Monitoring and explainability

A model can be robust and still be used on inputs it was never meant to handle. We develop methods that flag such inputs at runtime: out-of-distribution detectors that stay reliable under adversarial attack, using the instability of detection scores under smoothing, and detectors based on the distance to the decision boundary, computed with counterfactual explanations that also explain each decision.

We also study explanations for systems in which humans and neural agents interact, and verify formally that these explanations are robust.

Separation of in-distribution and out-of-distribution scores with ROSS (CVPR Findings 2026)
Separation of in-distribution and out-of-distribution scores with ROSS (CVPR Findings 2026)

Active researchers

2026

CVPR-F 2026

A Robust Out-of-Distribution Detection Framework via Synergistic Smoothing

M. Stoica, A. Hekal, A. Lomuscio

PDFarXiv
BibTeX
@article{DBLP:journals/corr/abs-2605-08191,
  author     = {Maria Stoica and Abdelrahman Hekal and Alessio Lomuscio},
  title      = {{A Robust Out-of-Distribution Detection Framework via Synergistic Smoothing}},
  journal    = {CoRR},
  volume     = {abs/2605.08191},
  year       = {2026},
  eprinttype = {arXiv},
  eprint     = {2605.08191},
  doi        = {10.48550/ARXIV.2605.08191}
}
IJCAI WS 2026

Out-of-Distribution Detection using Counterfactual Distance

M. Stoica, F. Leofante, A. Lomuscio

arXiv
BibTeX
@article{DBLP:journals/corr/abs-2508-10148,
  author     = {Maria Stoica and Francesco Leofante and Alessio Lomuscio},
  title      = {{Out-of-Distribution Detection using Counterfactual Distance}},
  journal    = {CoRR},
  volume     = {abs/2508.10148},
  year       = {2025},
  eprinttype = {arXiv},
  eprint     = {2508.10148},
  doi        = {10.48550/ARXIV.2508.10148}
}

2023

AAMAS 2023

Towards Robust Contrastive Explanations for Human-Neural Multi-agent Systems

F. Leofante, A. Lomuscio

PDFPublisher
BibTeX
@inproceedings{DBLP:conf/atal/LeofanteL23,
  author     = {Francesco Leofante and Alessio Lomuscio},
  title      = {{Towards Robust Contrastive Explanations for Human-Neural Multi-agent Systems}},
  booktitle  = {Proceedings of the 2023 International Conference on Autonomous Agents and Multiagent Systems, AAMAS 2023, London, United Kingdom, 29 May 2023 - 2 June 2023},
  pages      = {2343--2345},
  publisher  = {ACM},
  year       = {2023},
  url        = {https://dl.acm.org/doi/10.5555/3545946.3598928}
}
EUMAS 2023

Robust Explanations for Human-Neural Multi-agent Systems with Formal Verification

F. Leofante, A. Lomuscio

DOI
BibTeX
@inproceedings{DBLP:conf/eumas/LeofanteL23,
  author     = {Francesco Leofante and Alessio Lomuscio},
  title      = {{Robust Explanations for Human-Neural Multi-agent Systems with Formal Verification}},
  booktitle  = {Multi-Agent Systems - 20th European Conference, EUMAS 2023, Naples, Italy, September 14-15, 2023, Proceedings},
  series     = {Lecture Notes in Computer Science},
  volume     = {14282},
  pages      = {244--262},
  publisher  = {Springer},
  year       = {2023},
  doi        = {10.1007/978-3-031-43264-4_16}
}

Other areas