2026

Making Generative Models Know What They Don't Know via Hypothesis Testing

Leander Kurscheidt, Antonio Vergari

ICML 2026 Workshop on Testing Poster

Hypothesis-testing-based uncertainty estimation for generative models, aimed at making model confidence better aligned with what the model truly knows.

[OpenReview]

Making Generative Models Know What They Don't Know via Hypothesis Testing

Leander Kurscheidt, Antonio Vergari

ICML 2026 Workshop on Testing Poster

Hypothesis-testing-based uncertainty estimation for generative models, aimed at making model confidence better aligned with what the model truly knows.

[OpenReview]

The Theory and Practice of MAP Inference over Non-Convex Constraints

Leander Kurscheidt, Gabriele Masina, Roberto Sebastiani, Antonio Vergari

ICML 2026 Poster

A study of MAP inference in the presence of non-convex constraints, spanning theoretical and practical aspects of structured probabilistic inference.

[OpenReview]

The Theory and Practice of MAP Inference over Non-Convex Constraints

Leander Kurscheidt, Gabriele Masina, Roberto Sebastiani, Antonio Vergari

ICML 2026 Poster

A study of MAP inference in the presence of non-convex constraints, spanning theoretical and practical aspects of structured probabilistic inference.

[OpenReview]

2025

A Probabilistic Neuro-symbolic Layer for Algebraic Constraint Satisfaction

Leander Kurscheidt, Paolo Morettin, Roberto Sebastiani, Andrea Passerini, Antonio Vergari

UAI 2025 Oral Runner-up Best Student Paper

A probabilistic neuro-symbolic approach for algebraic constraint satisfaction, connecting learned representations with structured reasoning.

[OpenReview]

A Probabilistic Neuro-symbolic Layer for Algebraic Constraint Satisfaction

Leander Kurscheidt, Paolo Morettin, Roberto Sebastiani, Andrea Passerini, Antonio Vergari

UAI 2025 Oral Runner-up Best Student Paper

A probabilistic neuro-symbolic approach for algebraic constraint satisfaction, connecting learned representations with structured reasoning.

[OpenReview]