New system bridges spectral differences between major sky surveys, accelerating Milky Way research and exoplanet discovery
Researchers in China have developed an advanced artificial intelligence model capable of integrating stellar spectral data captured by different telescopes, overcoming one of astronomy’s most persistent technical barriers.
The model enables scientists to process and compare vast datasets gathered using varying observational methods, resolutions and wavelength ranges. The breakthrough demonstrates the expanding role of AI in managing large-scale astronomical information, as reported by Xinhua News Agency, a partner of TV BRICS.
Stellar spectra provide critical insights into a star’s temperature, chemical composition and surface gravity. By analysing this information, astronomers can reconstruct the evolutionary history of the Milky Way, from its earliest formation stages to the present day.
To address this challenge, a research team from the National Astronomical Observatories of the Chinese Academy of Sciences and the University of the Chinese Academy of Sciences applied principles similar to those used in large language models. Using a contrastive learning framework, the AI system autonomously learns the intrinsic relationships between spectral datasets from different instruments.
The model’s capabilities are expected to play a pivotal role in galactic archaeology – the study of ancient stars to understand the Milky Way’s formation and merger history. By rapidly sifting through enormous datasets, AI models can help detect extremely rare, metal-poor stars that serve as relics of the early universe.
The AI system has already been deployed in exploratory missions, including programmes focused on identifying planets similar to Earth. By accurately characterising the properties of planet-hosting stars, the model enhances the efficiency of screening potentially habitable worlds.
Researchers say the innovation marks a significant step towards fully integrated astronomical data ecosystems, where AI bridges technical divides and unlocks deeper insights into the structure and origins of our galaxy.