Update README.md

README.md

@@ -7,7 +7,9 @@ pipeline_tag: token-classification
 ---
 # bert-chunker-3
 
-bert-chunker-3 is a text chunker based on BERT with a classifier head to predict the start token of chunks (for use in RAG, etc), and using a sliding window it cuts documents of any size into chunks. Different from [bert-chunker-2](https://huggingface.co/tim1900/bert-chunker-2) and [bert-chunker](https://huggingface.co/tim1900/bert-chunker), the training data were labeled by a LLM, and it is **more stable**.
+bert-chunker-3 is a text chunker based on BERT with a classifier head to predict the start token of chunks (for use in RAG, etc), and using a sliding window it cuts documents of any size into chunks. 
+
+Different from [bert-chunker-2](https://huggingface.co/tim1900/bert-chunker-2) and [bert-chunker](https://huggingface.co/tim1900/bert-chunker), to overcome the data distribution shift, our training data were labeled by a LLM and trainng pipeline was improved, therefore it is **more stable**.
 
 ## Usage
 Run the following: