ExoNet: Calibrated Multimodal Deep Learning for TESS Exoplanet Candidate Vetting using Phase-Folded Light Curves, Stellar Parameters, and Multi-Head Attention

TL;DR

ExoNet is a multimodal deep learning framework that enhances TESS exoplanet candidate vetting by integrating light curves and stellar data with advanced neural architectures, achieving high accuracy and identifying promising candidates.

Contribution

Introduces ExoNet, a novel calibrated multimodal deep learning model combining CNNs and multi-head attention for improved exoplanet candidate classification.

Findings

Achieves Test AUC of 0.9549 and 86.3% accuracy on Kepler data.

Identifies 1,754 high-confidence signals and 52 habitable-zone candidates.

Full ablation shows each modality improves model performance.

Abstract

The discovery of exoplanets at scale has become one of the defining data science challenges in modern astrophysics. NASA's Transiting Exoplanet Survey Satellite (TESS) had catalogued over 7,800 planet candidates by early 2026, yet confirmation stands at fewer than 720. This paper introduces ExoNet, a multimodal deep learning framework that jointly processes phase-folded global and local light curve views alongside stellar parameter features through a calibrated late-fusion architecture combining 1D Convolutional Neural Networks, 8-head Multi-Head Attention over temporal feature maps, and a residual fusion head with post-hoc Temperature Scaling calibration. Trained on 7,585 labeled Kepler Objects of Interest, ExoNet achieves Test AUC = 0.9549 and 86.3% accuracy. Applied to 4,720 verified unconfirmed TESS Planet Candidates with TOI-TIC cross-identification verified against the NASA Exoplanet Archive, the model yields 1,754 high-confidence signals, 52 habitable-zone candidates, and six Earth-sized habitable-zone targets below 1.6 Earth radii. TOI-5728.01 and TOI-6716.01 emerge as the most Earth-like unconfirmed candidates. Full ablation confirms each modality improves AUC. Code and catalog are openly released.