Update README.md

README.md

@@ -11,7 +11,9 @@ pipeline_tag: token-classification
 
 ## Introduction
 
-bert-chunker 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. It was finetuned on top of [nreimers/MiniLM-L6-H384-uncased](https://huggingface.co/sentence-transformers/all-MiniLM-L6-v2). The whole training lasted for 10 minutes on a Nvidia P40 GPU with a 50 MB synthetized dataset. 
+bert-chunker 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. bert-chunker-3 is a text chunker based on BertForTokenClassification 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. We see it as an alternative of [semantic chunker](https://github.com/FullStackRetrieval-com/RetrievalTutorials/blob/main/tutorials/LevelsOfTextSplitting/5_Levels_Of_Text_Splitting.ipynb). Speciallly, bert-chunker not only works for the structured texts, but also the unstructured and messy texts. 
+
+It was finetuned on top of [nreimers/MiniLM-L6-H384-uncased](https://huggingface.co/sentence-transformers/all-MiniLM-L6-v2). The whole training lasted for 10 minutes on a Nvidia P40 GPU with a 50 MB synthetized dataset. 
 
 A new experimental version is [bert-chunker-2](https://huggingface.co/tim1900/bert-chunker-2).