Update pipeline tag to `graph-ml` and enhance model card introduction

README.md

@@ -1,56 +1,51 @@
 ---
-license: apache-2.0
+base_model:
+- Qwen/Qwen2.5-7B-Instruct
 datasets:
 - PKU-ML/Erdos
 - PKU-ML/Erdos-CoT
 language:
 - en
+library_name: transformers
+license: apache-2.0
 metrics:
 - accuracy
-base_model:
-- Qwen/Qwen2.5-7B-Instruct
-pipeline_tag: text-generation
+pipeline_tag: graph-ml
 tags:
 - graph
 - chat
-library_name: transformers
 ---
 
+# G1-7B: Teaching LLMs to Reason on Graphs with Reinforcement Learning
 
-# G1-7B
+This repository contains the G1-7B model, part of the G1 series of large language models presented in the paper [G1: Teaching LLMs to Reason on Graphs with Reinforcement Learning](https://huggingface.co/papers/2505.18499). G1 models are trained on the [Erdos](https://huggingface.co/datasets/PKU-ML/Erdos) benchmark for solving graph reasoning tasks, based on `Qwen2.5-Instruct`. The approach leverages Group Relative Policy Optimization (GRPO) for reinforcement learning, with supervised finetuning as a preliminary step.
 
-## Introduction
+Code: [https://github.com/PKU-ML/G1](https://github.com/PKU-ML/G1)
 
-G1 is the series of large language models trained on our benchmark [Erdos](https://huggingface.co/datasets/PKU-ML/Erdos) for solving graph reasoning tasks, based on Qwen2.5-Instruct. 
-We apply Group Relative Policy Optimization (GRPO) for reinforcement learning with supervised finetuning as a prelimary step.
-
-G1 brings the following improvements:
+## Introduction
 
-- **Significant improvement on graph reasoning**: G1 models achieve up to 46% improvement over baselines on Erdős, with the 7B variant matching OpenAI’s o3-mini and the 3B model surpassing Qwen2.5-72B-Instruct by notable margins.
-- **Strong Generalization to unseen graph tasks**: G1 exhibits zero-shot generalization on unseen graph tasks, improving performance on *other graph reasoning benchmarks* (GraphWiz, GraphArena) and *real-world graphs* (Cora, PubMed).
-- **NO Compromise on general reasoning**: Crucially, G1 preserves general reasoning ability (GSM8K, MATH, MMLU-Pro), proving its versatility.
+G1 brings the following improvements to graph reasoning with Large Language Models:
 
+-   **Significant improvement on graph reasoning**: G1 models achieve up to 46% improvement over baselines on Erdős, with the 7B variant matching OpenAI’s o3-mini and the 3B model surpassing Qwen2.5-72B-Instruct by notable margins.
+-   **Strong Generalization to unseen graph tasks**: G1 exhibits zero-shot generalization on unseen graph tasks, improving performance on *other graph reasoning benchmarks* (GraphWiz, GraphArena) and *real-world graphs* (Cora, PubMed).
+-   **NO Compromise on general reasoning**: Crucially, G1 preserves general reasoning ability (GSM8K, MATH, MMLU-Pro), proving its versatility.
 
 **This repo contains the G1-7B model**, which has the following features:
-- Type: Causal Language Models
-- Training Stage: SFT & RL
-- Architecture: the same with Qwen2.5-Instruct
-- Number of Parameters: 7.62B
-- Context Length: Full 32,768 tokens and generation 8192 tokens
-
-For more details, please refer to our [paper](https://arxiv.org/pdf/2505.18499) and [GitHub](https://github.com/PKU-ML/G1/tree/main).
-
+-   Type: Causal Language Models
+-   Training Stage: SFT & RL
+-   Architecture: the same with Qwen2.5-Instruct
+-   Number of Parameters: 7.62B
+-   Context Length: Full 32,768 tokens and generation 8192 tokens
 
 ## Requirements
 
-The model is trained based on Qwen/Qwen2.5-7B-Instruct. The code of Qwen2.5 has been in the latest Hugging face `transformers` and we advise you to use the latest version of `transformers`.
+The model is trained based on `Qwen/Qwen2.5-7B-Instruct`. The code of Qwen2.5 has been in the latest Hugging Face `transformers`, and we advise you to use the latest version of `transformers`.
 
 With `transformers<4.37.0`, you will encounter the following error:
 ```
 KeyError: 'qwen2'
 ```
 
-
 ## Quickstart
 
 Here provides a code snippet with `apply_chat_template` to show you how to load the tokenizer and model and how to generate contents.
@@ -73,10 +68,18 @@ model = AutoModelForCausalLM.from_pretrained(
 )
 tokenizer = AutoTokenizer.from_pretrained(model_name)
 
-prompt = "The task is to determine the degree centrality of a node in the graph.\n\n"\
-        "Degree centrality for a node is the fraction of nodes it is connected to.\n\n"\
-        "Here is an undirected graph containing nodes from 1 to 15. The edges are: (1, 15), (15, 11), (2, 3), (2, 6), (3, 6), (3, 7), (6, 7), (6, 8), (7, 8), (7, 14), (4, 10), (10, 5), (10, 12), (8, 14), (8, 9), (12, 11), (12, 13).\n\n"\
-        "Question: What is the degree centrality of node 2 in the graph?\n\n"\
+prompt = "The task is to determine the degree centrality of a node in the graph.
+
+"\
+        "Degree centrality for a node is the fraction of nodes it is connected to.
+
+"\
+        "Here is an undirected graph containing nodes from 1 to 15. The edges are: (1, 15), (15, 11), (2, 3), (2, 6), (3, 6), (3, 7), (6, 7), (6, 8), (7, 8), (7, 14), (4, 10), (10, 5), (10, 12), (8, 14), (8, 9), (12, 11), (12, 13).
+
+"\
+        "Question: What is the degree centrality of node 2 in the graph?
+
+"\
         "You need to format your answer as a float number."
 messages = [
     {"role": "user", "content": INSTRUCTION_TEMPLATE.format(instruction=prompt)}
@@ -103,17 +106,15 @@ response = tokenizer.batch_decode(generated_ids, skip_special_tokens=True)[0]
 print(response)
 ```
 
-
 ## Evaluation & Performance
 
-Detailed evaluation results are reported in this [📑 paper](https://arxiv.org/pdf/2505.18499).
-
+Detailed evaluation results are reported in the [paper](https://huggingface.co/papers/2505.18499).
 
 ## Citation
 
 If you find our work helpful, feel free to give us a cite.
 
-```
+```bibtex
 @article{guo2025g1,
   title={G1: Teaching LLMs to Reason on Graphs with Reinforcement Learning},
   author={Guo, Xiaojun and Li, Ang and Wang, Yifei and Jegelka, Stefanie and Wang, Yisen},