Suppression of $^{14}\mathrm{C}$ photon hits in large liquid scintillator detectors via spatiotemporal deep learning

TL;DR

This paper introduces deep learning models to identify and suppress $^{14}$C photon hits in liquid scintillator detectors, significantly improving energy resolution in neutrino experiments.

Contribution

It presents three novel deep learning models, including graph neural networks and Transformers, for tagging $^{14}$C photon hits at the hit level in liquid scintillator detectors.

Findings

Models achieve 25%-48% recall for $^{14}$C hits.

Misidentification of $e^+$ as $^{14}$C is kept below 1%.

Significant improvement in total charge resolution for overlapping hits.

Abstract

Liquid scintillator detectors are widely used in neutrino experiments due to their low energy threshold and high energy resolution. Despite the tiny abundance of $^{14}$C in LS, the photons induced by the $\beta$ decay of the $^{14}$C isotope inevitably contaminate the signal, degrading the energy resolution. In this work, we propose three models to tag $^{14}$C photon hits in $e^+$ events with $^{14}$C pile-up, thereby suppressing its impact on the energy resolution at the hit level: a gated spatiotemporal graph neural network and two Transformer-based models with scalar and vector charge encoding. For a simulation dataset in which each event contains one $^{14}$C and one $e^+$ with kinetic energy below 5 MeV, the models achieve $^{14}$C recall rates of 25%-48% while maintaining $e^+$ to $^{14}$C misidentification below 1%, leading to a large improvement in the resolution of total charge for events where $e^+$ and $^{14}$C photon hits strongly overlap in space and time.