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. This paper presents ExoNet, a multimodal deep learning framework that integrates phase-folded global and local light curve representations with stellar parameters using a late-fusion architecture combining 1D Convolutional Neural Networks and Multi-Head Attention. Trained on labeled Kepler data, ExoNet achieves strong classification performance and demonstrates effective generalization to TESS data. Applied to 200 unconfirmed TESS planet candidates, the model identifies multiple high-confidence candidates, including several within the habitable zone. The results highlight the effectiveness of multimodal fusion and attention mechanisms in automated exoplanet candidate validation.
Superluminous supernovae (SLSNe) are one of the most luminous stellar explosions known, yet they remain poorly understood. Because they are intrinsically rare, efficiently identifying them in the large alert streams produced by modern time-domain surveys is essential for enabling spectroscopic follow-up. We present NOMAI, a machine learning classifier designed to identify SLSN candidates directly from photometric alerts in the ZTF stream, using light curves accumulated over at least 30 days. It does not require any spectroscopic redshift and is running in real time within the Fink broker. ZTF light curves are transformed into a set of physically motivated features derived primarily from model-fitting procedures using SALT2 and Rainbow, a blackbody-based multi-band fitting framework. These features are used to train an XGBoost classifier on a curated dataset of labeled ZTF sources constructed using literature samples of SLSNe, along with TNS and internal ZTF labeled sources. The final training dataset contains 5280 unique sources, including 225 spectroscopically classified SLSNe. On the training sample, the classifier reaches 66% completeness and 58% purity. Deployed within the Fink broker, NOMAI has been running continuously since 18/12/2025 on the ZTF alert stream and publicly reports SLSN candidates every night by automatically posting them to dedicated communication channels. Based on this, we also report the first two-month as an evaluation period, where the classifier successfully recovered 22 of the 24 active SLSNe reported on the Transient Name Server. The achieved performances demonstrate that the classifier provides a valuable tool for experts to efficiently scan the alert stream and identify promising candidates. In the near future, NOMAI is intended to be adapted to operate on the Legacy Survey of Space and Time conducted by the Vera C. Rubin Observatory.
The plasma environment around Mars is highly variable because it is strongly influenced by the solar wind. Accurate identification of plasma regions around Mars is important for the community studying solar wind-Mars interactions, region-specific plasma processes, and atmospheric escape. In this study, we develop a machine-learning-based classifier to automatically identify three key plasma regions--solar wind, magnetosheath, and induced magnetosphere--using only ion omnidirectional energy spectra measured by the MAVEN Solar Wind Ion Analyzer (SWIA). Two neural network architectures are evaluated: a multilayer perceptron (MLP) and a convolutional neural network (CNN) that incorporates short temporal sequences. Our results show that the CNN can reliably distinguish the three plasma regions, whereas the MLP struggles to separate the solar wind and magnetosheath. Therefore, the CNN-based approach provides an efficient and accurate framework for large-scale plasma region identification at Mars and can be readily applied to future planetary missions.
Gravitational-wave events are interpreted in terms of Bayesian posteriors for their source properties inferred under unphysical reference priors. Though these parameter estimates are important intermediate data products for downstream analyses, across the catalog they provide generically biased sourced properties and are therefore unsuitable for direct astrophysical interpretation. Hierarchical parameter estimation is the solution, where joint analysis of the entire catalog of observations not only reduces statistical uncertainties but actually informs the correct prior. Population-informed source properties from there derived are naturally suited to astrophysical interpretation and catalog statistics, such as identification of exceptional events from previous and ongoing observing runs. Using the latest LIGO-Virgo-KAGRA data, we thus demonstrate that population inference is not optional to interpret gravitational-wave observations.
Weak gravitational lensing, the correlated distortion of background galaxy shapes by foreground structures, is a powerful probe of the matter distribution in our universe and allows accurate constraints on the cosmological model. In recent years, high-order statistics and machine learning (ML) techniques have been applied to weak lensing data to extract the nonlinear information beyond traditional two-point analysis. However, these methods typically rely on cosmological simulations, which poses several challenges: simulations are computationally expensive, limiting most realistic setups to a low training data regime; inaccurate modeling of systematics in the simulations create distribution shifts that can bias cosmological parameter constraints; and varying simulation setups across studies make method comparison difficult. To address these difficulties, we present the first weak lensing benchmark dataset with several realistic systematics and launch the FAIR Universe Weak Lensing Machine Learning Uncertainty Challenge. The challenge focuses on measuring the fundamental properties of the universe from weak lensing data with limited training set and potential distribution shifts, while providing a standardized benchmark for rigorous comparison across methods. Organized in two phases, the challenge will bring together the physics and ML communities to advance the methodologies for handling systematic uncertainties, data efficiency, and distribution shifts in weak lensing analysis with ML, ultimately facilitating the deployment of ML approaches into upcoming weak lensing survey analysis.