Update README.md

README.md

@@ -17,11 +17,9 @@ Z1-Coder
     <a href="#links" style="text-decoration: none; font-weight: bold;">Links</a> •
     <a href="#getting-started" style="text-decoration: none; font-weight: bold;">Getting Started</a> •
     <a href="#introduction" style="text-decoration: none; font-weight: bold;">Introduction</a> •
-    <a href="#evaluation" style="text-decoration: none; font-weight: bold;">Evaluation</a> •
     <a href="#citation" style="text-decoration: none; font-weight: bold;">Citation</a> 
   </p>
 </div>
-
 </div>
 
 
@@ -32,7 +30,7 @@ Z1-Coder
 
 # Links
 
-
+- [GitHub](https://github.com/Z1-Coder/Z1-Coder)
 - 🤗 [Z1-Coder models](https://huggingface.co/Z1-Coder)
 - 🤗 [Z1-Coder data](https://huggingface.co/Z1-Coder)
 
@@ -52,7 +50,6 @@ We use a learning rate of 5e-5 for the two training stages.
     <em>Figure 1: Comparison between Z1-Coder-7B and Qwen2.5-Coder-Instruct. </em>
 </p>
  -->
-
 To train Z1-Coder, we curate reasoning trajectories on code-related datasets and propose [self-invoking](https://github.com/CodeEval-Pro/CodeEval-Pro) evolving to further refine models' reasoning behaviour in code generation. 
 | Model                  | Trajectory Dataset Download       | Reference                      |
 |------------------------|-----------------------------------|--------------------------------|
@@ -68,7 +65,6 @@ We fine-tune Qwen-2.5-Coder-Base (1.5B and 7B) for two stages with two trajector
     <em>Figure 2: The pipeline of Z1-Coder training. </em>
 </p>
  -->
-
 <!-- # Evaluation
 Z1-Coder significantly outperforms other open-source models on different code generation benchmarks at a similar parameter size. Notably, Z1-Coder-7B surpasses the best 7B code LLMs Qwen2.5-Coder-7B-Instruct, with only its 1% post-training data. Z1-Coder-7B also achieves 20% pass@1 on LiveCodeBench and 51.4% on BigCodeBench, which performs comparable performance level compared to DeepseekCoder-33B-Instruct (21.5% and 51.1%) and LLaMA3.1-70B-Instruct (19.3% and 54.8 %). 
  -->
@@ -78,7 +74,6 @@ in blue, the second-best results are underlined. </em>
     <br>
     <img src="./assets/res1.png" width="700">
 </p> -->
-
 Z1-Coder-7B surpasses the best 7B code LLMs Qwen2.5-Coder-7B-Instruct, with only 1% its post-training data.
 
 | Model                  | Z1-Coder-7B                  | Qwen2.5-Coder-7B-Ins                     |
@@ -103,4 +98,4 @@ The code in this repository is mostly described in the post below. Please consid
   note         = {Accessed: 2025-01-17},
   year         = {2025}
 }
-```
+```