First event-by-event Cherenkov separation from sub-MeV electrons in liquid argon enables a proof-of-principle ALP search excluding new parameter space despite no observed excess.
GraphNeT: Graph neural networks for neutrino telescope event reconstruction
2 Pith papers cite this work, alongside 14 external citations. Polarity classification is still indexing.
2
Pith papers citing it
14
external citations · Pith
abstract
GraphNeT is an open-source python framework aimed at providing high quality, user friendly, end-to-end functionality to perform reconstruction tasks at neutrino telescopes using graph neural networks (GNNs). GraphNeT makes it fast and easy to train complex models that can provide event reconstruction with state-of-the-art performance, for arbitrary detector configurations, with inference times that are orders of magnitude faster than traditional reconstruction techniques. GNNs from GraphNeT are flexible enough to be applied to data from all neutrino telescopes, including future projects such as IceCube extensions or P-ONE. This means that GNN-based reconstruction can be used to provide state-of-the-art performance on most reconstruction tasks in neutrino telescopes, at real-time event rates, across experiments and physics analyses, with vast potential impact for neutrino and astro-particle physics.
fields
hep-ex 2years
2026 2representative citing papers
HEPTv2 achieves 98.6% double-majority tracking efficiency at 0.8% fake rate with ~15 ms inference and 0.4 GB memory on TrackML using an end-to-end point transformer with locality-sensitive hashing.
citing papers explorer
-
First Demonstration of a Hybrid Cherenkov and Scintillation Detector in a Proof-of-Principle Axion Search at a Beam Dump
First event-by-event Cherenkov separation from sub-MeV electrons in liquid argon enables a proof-of-principle ALP search excluding new parameter space despite no observed excess.
-
HEPTv2: End-to-End Efficient Point Transformer for Charged Particle Reconstruction
HEPTv2 achieves 98.6% double-majority tracking efficiency at 0.8% fake rate with ~15 ms inference and 0.4 GB memory on TrackML using an end-to-end point transformer with locality-sensitive hashing.