Update README.md

README.md

@@ -4,7 +4,7 @@
 
 [![arXiv](https://img.shields.io/badge/arXiv-2602.09892-b31b1b.svg)](https://arxiv.org/abs/2602.09892)
 [![Hugging Face Datasets](https://img.shields.io/badge/%F0%9F%A4%97%20Hugging%20Face-Datasets-blue)](https://huggingface.co/collections/AweAI-Team/scale-swe)
-[![Hugging Face Models](https://img.shields.io/badge/%F0%9F%A4%97%20Hugging%20Face-Models-yellow)](https://huggingface.co/AweAI-Team/Scale-SWE)
+[![Hugging Face Models](https://img.shields.io/badge/%F0%9F%A4%97%20Hugging%20Face-Models-yellow)](https://huggingface.co/AweAI-Team/Scale-SWE-Agent)
 [![Website](https://img.shields.io/badge/%F0%9F%8C%90_Project-Website-blue.svg)](https://aweai-team.github.io/projects/scaleswe/)
 [![License](https://img.shields.io/badge/License-CC%20BY%204.0-green.svg)](LICENSE)
 <br>
@@ -13,6 +13,7 @@
 </div>
 
 
+
 ## 🔥 Highlights
 
 - Source from 6M+ pull requests and 23000+ repositories.
@@ -26,8 +27,11 @@
 - **2026-02-26** 🚀 We released a portion of our data on [Hugging Face](https://huggingface.co/collections/AweAI-Team/scale-swe). This release includes **20,000 SWE task instances**—currently the largest **Real Executable** open-source SWE dataset available—alongside **71k distillation trajectories(3.5B)** from DeepSeek v3.2. **Much more data** will be released in the future.
 - **2026-02-10** 📝 Our paper [**"Immersion in the GitHub Universe: Scaling Coding Agents to Mastery"**](https://arxiv.org/abs/2602.09892) is now available on arXiv.
 
-## 📊 Data Format
 
+## FAQ
+- For evaluation of Scale-SWE-Data, you can use AweAgent and refer to this [evaluation script](https://github.com/AweAI-Team/AweAgent/blob/main/awe_agent/tasks/beyond_swe/evaluator.py).
+
+## 📊 Data Format
 
 | Field | Description |
 | :--- | :--- |
@@ -41,15 +45,21 @@
 | **`pr_commit`** | The commit hash of the pull request. |
 | **`parent_commit`** | The commit hash of the parent commit (base state). |
 | **`problem_statement`** | The issue description conveying the bug, provided to the model as input. |
-| **`f2p_patch`** | The developer-written test patch containing tests that fail before the fix (if available). |
-| **`f2p_script`** | The synthetic reproduction script generated by our unit-test creator agent. |
+| **`f2p_patch`** | The developer-written test patch containing tests that fail before the fix (if available). For evaluation, this patch should be applied. See [this script](https://github.com/AweAI-Team/AweAgent/blob/main/awe_agent/tasks/beyond_swe/evaluator.py). |
+| **`f2p_script`** | The synthetic reproduction script generated by our unit-test creator agent. Because a lot of high qaulity pull request do not have author written F2P, we can only synthetic F2P. This should be applied as test_fail_to_pass.py file just under repository directory. just before evaluation. See [this script](https://github.com/AweAI-Team/AweAgent/blob/main/awe_agent/tasks/beyond_swe/evaluator.py). |
 | **`FAIL_TO_PASS`** | Unit tests that fail on the buggy version but pass after the fix. |
 | **`PASS_TO_PASS`** | Unit tests that pass in both versions (regression tests). |
 | **`github_url`** | The URL of the original GitHub repository. |
-| **`pre_commands`** | These commands must be executed immediately upon entering the container to check out the correct commit. |
+| **`pre_commands`** | These commands **must** be executed immediately upon entering the container to check out the correct commit. |
 
-## 🤖 Results
-We fine-tuned Qwen-30B-A3B-Instruct on our synthesized trajectories.
+## Scale-SWE-Agent
+Please use [AweAgent](https://github.com/AweAI-Team/AweAgent) to inference Scale-SWE-Agent. Scale-SWE-Agent model parameter is avaliable at [Huggingface](https://huggingface.co/AweAI-Team/Scale-SWE-Agent). Key parameters can be seen below:
+
+| Parameter | Value |
+| :--- | :--- |
+| Max turns | 200 |
+| Max sequence length | 256k |
+| Temperature | 1 |
 
 
 ## 📖 Citation
@@ -66,7 +76,3 @@ If you find this project useful for your research, please consider citing our pa
       url={https://arxiv.org/abs/2602.09892}, 
 }
 ```
-
-## 📄 License
-
-This project is licensed under the CC BY 4.0 License - see the [LICENSE](LICENSE) file for details.