A survey on generative AI for detector effects unfolding in particle and nuclear physics
Authors:
- Tareq AlghamdiTATareq Alghamdi
- Tommaso VittoriniTVTommaso Vittorini
- Jitao XuJXJitao Xu
- Marco BattaglieriMBMarco Battaglieri
- Derek I. GlazierDGDerek I. Glazier
- Glòria MontañaGMGlòria Montaña
- Giorgio FotiGFGiorgio Foti
- Alessandro PilloniAPAlessandro Pilloni
- Nobuo SatoNSNobuo Sato
- Yaohang LiYLYaohang Li
- I-16146 Genova Infn Sezione di GenovaIGI-16146 Genova Infn Sezione di Genova
- Newport News Thomas Jefferson National Accelerator FacilityNFNewport News Thomas Jefferson National Accelerator Facility
Open Access
- 1 January 2026
- journal article
- Published by American Institute of Mathematical Sciences (AIMS) in Applied Computing and Intelligence
- Vol. 6 (2) , 138-156
Abstract
In particle and nuclear physics, “detector effects unfolding” can be viewed as a highdimensional inverse problem whose goal is to recover the true event distributions from observed experimental data corrupted by detector-induced distortions. Recent advances in generative AI have positioned data-driven and machine learning-based approaches as powerful alternatives to traditional unfolding techniques, offering superior scalability to high-dimensional data, capability of learning complex detector responses, and the ability to operate directly at the event level. We survey state-ofthe-art generative AI-based models for detector folding and unfolding. We review existing architectures and training strategies, and highlight recent methodological advances and open challenges. Through a detailed discussion of latent space representations, uncertainty quantification, physics insights, and background removals, we demonstrate the potential to significantly advance the field.Keywords
This publication has 42 references indexed in Scilit:
- Converting sWeights to probabilities with density ratiosComputer Physics Communications, 2025
- Point cloud-based diffusion models for the Electron-Ion ColliderPhysics Letters B, 2025
- Physics-driven learning for inverse problems in quantum chromodynamicsNature Reviews Physics, 2025
- CaloClouds: fast geometry-independent highly-granular calorimeter simulationJournal of Instrumentation, 2023
- Generative models uncertainty estimationJournal of Physics: Conference Series, 2023
- Ensemble Models for Calorimeter SimulationsJournal of Physics: Conference Series, 2023
- Decoding Photons: Physics in the Latent Space of a BIB-AE Generative NetworkEPJ Web of Conferences, 2021
- Mining gold from implicit models to improve likelihood-free inferenceProceedings of the National Academy of Sciences, 2020
- Learning Particle Physics by Example: Location-Aware Generative Adversarial Networks for Physics SynthesisComputing and Software for Big Science, 2017
- : A statistical tool to unfold data distributionsNuclear Instruments and Methods in Physics Research Section A: Accelerators, Spectrometers, Detectors and Associated Equipment, 2005