Pietro Barbiero

Interpretable Hierarchical Concept Reasoning through Graph Learning

Scalable Evaluation and Neural Models for Compositional Generalization

Deferring Concept Bottleneck Models: Learning to Defer Interventions to Inaccurate Experts

Causal Concept Graph Models: Beyond Causal Opacity in Deep Learning

FLUX: Efficient Descriptor-Driven Clustered Federated Learning under Arbitrary Distribution Shifts

Counterfactual Concept Bottleneck Models

Avoiding Leakage Poisoning: Concept Interventions Under Distribution Shifts

Causally Reliable Concept Bottleneck Models

Linearly Interpretable Concept Embedding Model for Text Classification