Suppression of $^{14}\mathrm{C}$ photon hits in large liquid scintillator detectors via spatiotemporal deep learning
This work addresses a practical bottleneck in neutrino experiments by improving energy resolution through hit-level suppression of 14C background, but the improvement is incremental and domain-specific.
The authors propose deep learning models (gated spatiotemporal graph neural network and Transformer-based models) to tag and suppress 14C photon hits in liquid scintillator detectors, achieving 25%-48% 14C recall with <1% misidentification, improving energy resolution for overlapping events.
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 $β$ 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.