressivity and interpretability of symbolic reasoning, they open up exciting opportunities for intelligent systems that can reason over structured knowledge. However, existing evaluation frameworks are heavily tailored to traditional symbolic systems and fall short in capturing the unique strengths and limitations of hybrid reasoning approaches. This project aims to design and implement a comprehensive benchmarking framework specifically for neuro-symbolic ontology reasoners, with a focus on core reasoning and querying tasks.