Francesco Leofante

Francesco Leofante is a Research Fellow at Imperial College London, affiliated with the Centre for Explainable AI. His work focuses on Explainable AI (XAI), particularly counterfactual explanations with formal robustness guarantees against perturbations.

• Imperial College Research Fellowship
• DAAD AINet Fellowship (Safety and Security in AI)
• Imperial PFDC Supporting Research Staff and Students Award 2023

Research Interests center on Explainable AI, emphasizing counterfactual explanations, robustness, model multiplicity, and formal verification in critical systems like energy and aviation. His work bridges AI, formal methods, and human-AI collaboration.

Publications include studies on robust counterfactual explanations, parametric ReLUs for verification, and multi-agent systems with formal guarantees. These appear in top venues like AAAI, KR, IJCAI, and AAMAS, often addressing AI safety and trustworthy systems.

Scientific Awards include the Imperial PFDC Supporting Research Staff and Students Award 2023, DAAD AINet Fellowship, and Imperial College Research Fellowship. He also contributes to workshops and program committees at conferences like AAAI, KR, and IJCAI.

Future Work includes expanding robust XAI into critical infrastructure systems (energy, aviation) and developing tools like OMTPlan for AI planning and verification.