publications

Publications in reversed chronological order.

2026

  1. TPM 2026
    Rule-Selection Surrogates for Tractable Neurosymbolic Inference
    David Debot, and Giuseppe Marra
    TPM 2026, 2026
    PDF
  2. ICML
    Prototype-Grounded Concept Models for Verifiable Concept Alignment
    Stefano Colamonaco, David Debot, Pietro Barbiero, and Giuseppe Marra
    ICML 2026, 2026
    Code
  3. NeurIPS
    Neurosymbolic Object-Centric Learning with Distant Supervision
    Stefano Colamonaco, David Debot, and Giuseppe Marra
    NeurIPS, 2026

2025

  1. AAAI
    Neurosymbolic Reinforcement Learning: Playing MiniHack With Probabilistic Logic Shields
    David Debot, Gabriele Venturato, Giuseppe Marra, and Luc De Raedt
    In Proceedings of the AAAI Conference on Artificial Intelligence, 2025
    Video Code Website
  2. Preprint
    Interpretable Hierarchical Concept Reasoning through Attention-Guided Graph Learning
    David Debot, Pietro Barbiero, Gabriele Dominici, and Giuseppe Marra
    arXiv preprint arXiv:2506.21102 (under review), 2025
  3. Preprint
    Quantifying the Accuracy-Interpretability Trade-Off in Concept-Based Sidechannel Models
    David Debot, and Giuseppe Marra
    arXiv preprint arXiv:2510.05670 (under review), 2025

2024

  1. NeurIPS
    cmr_preview.png
    Interpretable Concept-Based Memory Reasoning
    David Debot, Pietro Barbiero, Francesco Giannini, Gabriele Ciravegna, Michelangelo Diligenti, and Giuseppe Marra
    In Thirty-eight Conference on Neural Information Processing Systems, 2024
    PDF Video Blog Code Poster
  2. eXCV at ECCV
    Integrating Local and Global Interpretability for Deep Concept-Based Reasoning Models
    David Debot, and Giuseppe Marra
    In European Conference on Computer Vision, 2024