Contrastive pretraining on sparse voxel representations of simulated neutrino events yields classifiers substantially more stable than an augmented baseline under simulated detector parameter shifts.
Alonso-Monsalve, D
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
hep-ex 1years
2025 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
Contrastive Learning for Robust Representations of Neutrino Data
Contrastive pretraining on sparse voxel representations of simulated neutrino events yields classifiers substantially more stable than an augmented baseline under simulated detector parameter shifts.