EncoTESS is a compact Time Series Foundation Model trained on TESS 2‑minute light curves that encodes stellar variability into a fixed‑size latent space, handling noise, irregular sampling, and data gaps. It improves age estimation for young K and M stars (≤100 Myr) and older M stars (≤1 Gyr) by outperforming traditional rotation period and variability amplitude indicators. The model’s lightweight architecture (~1 % of typical TSFMs) allows deployment on standard laptops and can be extended to other TESS cadences and missions like Kepler and PLATO.
arXiv:2510. 06200v4 Announce Type: replace-cross Abstract: Current time series foundation model (TSFM) training corpora largely omit data with certain complexities like irregular temporal sampling.
DELOS is a deep‑learning framework that uses contrastive scoring to perform blind searches for shallow transits in Kepler photometry. It combines GPU‑accelerated phase folding, optimized phase binning, and a custom one‑dimensional convolutional encoder to produce a transit‑likeness score periodogram without relying on pre‑detected events. In tests on synthetic data and controlled injection‑recovery experiments, DELOS outperforms traditional methods (BLS and TLS) in precision‑recall and speed, and successfully recovers all known shallow intermediate‑to‑long‑period transit signals in a selected Kepler sample.
arXiv:2606. 28446v1 Announce Type: cross Abstract: Light curves describe temporal variations in the brightness of celestial objects.
arXiv:2606. 07771v1 Announce Type: cross Abstract: Foundation models for astronomical surveys offer powerful learned representations that can be transferred to downstream regression tasks such as galaxy property estimation.
arXiv:2607. 00228v1 Announce Type: cross Abstract: Modern time-domain surveys such as the Zwicky Transient Facility (ZTF) generate hundreds of thousands of alerts each night, making real-time decisions for follow-up observations a central challenge in time-domain astronomy.
arXiv:2606. 00219v1 Announce Type: cross Abstract: We are witnessing a surge in observations of the cosmic dawn (CD) and epoch of reionisation (EoR), driving an increasing demand for fast and robust theoretical interpretation frameworks.
The paper demonstrates that a self‑supervised Vision Transformer (ViT) pretrained on a fast, low‑cost semi‑numerical simulator can produce data summaries that transfer across different simulators without retraining. In 21cm cosmology, the ViT—named SKATR—pretrained on 67,000 21cmFAST lightcones is applied unchanged to hydrodynamical Loreli II lightcones, enabling accurate inference of five astrophysical parameters with fewer radiative‑transfer simulations than a fully‑supervised baseline. SKATR remains accurate, informative, and calibrated even under realistic SKA antenna array noise, outperforming supervised models retrained on noisy data.
arXiv:2606. 02778v1 Announce Type: cross Abstract: I present EXOVEIL, a transit detection system that learns what a star's brightness should look like and flags when reality disagrees.
arXiv:2605. 27527v2 Announce Type: replace-cross Abstract: Astrophysical observations from Earth are subject to weather, environmental, and scientific constraints that lead to sparse, irregular light curves.
arXiv:2608. 19419v1 Announce Type: cross Abstract: Microlensing can reveal populations of faint compact objects that are otherwise difficult to detect.
arXiv:2606. 18464v1 Announce Type: cross Abstract: Detecting the tiny Doppler shifts induced by Earth-mass planets in stellar radial-velocity measurements remains extremely challenging due to stellar activity.
The study audits the AION-1 foundation model, a 39‑modality transformer trained on over 200 million astronomical objects, and finds that its reliance on a survey detection channel—specifically the segmentation map—introduces a severe systematic bias. By keeping image tokens unchanged and editing only the segmentation map, all model outputs (flux, size, ellipticity, redshift) shift by factors of 110–4400 compared to a placebo, revealing that the model’s predictions are driven more by detection gating than by the actual light distribution. This bias propagates into cosmological analyses, shifting tomographic mean redshifts by a median 0.71 × the LSST DESC requirement and exceeding it in multiple assignments, while removing the detection channel eliminates the effect without measurable cost. whyItMatters":"The bias in the detection channel directly inflates errors in key astronomical measurements, potentially compromising the precision of cosmological studies that rely on accurate redshift estimates."