Add link to Kaiyuan-Spark repo on GitHub

README.md

@@ -22,9 +22,11 @@ it achieves performance competitive with current state-of-the-art fully open mod
   <img alt="Model Performance Comparison" src="model_performance_comparison.svg"/>
 </center>
 
-The dataset used to train Kaiyuan-2B is published at <https://huggingface.co/datasets/thu-pacman/PCMind-2.1-Kaiyuan-2B>.
+Other artifacts:
 
-The _PCMind-2.1-Kaiyuan-2B Technical Report_ is published at <https://arxiv.org/abs/2512.07612>. 
+* The dataset used to train Kaiyuan-2B is published at <https://huggingface.co/datasets/thu-pacman/PCMind-2.1-Kaiyuan-2B>.
+* The _PCMind-2.1-Kaiyuan-2B Technical Report_ is published at <https://arxiv.org/abs/2512.07612>.
+* The data preprocessing framework built on PySpark is published at <https://github.com/thu-pacman/Kaiyuan-Spark>.
 
 ## Introduction
 
@@ -78,7 +80,8 @@ Please cite [our technical report](https://arxiv.org/abs/2512.07612) if you use
 
 ## License
 
-The code and model weights of Kaiyuan-2B are licensed under [Apache-2.0 License](LICENSE) with the following copyright notice.
+All artifacts (including code, model weights, and training data) of Kaiyuan-2B
+are licensed under [Apache-2.0 License](LICENSE) with the following copyright notice:
 
 ```text
 Copyright 2025 Tsinghua University & Peng Cheng Laboratory