Update README.md

Author: xiaoyao4573

apps/protein_folding/helixfold3/README.md

@@ -2,7 +2,7 @@
 
 The AlphaFold series has transformed protein structure prediction with remarkable accuracy, often matching experimental methods. While AlphaFold2 and AlphaFold-Multimer are open-sourced, facilitating rapid and reliable predictions, [AlphaFold3](https://doi.org/10.1038/s41586-024-07487-w) remains partially accessible and has not been open-sourced, restricting further development.
 
-The PaddleHelix team is working on [HelixFold3](./helixfold3_report.pdf) to replicate the advanced capabilities of AlphaFold3. Insights from the AlphaFold3 paper inform our approach and build on our prior work with [HelixFold](https://arxiv.org/abs/2207.05477), [HelixFold-Single](https://doi.org/10.1038/s42256-023-00721-6), [HelixFold-Multimer](https://arxiv.org/abs/2404.10260), and [HelixDock](https://arxiv.org/abs/2310.13913). Currently, HelixFold3's accuracy in predicting the structures of small molecule ligands, nucleic acids (including DNA and RNA), and proteins is comparable to that of AlphaFold3. We are committed to continuously enhancing the model's performance and rigorously evaluating it across a broader range of biological molecules. Please refer to our [HelixFold3 technical report](./helixfold3_report.pdf) for more details. **The initial version of the HelixFold3 server, designed for biomolecular structure prediction, is also available on the PaddleHelix website (https://paddlehelix.baidu.com/app/all/helixfold3/forecast).** We encourage everyone to explore its capabilities and leverage it for impactful and innovative research.
+The PaddleHelix team is working on [HelixFold3](./helixfold3_report.pdf) to replicate the advanced capabilities of AlphaFold3. Insights from the AlphaFold3 paper inform our approach and build on our prior work with [HelixFold](https://arxiv.org/abs/2207.05477), [HelixFold-Single](https://doi.org/10.1038/s42256-023-00721-6), [HelixFold-Multimer](https://arxiv.org/abs/2404.10260), and [HelixDock](https://arxiv.org/abs/2310.13913). Currently, HelixFold3's accuracy in predicting the structures of small molecule ligands, nucleic acids (including DNA and RNA), and proteins is comparable to that of AlphaFold3. We are committed to continuously enhancing the model's performance and rigorously evaluating it across a broader range of biological molecules. Please refer to our [HelixFold3 technical report](https://arxiv.org/pdf/2408.16975) for more details. **The initial version of the HelixFold3 server, designed for biomolecular structure prediction, is also available on the PaddleHelix website (https://paddlehelix.baidu.com/app/all/helixfold3/forecast).** We encourage everyone to explore its capabilities and leverage it for impactful and innovative research.
 
 
 <!-- <img src="./demo_output/6zcy_demo_result.png" alt="demo" align="middle" style="margin-left: 25%; margin-right: 25%; width: 50%; margin-bottom: 20px;" /> -->