更新文本

text_content.py

@@ -3,7 +3,8 @@ Based on the DomainEval benchmark, we evaluate code generation ability of differ
             
 More details about how to evaluate the LLM are available in the [DomainEval GitHub repository](https://github.com/domaineval/DomainEval).
 
-For a complete description of DomainEval benchmark and related experimental analysis, please refer to the paper: [DOMAINEVAL: An Auto-Constructed Benchmark for Multi-Domain Code Generation](https://arxiv.org/abs/2408.13204). [![](https://img.shields.io/badge/arXiv-2408.13204-b31b1b.svg)](https://arxiv.org/abs/2408.13204)
+For a complete description of DomainEval benchmark and related experimental analysis, please refer to the paper: 
+[DOMAINEVAL: An Auto-Constructed Benchmark for Multi-Domain Code Generation](https://arxiv.org/abs/2408.13204). [![](https://img.shields.io/badge/arXiv-2408.13204-b31b1b.svg)](https://arxiv.org/abs/2408.13204)
 
 **_Latest News_** 🔥
 - [24/08/26] We release our DomainEval benchmark, leaderboard and paper.
@@ -32,6 +33,6 @@ NOTES_TEXT = """
 - Evaluate using pass@k as the evaluation metric.
 - `Mean` denotes the macro average results of pass@k across 6 different domains.
 - `Std` denotes the standard deviation of pass@k across 6 different domains.
-- you can choose differt pass@k in `⏬ Pass@k`.
+- You can choose differt pass@k in `⏬ Pass@k`.
 - `⏬ Domains` can choose domains you want to show in the leaderboard.
 """