Update README.md

README.md

@@ -209,18 +209,10 @@ dataset_info:
 
 datasets:
   - legacy
-
-citation: |
-  @inproceedings{koblischke2025why,
-    title={Why wait for human annotations when you have AI? Semantic searching scientific images with synthetic labels},
-    author={Nolan Koblischke and Liam Parker and Francois Lanusse and Irina Espejo Morales and Jo Bovy and Shirley Ho},
-    booktitle={NeurIPS 2025 AI for Science Workshop},
-    year={2025},
-    url={https://openreview.net/forum?id=j8Qxvb37HQ}
-  }
 ---
 
 # Description Benchmark based on GalaxyZoo
+[![arXiv](https://img.shields.io/badge/arXiv-2512.11982-b31b1b.svg)](https://arxiv.org/abs/2512.11982)
 
 A small sample of the GZ-DECaLS dataset to cheaply test the ability of vision-language models to describe galaxy images.
 
@@ -230,12 +222,14 @@ The underlying catalog data are derived from the Galaxy Zoo DECaLS: Detailed Vis
 
 # Embeddings Citation
 ```
-@inproceedings{koblischke2025why,
-title={Why wait for human annotations when you have AI? Semantic searching scientific images with synthetic labels},
-author={Nolan Koblischke and Liam Parker and Francois Lanusse and Irina Espejo Morales and Jo Bovy and Shirley Ho},
-booktitle={NeurIPS 2025 AI for Science Workshop},
-year={2025},
-url={https://openreview.net/forum?id=j8Qxvb37HQ}
+@misc{koblischke2025semantic,
+      title={Semantic search for 100M+ galaxy images using AI-generated captions}, 
+      author={Nolan Koblischke and Liam Parker and Francois Lanusse and Irina Espejo Morales and Jo Bovy and Shirley Ho},
+      year={2025},
+      eprint={2512.11982},
+      archivePrefix={arXiv},
+      primaryClass={astro-ph.IM},
+      url={https://arxiv.org/abs/2512.11982}, 
 }
 ```