Improve dataset card: Add paper/code links, task categories, and sample usage

README.md

@@ -3,8 +3,25 @@ language:
 - en
 size_categories:
 - 10K<n<100K
+task_categories:
+- text-generation
+- reinforcement-learning
 ---
 
-This is the RL training dataset from RePro: Training Language Models to Faithfully Recycle the Web for Pretraining.
+# RePro RL Training Dataset
 
-Each entry contains two columns, dataman_score and text.
+This repository contains the Reinforcement Learning (RL) training dataset used in the paper [RePro: Training Language Models to Faithfully Recycle the Web for Pretraining](https://huggingface.co/papers/2510.10681).
+
+RePro proposes a novel web recycling method that trains a relatively small Language Model (LM) with reinforcement learning to generate effective and faithful rephrasings of pretraining data. This dataset serves as the RL training data for the rephraser model.
+
+Each entry contains two columns: `dataman_score` and `text`.
+
+**Code Repository:** [https://github.com/cxcscmu/RePro](https://github.com/cxcscmu/RePro)
+
+### Sample Usage
+
+This dataset is utilized for training the RePro rephraser model. According to the project's GitHub repository, you can initiate the rephraser training with the following command:
+
+```bash
+bash scripts/rl.sh
+```