Welcome!
We are part of the ART(ificial Intelligence) group at University of Rome Tor Vergata
Human Centric ART wants to put power back into human hands by pursuing a more reasonable interplay between deductive and inductive teaching and by repaying all the actors producing knowledge for these machines. Fortunately, there are others with this research objective, even if it is framed with different wordings, for example, neuro-symbolic AI, knowledge-enhanced neural networks, and knowledge-aware attentional neural networks.
To work on the above goal, Human Centric ART is currently exploring and will explore these areas:
- Accountability of machine learning inferences
- Actionable explainability for machine learning and neural networks
- Exploiting prior knowledge in neural networks
We are looking for passionate new PhD students, Postdocs, and Master students to join the team (more info) !
News
June 30, 2025
Don’t miss the Elena S. Ruzzetti and Fabio Massimo Zanzotto’s tutorial at IJCNN 2025.
May, 2025
3 Papers accepted at ACL 2025! Check out here the papers:
- Position Paper: MeMo: Towards Language Models with Associative Memory Mechanisms (Zanzotto et al., Findings 2025)
- R2-MultiOmnia: Leading Multilingual Multimodal Reasoning via Self-Training (Ranaldi et al., ACL 2025)
- Private Memorization Editing: Turning Memorization into a Defense to Strengthen Data Privacy in Large Language Models (Ruzzetti et al., ACL 2025)
We will partecipate in the L2M2 Workshop and we will present Protoknowledge Shapes Behaviour of LLMs in Downstream Tasks: Memorization and Generalization with Knowledge Graphs (Ranaldi et al., 2025)
Feel free to reach out in Vienna! See you there!