Add citation info

README.md

@@ -15,7 +15,7 @@ datasets:
 Pretrained model on Uzbek language (Cyrillic script) using a masked
 language modeling and next sentence prediction objectives.
 
-### How to use
+## How to use
 
 You can use this model directly with a pipeline for masked language modeling:
 
@@ -61,3 +61,15 @@ You can use this model directly with a pipeline for masked language modeling:
 ## Training data
 
 UzBERT model was pretrained on \~625K news articles (\~142M words).
+
+## BibTeX entry and citation info
+```bibtex
+@misc{mansurov2021uzbert,
+      title={{UzBERT: pretraining a BERT model for Uzbek}},
+      author={B. Mansurov and A. Mansurov},
+      year={2021},
+      eprint={2108.09814},
+      archivePrefix={arXiv},
+      primaryClass={cs.CL}
+}
+```