A Hybrid Framework for Kilonova Anomaly Detection using Single-Epoch SEDs from the 7-Dimensional Telescope Astropy ( 5 ; 3 ; 4 ) , NumPy ( 36 ) , SciPy ( 80 ) , Source Extractor ( 9 ) , SWarp ( 11 ) , SCAMP ( 10 ) , XGBoost ( 18 ) , LightGBM ( 47 ) , CatBoost ( 79 ) , Isolation Forest ( 56 ) .
Gregory S. H. Paek, Myungshin Im, Seo-Won Chang, Hyeonho Choi, Ji Hoon Kim · 2026
Read A Hybrid Framework for Kilonova Anomaly Detection using Single-Epoch SEDs from the 7-Dimensional Telescope Astropy ( 5 ; 3 ; 4 ) , NumPy ( 36 ) , SciPy ( 80 ) , Source Extractor ( 9 ) , SWarp ( 11 ) , SCAMP ( 10 ) , XGBoost ( 18 ) , LightGBM ( 47 ) , CatBoost ( 79 ) , Isolation Forest ( 56 ) . by Gregory S. H. Paek, Myungshin Im, Seo-Won Chang, Hyeonho Choi, Ji Hoon Kim (2026) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2603-13814.txt.
Introduction
Paek and co-authors build a hybrid system to spot kilonovae in data from the 7-Dimensional Telescope. It pairs an isolation forest, which flags odd events, with a gradient-boosted classifier that sorts eight common transient types. Training uses simulated telescope photometry that matches each filter's sensitivity. The classifier reaches a macro F1 score near 0.80, and the anomaly detector recovers over 90% of detectable kilonovae. The authors also show that half the filters keep nearly full accuracy.
Catalog record
| CREATOR | Gregory S. H. Paek, Myungshin Im, Seo-Won Chang, Hyeonho Choi, Ji Hoon Kim |
|---|---|
| YEAR | 2026 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 120538 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2603.13814