Research
My doctoral research applies machine learning to inference problems in hadronic physics — recovering physical distributions from raw scattering events, representing parton distributions with neural networks rather than hand-chosen functional forms, and doing active inference when the available priors are incomplete.
Before the Ph.D., my research was in combinatorial optimization: learning which search heuristic to apply, and finding promising regions of search spaces too large to explore exhaustively.
Conference Papers
Published in Conference on Physics and AI (PAI26), 2026
A conditional diffusion model that learns transverse momentum distributions directly from raw scattering events, without an intermediate binning or unfolding step.
Recommended citation: Jitao Xu, C. Cocuzza, K. Braga, D. Lersch, N. Sato, Y. Li. (2026). "Learning Transverse Momentum Distributions from Raw Scattering Events via Conditional Diffusion." Conference on Physics and AI (PAI26).
Published in Proceedings of the Thirty-Fifth International Joint Conference on Artificial Intelligence (IJCAI-ECAI 2026), Main Track, 2026
Diffusion-based active inference for inverse problems where the prior information available is incomplete. Accepted to the main track of IJCAI-ECAI 2026.
Recommended citation: Jitao Xu, N. Sato, Y. Li. (2026). "Active Diffusion-Based Inference for Ill-Posed Inverse Problems Under Incomplete Priors." Proceedings of the Thirty-Fifth International Joint Conference on Artificial Intelligence (IJCAI-ECAI 2026), Main Track.
Published in 2025 IEEE 37th International Conference on Tools with Artificial Intelligence (ICTAI), 2025
Generative surrogate models that make event-level analysis tractable in hadronic physics.
Recommended citation: T. Alghamdi, Jitao Xu, N. Ramachandra, N. Sato, Y. Li. (2025). "Towards an Event-Level Analysis in Hadronic Physics Using Generative AI-Based Surrogates." 2025 IEEE 37th International Conference on Tools with Artificial Intelligence (ICTAI).
Published in Proceedings of the AAAI Conference on Artificial Intelligence, 39(11), 2025
A learning-based framework that predicts which variable ordering heuristic to apply at each decision point in a CSP solver. AAAI 2025.
Recommended citation: Jitao Xu, Y. Wu, H. Li, M. Yin. (2025). "Prediction-Based Adaptive Variable Ordering Heuristics for Constraint Satisfaction Problems." Proceedings of the AAAI Conference on Artificial Intelligence, 39(11), 11390.
Journal Articles and Preprints
Published in arXiv preprint arXiv:2605.06994, 2026
A neural network parameterisation of generalized parton distributions. arXiv preprint.
Recommended citation: Jitao Xu, H. Jang, Z. Panjsheeri, G. W. Chern, Y. Li, S. Liuti, D. Adams, et al. (2026). "Neural Network Representation of Generalized Parton Distributions (NNGPD)." arXiv preprint arXiv:2605.06994.
Download Paper
Published in Applied Soft Computing, 164, 111934, 2024
A hybrid large-neighbourhood-search and evolutionary method that established new lower bounds on ten hard benchmark instances.
Recommended citation: Jitao Xu, H. Li, M. Yin. (2024). "Finding and Exploring Promising Search Space for the 0-1 Multidimensional Knapsack Problem." Applied Soft Computing, 164, 111934.
Download Paper
