Update README.md

README.md

@@ -13,8 +13,7 @@ tags:
 ---
 
 # Lean-ByT5
-
-**Lean-ByT5** is a ByT5-small model fine-tuned for **Lean theorem proving**.
+**Lean-ByT5** is a ByT5-small model fine-tuned on the Lean-Mathlib data for **Lean theorem proving**.
 
 The model performs **sequence-to-sequence generation**, generating Lean tactics for a given
 statement goal.
@@ -26,7 +25,6 @@ statement goal.
 - PyTorch Lightning
 
 ## Related Paper & Citation
-
 For more details on the training methodology, data augmentation strategy, and dynamic sampling method used for Lean theorem proving, please refer to our paper:
 
 **Enhancing Neural Theorem Proving through Data Augmentation and Dynamic Sampling Method**  
@@ -36,7 +34,6 @@ Rahul Vishwakarma, Subhankar Mishra
 **Paper:** https://arxiv.org/abs/2312.14188
 
 If you use this model or build upon this work, please cite:
-
 ```bibtex
 @article{vishwakarma2023enhancing,
   title={Enhancing Neural Theorem Proving through Data Augmentation and Dynamic Sampling Method},