ExoNet Archives

ExoNet architecture. Global and local CNN encoders each produce a 256-d embedding; the stellar MLP produces 128-d. All three are concatenated (640-d) and passed through a residual fusion MLP before the binary classifier head. -- astro-ph.EP

Astronomy & Telescopes, Exoplanets, -moons, -comets, Status Report, Stellar Cartography

ExoNet: Multimodal Deep Learning for TESS Exoplanet Candidate Identification via Phase-Folded Light Curves, Stellar Parameters, and Multi-Head Attention Fusion

by Keith CowingApril 22, 2026April 22, 2026

NASA’s Transiting Exoplanet Survey Satellite (TESS) has identified thousands of exoplanet candidates, yet many remain unconfirmed due to the limitations of manual vetting processes.

ExoNet

ExoNet: Multimodal Deep Learning for TESS Exoplanet Candidate Identification via Phase-Folded Light Curves, Stellar Parameters, and Multi-Head Attention Fusion

The Long-term Persistence Of Microbial Biospheres Introduces A Temporal Bias 1 Into The Drake Equation: Implications For The Search For Extraterrestrial Life

A Dual-mineral Protection Mechanism For Organic Matter Preservation And Its Implications For Mars Habitability And Biosignature Detection

POSTELLAR: Posterior Stellar Spectrum Sampling – An Alternative to Approximate Stellar Spectra for Exoplanetary Analysis

Interplanetary Dust Ablation Seeds Silicate Clouds in Exoplanet Atmospheres

Sea Ice As An Origin Of Life Location For Hycean And Ocean Worlds