A JEPA-based model with domain-informed multi-view self-distillation learns light-curve representations that outperform hand-crafted features on 15 of 16 StarEmbed metrics and adapts competitively to other irregular time-series datasets.
Title resolution pending
2 Pith papers cite this work. Polarity classification is still indexing.
2
Pith papers citing it
fields
astro-ph.IM 2years
2026 2verdicts
UNVERDICTED 2representative citing papers
AstroSkyFlow generates simulated time-series astronomical images with realistic noise and variability, outperforming SkyMaker in noise and PSF reproduction while recovering injected signals such as exoplanet transits and asteroid trails.
citing papers explorer
-
Domain-Informed Multi-View Self-Distillation for Astronomical Light-Curve Representation Learning with JEPA
A JEPA-based model with domain-informed multi-view self-distillation learns light-curve representations that outperform hand-crafted features on 15 of 16 StarEmbed metrics and adapts competitively to other irregular time-series datasets.
-
AstroSkyFlow: an astronomical sky image flow simulator for time domain survey validation and machine learning
AstroSkyFlow generates simulated time-series astronomical images with realistic noise and variability, outperforming SkyMaker in noise and PSF reproduction while recovering injected signals such as exoplanet transits and asteroid trails.