Speakers
Description
Multi-messenger astronomy relies on three independent alert systems - gravitational wave detectors (LIGO/Virgo/KAGRA), high-energy neutrino observatories (IceCube), and gamma-ray monitors (Fermi-GBM), each firing alerts only when their own threshold is individually crossed. The problem is that when all three show near-threshold activity around the same time, no existing pipeline recognizes it as potentially significant. Temporally coincident sub-threshold signals across all three messengers could indicate a genuine astrophysical event like a faint kilonova or a distant neutron star merger which arent loud enough for any single system to catch on its own. We propose an ML-based fusion model that takes in near-threshold data from all three streams simultaneously and outputs a joint significance score at the trigger stage, before full offline parameter estimation. Using archival public data from GWTC skymaps, IceCube public alerts, and the Fermi-GBM catalog, we simulate a low-latency coincidence scenario and train the model to distinguish true astrophysical coincidences from accidental ones. We compare against standard single-channel thresholding and show that joint fusion recovers sub-threshold events that current pipelines miss entirely. With next-generation facilities like Einstein Telescope and IceCube-Gen2 pushing data rates higher, real-time cross-messenger trigger fusion is only going to become more critical.
| Do you plan to submit a 4-page extended abstract on OpenReview (only for Presentations/Posters)? | Maybe |
|---|