Improve dataset card: Add metadata, HF paper link, and sample usage

README.md

@@ -1,20 +1,28 @@
 ---
 license: apache-2.0
 pretty_name: ImitSAT KeyTrace
-# Show only PREVIEW files in the Viewer (full buckets remain downloadable)
+language:
+- en
+task_categories:
+- reinforcement-learning
+tags:
+- sat
+- boolean-satisfiability
+- imitation-learning
 configs:
 - config_name: preview
   default: true
   data_files:
   - split: train
     path:
-      - preview/sat_5_15_preview.jsonl
-      - preview/sat_16_30_preview.jsonl
-      - preview/sat_31_60_preview.jsonl
-      - preview/sat_50_50_preview.jsonl
-      - preview/sat_61_100_preview.jsonl
-      - preview/sat_100_100_preview.jsonl
+    - preview/sat_5_15_preview.jsonl
+    - preview/sat_16_30_preview.jsonl
+    - preview/sat_31_60_preview.jsonl
+    - preview/sat_50_50_preview.jsonl
+    - preview/sat_61_100_preview.jsonl
+    - preview/sat_100_100_preview.jsonl
 ---
+
 <p align="center">
   <h1 align="center"><em>ImitSAT‑KeyTrace</em> (Download & Preview)</h1>
   <div align="center">
@@ -22,6 +30,10 @@ configs:
       <img src="https://img.shields.io/badge/arXiv-Preprint-brightgreen.svg" alt="arXiv Preprint">
     </a>
     &nbsp;
+    <a href="https://huggingface.co/papers/2509.25411">
+      <img src="https://img.shields.io/badge/HF-Paper-blue" alt="Hugging Face Paper">
+    </a>
+    &nbsp;
     <a href="https://github.com/zewei-Zhang/ImitSAT">
       <img src="https://img.shields.io/badge/GitHub-Repo-black?logo=github" alt="GitHub repository">
     </a>
@@ -32,7 +44,7 @@ configs:
   </div>
 </p>
 
-**Summary.** This repository hosts the bucketed training data for **ImitSAT**, a learned branching policy for CDCL SAT solving.  
+**Summary.** This repository hosts the bucketed training data for **ImitSAT**, a learned branching policy for CDCL SAT solving, presented in the paper [Boolean Satisfiability via Imitation Learning](https://huggingface.co/papers/2509.25411).  
 Each bucket is a gzip‑compressed JSON Lines file (`*.jsonl.gz`) of **KeyTrace** training examples for random 3‑SAT with planted assignments.
 
 > ⚠️ **Use our code to load & preprocess.** We intentionally **do not** provide a `datasets.load_dataset` builder here.  
@@ -52,6 +64,26 @@ git clone https://huggingface.co/datasets/zeweizhang/ImitSAT-KeyTrace
 - [sat_100_100.jsonl.gz](https://huggingface.co/datasets/zeweizhang/ImitSAT-KeyTrace/resolve/main/sat_100_100.jsonl.gz)
 
 
+## Sample Usage
+
+This dataset is designed to be used with the ImitSAT code repository for training. The official loader for this dataset is `ImitSAT/ImitSAT_dataset.py` within the [ImitSAT GitHub repository](https://github.com/zewei-Zhang/ImitSAT).
+
+### Environment Setup
+First, set up the required Python environment as specified in the GitHub repository:
+```bash
+conda create -n imitsat python=3.10 -y
+conda activate imitsat
+pip install -r requirements.txt
+```
+
+### Training ImitSAT
+After setting up the environment and downloading the `ImitSAT-KeyTrace` dataset (as shown in the Download section above), you can train the ImitSAT model by pointing your training configuration's `dataset_path` to the downloaded folder. An example training command from the ImitSAT repository is:
+```bash
+python  train_ImitSAT.py  --config ./model_config/ImitSAT_config.json
+# (Optional) resume from a folder containing ImitSAT.npz and tokenizer/
+python  train_ImitSAT.py  --config ./model_config/ImitSAT_config.json  --resume_dir ./model_ckpt/
+```
+
 ## Preview
 
 The Dataset Viewer shows only small, human‑readable files under preview/ (100 examples each)
@@ -68,4 +100,4 @@ The Dataset Viewer shows only small, human‑readable files under preview/ (100
       primaryClass={cs.AI},
       url={https://arxiv.org/abs/2509.25411}, 
 }
-```
+```