Artificial Intelligence Discovered 247 New Cosmic Lenses express

Original: "Searching for Galaxy Cluster-Scale Strong lenses from the DESI Legacy Imaging Surveys"
arXiv:2601.11054 · 2026-01-16 · CC BY 4.0 · 1 min
A neural network identified 247 previously unknown gravitational lenses — nature's telescopes that bend light from distant galaxies.

📄 Showing the "Simple" version — "Advanced" is not ready yet. Add it to favorites to help prioritize it.

A galaxy cluster's gravity warps spacetime, acting like a colossal lens: it focuses and amplifies light from distant objects, as predicted by Einstein. Such lenses are rare, but they let us glimpse what lurks at the edges of the cosmos.

To find them, a neural network was trained on tens of thousands of synthetic and real images, then set to sift through 540,432 snapshots. The AI hunted for barely visible arcs of light. Of 485 suspicious objects, 247 turned out to be new—16 with bright, crisp arcs.

The most unexpected find: one lens gave itself away with a microscopic arc that was previously dismissed as an image defect. Now these lenses help astrophysicists: by analyzing the light bending, they calculate how much invisible dark matter hides in the clusters, and how dark energy accelerates the Universe's expansion. The data is open to all observatories worldwide.

Original abstract

as written by the authors arXiv:2601.11054

Galaxy cluster-scale strong gravitational lensing systems are rare yet valuable tools for investigating the properties of dark matter and dark energy, as well as providing the opportunity to study the distant universe at flux levels and spatial resolutions that would otherwise be unavailable. Large-scale imaging surveys present unprecedented opportunities to expand the sample of cluster lenses. In this study, we adopt a deep learning-based approach to identify cluster lenses from the DESI Legacy Imaging Surveys, utilizing the catalog of galaxy cluster candidates identified by Zou et al. (2021). Our lens-finder employs a ResNet-18 architecture, trained with mock images of cluster lenses as positives and observational images of cluster scale non-lenses as negatives. We do an iterative operation to increase the completeness of our work, namely adding the found true positive samples back to the training set and training again for several times. Human inspection is conducted to further refine the candidates, categorizing them into grades (A, B, C) according to the significance of the strongly lensed arcs. Reviewing all 540,432 objects in Zou's catalog, we discover 485 high-confidence cluster lens candidates with a cluster $$M_{500}$$ range of $$10^{13.67\sim14.97}M_{\odot}$$ and a Brightest Central Galaxy (BCG) redshift range of $$0.04\sim0.89$$. After excluding the lens candidates listed in previous studies, we identify 247 newly discovered cluster lens candidates, including 16 grade A, 90 grade B, and 141 grade C. This catalog of cluster lens candidates is publicly available online, and follow-up observations are encouraged to confirm and conduct thorough investigations of these systems.

Gravitational lensing sometimes produces 'Einstein rings'—perfect circles of light when the lens, source, and observer align just right. In these rings, you can even make out individual star clusters in distant galaxies.

Scientists
Fritz ZwickyJacob BekensteinKarl SchwarzschildJoseph WeberCharles TownesEmmy Noether
Concepts
galaxy cluster globular cluster desi lsst large magellanic cloud large-scale structure catalog strong lensing gravitational lensing modeling image classification gravitational microlensing microlensing clustering galactic center galactic bulge constellation apparent magnitude satellite galaxy dense clump red dot galaxies