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 simulated test dataset in which each event contains one $^{14}\mathrm{C}$ and one e$^+$ with kinetic energy below 5 MeV, the models achieve $^{14}\mathrm{C}$ recall rates of 25%-48% while maintaining $\mathrm{e}^{+}$ to $^{14}\mathrm{C}$ misidentification below 1%, leading to a large improvement in the resolution of total charge for events where $\mathrm{e}^{+}$ and $^{14}\mathrm{C}$ photon hits strongly overlap in space and time.

翻译：暂无翻译