Improve model card: Add `library_name` and prominent links to paper and GitHub

This PR enhances the model card by:

- Adding `library_name: transformers` to the metadata. This is supported by the `config.json` (`"architectures": ["Qwen3ForCausalLM"]`, `"model_type": "qwen3"`, `"transformers_version": "4.53.0"`), which enables the "How to use" widget for direct integration with the Hugging Face `transformers` library.
- Adding prominent links to the paper ([rStar2-Agent: Agentic Reasoning Technical Report](https://huggingface.co/papers/2508.20722)) and the GitHub repository ([https://github.com/microsoft/rStar](https://github.com/microsoft/rStar)) at the top of the model card content for better visibility and easier access to the research and code.
- The existing "Usage", "Citation", and "License" sections remain unchanged as they are already comprehensive and accurate.

These updates will improve the discoverability and usability of the model for the community.

README.md

@@ -1,24 +1,29 @@
 ---
+language:
+- en
+- zh
 license: mit
+pipeline_tag: text-generation
 tags:
 - reinforcement-learning
 - agentic-reasoning
 - math-reasoning
 - tool-use
-language:
-- en
-- zh
-pipeline_tag: text-generation
+library_name: transformers
 ---
+
 # rStar2-Agent-14B: Advanced Agentic Reasoning Model
 
+This model is part of the research presented in the paper [rStar2-Agent: Agentic Reasoning Technical Report](https://huggingface.co/papers/2508.20722).
+Find the official code and training recipes on the [GitHub repository](https://github.com/microsoft/rStar).
+
 ## Model Description
 
 This is a reproduced version of rStar2-Agent, a 14B parameter math reasoning model that achieves performance comparable to 67B DeepSeek-R1 through pure agentic reinforcement learning. The model excels at planning, reasoning, and autonomously using coding tools to efficiently explore, verify, and reflect for complex problem-solving.
 
 ## Usage
 
-This is an example usage. To reproduce the math evaluation results in technical report, please refer to [@microsoft/rstar](https://github.com/microsoft/rstar).
+This is an example usage. To reproduce the math evaluation results in technical report, please refer to [@microsoft/rstar](https://github.com/microsoft/rStar).
 
 ### 1. Start SGLang Server
 
@@ -56,7 +61,11 @@ tools = [
         "type": "function", 
         "function": {
             "name": "execute_python_code_with_standard_io",
-            "description": "Execute Python code with standard input and capture standard output.\nThis function takes a Python code string and an input string, provides the input string\nthrough standard input (stdin) to the code, and captures and returns any output produced\nthrough standard output (stdout). If the executed code raises an exception, the error\nmessage will be captured and returned instead.",
+            "description": "Execute Python code with standard input and capture standard output.
+This function takes a Python code string and an input string, provides the input string
+through standard input (stdin) to the code, and captures and returns any output produced
+through standard output (stdout). If the executed code raises an exception, the error
+message will be captured and returned instead.",
             "parameters": {
                 "type": "object",
                 "properties": {
@@ -144,7 +153,8 @@ while True:
         for tool_call in response.choices[0].message.tool_calls:
             function_args = json.loads(tool_call.function.arguments)
             
-            print(f">>> Executing Code:\n{function_args['code']}")
+            print(f">>> Executing Code:
+{function_args['code']}")
             input_text = function_args.get('input', '')
             print(f">>> With Input: {input_text if input_text else '(no input)'}")
             
@@ -186,4 +196,4 @@ If you use this model in your research, please cite:
 
 ## License
 
-This model is released under the MIT License.
+This model is released under the MIT License.