Update README.md

README.md

@@ -12,9 +12,11 @@ base_model:
 # Model Card for Model ID
 
 <!-- Provide a quick summary of what the model is/does. -->
-This model is a personalized reward model for pluralistic alignment and also a demonstration for our [paper](https://pal-alignment.github.io/).
-We train the PAL-B-Large model on a variant of Reddit TL;DR summary dataset with 10 users who provide the most amount of feedbacks and achieve higher performance compared with vanilla homogeneous reward model.
+This model is a personalized reward model for pluralistic alignment and serves as a demonstration for our [paper](https://pal-alignment.github.io/). 
 
+We train the PAL-B-Large model on a variant of Reddit TL;DR summary dataset, incorporating feedback from the 10 most active users. 
+
+Our approach outperforms the standard homogeneous reward model, demonstrating improved performance with our proposed Pluralistic Alignment method.
 
 ## Model Details