aiqbench analysis
The paper uses a generative neural surrogate for neutrino-nucleus scattering events and fine-tunes it to published MicroBooNE cross-section measurements, without needing event-level experimental data. The main ML question is whether calibrating a generative model to aggregate distributions improves unseen observables and transfers across beam fluxes. Those are useful questions for ML readers, though the claims rest on one experiment and one interaction model.
Machine-learning-based neutrino-event generation constrained by experimental measurements
arXiv:2610.20870v1 Announce Type: cross Abstract: We present a generative surrogate for charged-current \(\nu_\mu\)-nucleus scattering that can be calibrated to measured differential cross sections without event-level experimental samples. Starting from a theoretical interaction model, we fine-tune the surrogate to a MicroBooNE measurement and show that this also improves the description of observables not used in the fit. The calibrated model can also be applied to measurements from different neutrino fluxes without retuning, while retaining event-level predictions of outgoing-muon kinematics. Our results show that experimental cross sections can directly constrain generative event models while preserving their applicability across different observables and beam configurations.
Abstract:We present a generative surrogate for charged-current \(\nu_\mu\)-nucleus scattering that can be calibrated to measured differential cross sections without event-level experimental samples. Starting from a theoretical interaction model, we fine-tune the surrogate to a MicroBooNE measurement and show that this also improves the description of observables not used in the fit. The calibrated model can also be applied to measurements from different neutrino fluxes without retuning, while retaining event-level predictions of outgoing-muon kinematics. Our results show that experimental cross sections can directly constrain generative event models while preserving their applicability across different observables and beam configurations.
High Energy Physics - Phenomenology (hep-ph); Machine Learning (cs.LG); High Energy Physics - Experiment (hep-ex); Nuclear Experiment (nucl-ex); Nuclear Theory (nucl-th)
arXiv:2610.20870 [hep-ph]
(or arXiv:2610.20870v1 [hep-ph] for this version)
https://doi.org/10.48550/arXiv.2610.20870 arXiv-issued DOI via DataCite (pending registration)
From: Krzysztof M. Graczyk [view email] [v1] Wed, 14 Oct 2026 19:59:28 UTC (2,994 KB)
Full text under the publisher's open licence · 147 words
Related coverage
Researchers classify neutrino events with a quantum computer
Quantum Zeitgeist · Aug 23, 2026
Markov Chain Decoders Overcome the Heavy-Tail Limitations of Lipschitz Generative Models
arXiv · Jun 5, 2026
From Correlation to Causation in Lane Change Prediction for Automated Driving: A Causal Explanation Framework
arXiv · Jun 16, 2026
Geospatial Diffusion-based Evolution Synthesis (GeoDES) for Storm-Centered Weather Augmentation
arXiv · Jul 23, 2026