Curriculum Vitae

Education

Academic Experience

Research Interests

• Deep generative models: flow matching, diffusion, and energy-based models
• Compositional and controllable generation: training-free composition, guidance, multi-condition sampling
• Graph machine learning: message passing, long-range information propagation, benchmarking
• Generative models for molecules: property optimisation and molecular design
• AI for health: biosignals and wearable/smartphone sensor data

Publications

Abstracts and links are on the publications page.

1. [MGB26b]
   L. Miglior*, A. Gravina*, D. Bacciu. Stable transformers for graph generation. arXiv:2609.39739, 2026.
2. [MGB26a]
   L. Miglior, V. Gervasi, D. Bacciu. CompFlow: Composing velocity fields for multi-condition generation. CompLearn Workshop at ICML, 2026.
3. [MTGB26]
   L. Miglior, M. Tolloso, A. Gravina, D. Bacciu. Can you hear me now? A benchmark for long-range graph propagation. ICLR, 2026.
4. [SMG⁺25]
   L. Simone, L. Miglior, V. Gervasi, et al. Deep learning framework for cardiorespiratory disease detection using smartphone IMU sensors. Computers in Biology and Medicine, 2025.
5. [MSPB25]
   L. Miglior, L. Simone, M. Podda, D. Bacciu. Towards efficient molecular property optimization with graph energy based models. ESANN, 2025.
6. [VGM⁺24]
   E. Vignali, E. Gasparotti, L. Miglior, et al. A new smartphone-based method for remote health monitoring: Assessment of respiratory kinematics. Electronics, 2024.
7. [VMG⁺23]
   E. Vignali, L. Miglior, E. Gasparotti, et al. An application for remote respiratory patterns detection in post-surgery patients. 28th Congress of the European Society of Biomechanics, 2023.

Conferences & Workshops