# StarEmbed: Benchmarking Time Series Foundation Models on Astronomical Observations of Variable Stars

arXiv: https://arxiv.org/abs/2510.06200 · HTML: https://mlsanity.com/paper/2510.06200/

## Key facts

- Cited by 1 papers in the map (data through 14 Oct 2026).
- Builds on 26 of its 69 references that are in the map.
- First posted on arXiv 7 Oct 2025.

## Builds on (most cited first, 26 of 26)

- [Emerging Properties in Self-Supervised Vision Transformers](https://mlsanity.com/paper/2104.14294/) (2021)
- [A Time Series is Worth 64 Words: Long-term Forecasting with Transformers](https://mlsanity.com/paper/2211.14730/) (2022)
- [iBOT: Image BERT Pre-Training with Online Tokenizer](https://mlsanity.com/paper/2111.07832/) (2021)
- [Quantifying Attention Flow in Transformers](https://mlsanity.com/paper/2005.00928/) (2020)
- [Chronos: Learning the Language of Time Series](https://mlsanity.com/paper/2403.07815/) (2024)
- [Unified Training of Universal Time Series Forecasting Transformers](https://mlsanity.com/paper/2402.02592/) (2024)
- [TS2Vec: Towards Universal Representation of Time Series](https://mlsanity.com/paper/2106.10466/) (2021)
- [Text and Code Embeddings by Contrastive Pre-Training](https://mlsanity.com/paper/2201.10005/) (2022)
- [Time-MoE: Billion-Scale Time Series Foundation Models with Mixture of Experts](https://mlsanity.com/paper/2409.16040/) (2024)
- [VisionTS: Visual Masked Autoencoders Are Free-Lunch Zero-Shot Time Series Forecasters](https://mlsanity.com/paper/2408.17253/) (2024)
- [Image Clustering with External Guidance](https://mlsanity.com/paper/2310.11989/) (2023)
- [Deep Transformation-Invariant Clustering](https://mlsanity.com/paper/2006.11132/) (2020)
- [LSEnet: Lorentz Structural Entropy Neural Network for Deep Graph Clustering](https://mlsanity.com/paper/2405.11801/) (2024)
- [Scaling-laws for Large Time-series Models](https://mlsanity.com/paper/2405.13867/) (2024)
- [RAPID: Early Classification of Explosive Transients using Deep Learning](https://mlsanity.com/paper/1904.00014/) (2019)
- [Alert Classification for the ALeRCE Broker System: The Real-time Stamp Classifier](https://mlsanity.com/paper/2008.03309/) (2020)
- [Scalable End-to-end Recurrent Neural Network for Variable star classification](https://mlsanity.com/paper/2002.00994/) (2020)
- [Maven: A Multimodal Foundation Model for Supernova Science](https://mlsanity.com/paper/2408.16829/) (2024)
- [AstroMLab 2: AstroLLaMA-2-70B Model and Benchmarking Specialised LLMs for Astronomy](https://mlsanity.com/paper/2409.19750/) (2024)
- [Deep Attention-Based Supernovae Classification of Multi-Band Light-Curves](https://mlsanity.com/paper/2201.08482/) (2022)
- [Representation Learning for Time-Domain High-Energy Astrophysics: Discovery of Extragalactic Fast X-ray Transient XRT 200515](https://mlsanity.com/paper/2412.01150/) (2024)
- [A Classifier-Based Approach to Multi-Class Anomaly Detection for Astronomical Transients](https://mlsanity.com/paper/2403.14742/) (2024)
- [Rotary Masked Autoencoders are Versatile Learners](https://mlsanity.com/paper/2505.20535/) (2025)
- [The Tiny Time-series Transformer: Low-latency High-throughput Classification of Astronomical Transients using Deep Model Compression](https://mlsanity.com/paper/2303.08951/) (2023)
- [The Scaling Law in Stellar Light Curves](https://mlsanity.com/paper/2405.17156/) (2024)
- [ORACLE: A Real-Time, Hierarchical, Deep-Learning Photometric Classifier for the LSST](https://mlsanity.com/paper/2501.01496/) (2025)

## Built on by (most cited first, 1 of 1)

- [Domain-Informed Multi-View Self-Distillation for Astronomical Light-Curve Representation Learning with JEPA](https://mlsanity.com/paper/2606.28446/) (2026)

## Abstract (arXiv)

Current time series foundation model (TSFM) training corpora largely omit data with certain complexities like irregular temporal sampling. Astronomical time series of stellar fluxes (light curves) are available in immense quantities and exhibit irregular sampling, multiple variates, and heteroskedasticity. We introduce StarEmbed, the first public benchmark for light curves comprised of real observations of ~40,000 stars expert-labeled across seven classes and evaluations in clustering, classification, and out-of-distribution (OOD) source detection. We benchmark TSFMs with differing architecture and training strategies as well as domain-specific transformers. Our results demonstrate that the Chronos family, despite being pre-trained on regularly sampled non-astronomical data, yields state-of-the-art (SOTA) performance in light curve clustering and OOD detection. While no TSFM strictly surpasses the classification performance of the long-established domain baseline, they do demonstrate excellent generalization abilities. StarEmbed marks a step toward universal light curve embeddings and improved TSFM performance on challenging data.