Francesco Giannini — accepted AI-conference papers

2026

DeepProofLog: Efficient Proving in Deep Stochastic Logic Programs

Ying Jiao, Rodrigo Castellano Ontiveros, Luc De Raedt, Marco Gori, Francesco Giannini, Michelangelo Diligenti +1

AAAI 2026technical

Neurosymbolic (NeSy) AI combines neural architectures and symbolic reasoning to improve accuracy, interpretability, and generalization. While logic inference on top of subsymbolic modules has been shown to effectively guarantee these properties, this often comes at the cost of reduced scalability, w

Cited by 0 Source ↗PDF ↗Scholar ↗

2026

Mixture of Concept Bottleneck Experts

Francesco De Santis, Gabriele Ciravegna, Giovanni De Felice, Arianna Casanova, Francesco Giannini, Michelangelo Diligenti +4

ICML 2026spotlight

Concept Bottleneck Models (CBMs) promote interpretability by grounding predictions in human-understandable concepts. However, existing CBMs typically fix their task predictor to a single linear or Boolean expression, limiting both predictive accuracy and adaptability to diverse user needs. We propos…

Cited by 0 Source ↗Scholar ↗

2025

Counterfactual Concept Bottleneck Models

Gabriele Dominici, Pietro Barbiero, Francesco Giannini, Martin Gjoreski, Giuseppe Marra, Marc Langheinrich

ICLR 2025poster

Current deep learning models are not designed to simultaneously address three fundamental questions: predict class labels to solve a given classification task (the "What?"), simulate changes in the situation to evaluate how this impacts class predictions (the "How?"), and imagine how the scenario sh…

Cited by 2 Source ↗PDF ↗code ↗Scholar ↗

2025

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

Andrea Pugnana, Riccardo Massidda, Francesco Giannini, Pietro Barbiero, Mateo Espinosa Zarlenga, Roberto Pellungrini +3

NeurIPS 2025poster

Concept Bottleneck Models (CBMs) are interpretable machine learning models that ground their predictions on human-understandable concepts, allowing for targeted interventions in their decision-making process. However, when intervened on, CBMs assume the availability of humans that can identify the n…

Cited by 0 Source ↗Scholar ↗

2025

Grounding Methods for Neural-Symbolic AI

Rodrigo Castellano Ontiveros, Francesco Giannini, Marco Gori, Giuseppe Marra, Michelangelo Diligenti

IJCAI 2025

A large class of Neural-Symbolic (NeSy) methods employs a machine learner to process the input entities, while relying on a reasoner based on First-Order Logic to represent and process more complex relationships among the entities. A fundamental role for these methods is played by the process of log

Cited by 0 Source ↗PDF ↗code ↗Scholar ↗

2024

Interpretable Concept-Based Memory Reasoning

David Debot, Pietro Barbiero, Francesco Giannini, Gabriele Ciravegna, Michelangelo Diligenti, Giuseppe Marra

NeurIPS 2024poster

The lack of transparency in the decision-making processes of deep learning systems presents a significant challenge in modern artificial intelligence (AI), as it impairs users’ ability to rely on and verify these systems. To address this challenge, Concept Bottleneck Models (CBMs) have made signific…

Cited by 6 Source ↗PDF ↗code ↗Scholar ↗

2024

Relational Concept Bottleneck Models

Pietro Barbiero, Francesco Giannini, Gabriele Ciravegna, Michelangelo Diligenti, Giuseppe Marra

NeurIPS 2024poster

The design of interpretable deep learning models working in relational domains poses an open challenge: interpretable deep learning methods, such as Concept Bottleneck Models (CBMs), are not designed to solve relational problems, while relational deep learning models, such as Graph Neural Networks (…

Cited by 4 Source ↗PDF ↗code ↗Scholar ↗

2023

Interpretable Graph Networks Formulate Universal Algebra Conjectures

Francesco Giannini, Stefano Fioravanti, Oguzhan Keskin, Alisia Maria Lupidi, Lucie Charlotte Magister, Pietro Lio +1

NeurIPS 2023poster

The rise of Artificial Intelligence (AI) recently empowered researchers to investigate hard mathematical problems which eluded traditional approaches for decades. Yet, the use of AI in Universal Algebra (UA)---one of the fields laying the foundations of modern mathematics---is still completely unexp…

Cited by 6 Source ↗PDF ↗Scholar ↗

2023

Interpretable Neural-Symbolic Concept Reasoning

Pietro Barbiero, Gabriele Ciravegna, Francesco Giannini, Mateo Espinosa Zarlenga, Lucie Charlotte Magister, Alberto Tonda +4

ICML 2023poster

Deep learning methods are highly accurate, yet their opaque decision process prevents them from earning full human trust. Concept-based models aim to address this issue by learning tasks based on a set of human-understandable concepts. However, state-of-the-art concept-based models rely on high-dime…

Cited by 28 Source ↗PDF ↗code ↗Scholar ↗

2022

Concept Embedding Models: Beyond the Accuracy-Explainability Trade-Off

Mateo Espinosa Zarlenga, Pietro Barbiero, Gabriele Ciravegna, Giuseppe Marra, Francesco Giannini, Michelangelo Diligenti +6

NeurIPS 2022accept

Deploying AI-powered systems requires trustworthy models supporting effective human interactions, going beyond raw prediction accuracy. Concept bottleneck models promote trustworthiness by conditioning classification tasks on an intermediate level of human-like concepts. This enables human intervent…

Cited by 179 Source ↗PDF ↗code ↗Scholar ↗

2022

Entropy-Based Logic Explanations of Neural Networks

Pietro Barbiero, Gabriele Ciravegna, Francesco Giannini, Pietro Lió, Marco Gori, Stefano Melacci

AAAI 2022technical

Explainable artificial intelligence has rapidly emerged since lawmakers have started requiring interpretable models for safety-critical domains. Concept-based neural networks have arisen as explainable-by-design methods as they leverage human-understandable symbols (i.e. concepts) to predict class m…

Cited by 105 Source ↗PDF ↗code ↗Scholar ↗

2022

Extending Logic Explained Networks to Text Classification

Rishabh Jain, Gabriele Ciravegna, Pietro Barbiero, Francesco Giannini, Davide Buffelli, Pietro Lio

EMNLP 2022main

Recently, Logic Explained Networks (LENs) have been proposed as explainable-by-design neural models providing logic explanations for their predictions.However, these models have only been applied to vision and tabular data, and they mostly favour the generation of global explanations, while local on…

Cited by 15 Source ↗PDF ↗Scholar ↗

2020

Human-Driven FOL Explanations of Deep Learning

Gabriele Ciravegna, Francesco Giannini, Marco Gori, Marco Maggini, Stefano Melacci

IJCAI 2020poster

Deep neural networks are usually considered black-boxes due to their complex internal architecture, that cannot straightforwardly provide human-understandable explanations on how they behave. Indeed, Deep Learning is still viewed with skepticism in those real-world domains in which incorrect predict…

Cited by 0 Source ↗PDF ↗Scholar ↗