tim1900
/

bert-chunker-3

Token Classification

Model card Files Files and versions

tim1900 commited on 15 days ago

Commit

cac741b

·

verified ·

1 Parent(s): 563c4f6

Update README.md

Files changed (1) hide show

README.md +2 -1

README.md CHANGED Viewed

@@ -14,7 +14,8 @@ bert-chunker-3 is a text chunker based on BertForTokenClassification to predict
 Different from [bc-2](https://huggingface.co/tim1900/bert-chunker-2) and [bc](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**. It has a **competitive** [**performance**](#evaluation).
 Updates :
-- 2025.5.12: an experimental script that **supports specifying the maximum tokens per chunk** is available now [below](#experimental), evaluation is at [**evaluation**](#evaluation).
 ## Usage
 Run the following:

 Different from [bc-2](https://huggingface.co/tim1900/bert-chunker-2) and [bc](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**. It has a **competitive** [**performance**](#evaluation).
 Updates :
+- 2026.1.1: Newest version is [bc-3.5](https://huggingface.co/tim1900/bert-chunker-3.5). Better performance, less bugs, simpler code and strategy.
+- 2025.5.12: An experimental script that **supports specifying the maximum tokens per chunk** is available now [below](#experimental), evaluation is at [**evaluation**](#evaluation).
 ## Usage
 Run the following: