Link model to paper and add Arxiv metadata

README.md

@@ -1,19 +1,22 @@
 ---
 language:
 - en
+library_name: transformers
 license: apache-2.0
+pipeline_tag: text-generation
 tags:
 - language-model
 - sample-efficient
 - pretraining
 - transformer
-library_name: transformers
-pipeline_tag: text-generation
+arxiv: 2602.02522
 ---
 
 # IMU-1 Base
 
-A sample-efficient 430M parameter language model trained on 72B tokens that approaches the benchmark performance of models trained on 56× more data.
+This repository contains the IMU-1 Base model, a sample-efficient 430M parameter language model introduced in the paper [IMU-1: Sample-Efficient Pre-training of Small Language Models](https://huggingface.co/papers/2602.02522).
+
+IMU-1 is trained on 72B tokens and approaches the benchmark performance of models trained on 56× more data.
 
 ## Model Details
 
@@ -95,14 +98,14 @@ print(tokenizer.decode(outputs[0]))
 ## Citation
 
 ```bibtex
-@misc{imu1_2025,
-  title={Sample Efficient Language Model Pre-training},
+@article{panchuk2025imu1,
+  title={IMU-1: Sample-Efficient Pre-training of Small Language Models},
   author={George Panchuk},
-  year={2025},
-  url={https://huggingface.co/thepowerfuldeez/imu1_base}
+  journal={arXiv preprint arXiv:2602.02522},
+  year={2025}
 }
 ```
 
 ## License
 
-Apache 2.0
+Apache 2.0