Astronomers at the University of Warwick have used a new AI system called RAVEN to confirm more than 100 exoplanets from NASA's Transiting Exoplanet Survey Satellite (TESS) data. The discoveries include 31 newly identified worlds, many orbiting close to their stars, along with thousands of candidates. The findings reveal rare planet types and precise measurements of planetary occurrence rates around Sun-like stars.
Researchers applied the RAVEN pipeline to observations of over 2.2 million stars from TESS's first four years, focusing on planets with orbits shorter than 16 days. The system validated 118 new planets and over 2,000 high-quality candidates, nearly 1,000 of which were previously unknown. Among the confirmed worlds are ultra-short-period planets orbiting in under 24 hours and those in the 'Neptunian desert,' a region where such planets are scarce according to theory. The studies also identified tightly packed multi-planet systems with previously undetected pairs around the same star. The results appear in the Monthly Notices of the Royal Astronomical Society (MNRAS), volume 548 issue 3 and volume 546 issue 2, both published in 2026. RAVEN improves detection by training machine learning models on simulated data to distinguish true planetary transits from false signals like eclipsing binary stars. It handles the full process from signal detection to statistical validation, reducing biases and enabling reliable population studies. Dr. Marina Lafarga Magro, a postdoctoral researcher at Warwick and first author of the discovery paper, said: 'Using our newly developed RAVEN pipeline, we were able to validate 118 new planets, and over 2,000 high-quality planet candidates, nearly 1,000 of them entirely new.' Dr. Andreas Hadjigeorghiou, who led RAVEN's development, explained: 'The challenge lies in identifying if the dimming is indeed caused by a planet in orbit around the star or by something else, like eclipsing binary stars, which is what RAVEN tries to answer.' The analysis shows that 9-10% of Sun-like stars host close-in planets, aligning with NASA's Kepler mission but with uncertainties reduced by up to a factor of ten. Neptunian desert planets appear around just 0.08% of such stars. Dr. Kaiming Cui, first author of the demographics study, noted: 'For the first time, we can put a precise number on just how empty this 'desert' is.' Dr. David Armstrong, senior co-author, added that RAVEN produces datasets reliable enough to map planet prevalence. The team has released interactive catalogs for further research, aiding future missions like ESA's PLATO.