Fix bilingual README links: use absolute HF URLs so EN/ZH switch and internal links work on Hugging Face

README.md

@@ -10,28 +10,30 @@ tags:
 pipeline_tag: text-generation
 ---
 
+<table width="100%" border="0" cellspacing="0" cellpadding="0">
+<tr>
+<td valign="middle">
+
 # UltraX: Refining Pre-Training Data at Scale with Adaptive Programmatic Editing
 
-<p align="center">
-  <a href="https://huggingface.co/collections/openbmb/ultradata">
-    <img src="https://raw.githubusercontent.com/OpenBMB/UltraX/main/assets/ultradata-logo.png" alt="OpenBMB UltraData" width="350"/>
-  </a>
-  <br/>
-  <a href="https://huggingface.co/collections/openbmb/ultradata">
-    <img src="https://raw.githubusercontent.com/OpenBMB/UltraX/main/assets/ultrax-logo.png" alt="UltraX Logo" width="350"/>
-  </a>
-</p>
+</td>
+<td width="280" align="right" style="padding-top: 14px;" valign="top">
+<a href="https://huggingface.co/collections/openbmb/ultradata"><img src="https://huggingface.co/openbmb/UltraX-0.6B-Preview/resolve/main/assets/ultradata-logo.png" width="250"/></a>
+</td>
+</tr>
+</table>
 
 <p align="center">
 <a href="https://arxiv.org/abs/2607.08646">📜 Paper</a> |
 <a href="https://github.com/openbmb/UltraX">💻 Code</a> |
 <a href="https://huggingface.co/datasets/openbmb/UltraX-Preview">🤗 Datasets</a> |
-<a href="https://huggingface.co/collections/openbmb/ultradata">📦 UltraData Collection</a>
+<a href="https://huggingface.co/collections/openbmb/ultradata">📦 UltraData Collection</a> |
+<a href="https://opensource.org/license/apache-2-0">📄 License: Apache 2.0</a>
 </p>
 
 <p align="center">
 English |
-<a href="README_ZH.md">中文</a>
+<a href="https://huggingface.co/openbmb/UltraX-0.6B-Preview/blob/main/README_ZH.md">中文</a>
 </p>
 
 ## 📚 Introduction
@@ -48,8 +50,7 @@ This repository provides the **UltraX refinement models**, which are trained via
 
 ## 📢 News
 
-- **[2026.07.13]** UltraX codebase, refinement model, and refined datasets are now available on GitHub and Hugging Face. 🚀🚀🚀
-- **[2025.07.10]** UltraX technical report is available on [arXiv](https://arxiv.org/abs/2607.08646). 🔥🔥🔥
+- **[2026.07]** 🎉 We release the UltraX codebase, refinement models, and refined datasets.
 
 ## 💡 Highlights
 
@@ -60,7 +61,7 @@ This repository provides the **UltraX refinement models**, which are trained via
 ## 🔬 Pipeline Overview
 
 <div align="center">
-  <img src="https://raw.githubusercontent.com/OpenBMB/UltraX/main/assets/ultrax_pipeline.png" width="800"/>
+  <img src="https://huggingface.co/openbmb/UltraX-0.6B-Preview/resolve/main/assets/ultrax_pipeline.png" width="800"/>
 </div>
 
 ## 🔍 Model Details
@@ -90,11 +91,11 @@ All models are trained with the following configuration:
 Performance comparison on FineWeb (20B tokens, 1B MiniCPM, 10 benchmarks, zero-shot):
 
 <div align="center">
-  <img src="https://raw.githubusercontent.com/OpenBMB/UltraX/main/assets/results.png" width="900"/>
+  <img src="https://huggingface.co/openbmb/UltraX-0.6B-Preview/resolve/main/assets/results.png" width="900"/>
 </div>
 
 <div align="center">
-  <img src="https://raw.githubusercontent.com/OpenBMB/UltraX/main/assets/fineweb_token_curve.png" alt="FineWeb Token Curve" width="450"/>
+  <img src="https://huggingface.co/openbmb/UltraX-0.6B-Preview/resolve/main/assets/fineweb_token_curve.png" alt="FineWeb Token Curve" width="450"/>
   <p><i>Average downstream performance on FineWeb under different training token budgets.</i></p>
 </div>
 
@@ -126,8 +127,8 @@ remove_lines(12, 15)
 ```python
 from transformers import AutoModelForCausalLM, AutoTokenizer
 
-model = AutoModelForCausalLM.from_pretrained("openbmb/UltraX-0.6B-Preview")
-tokenizer = AutoTokenizer.from_pretrained("openbmb/UltraX-0.6B-Preview")
+model = AutoModelForCausalLM.from_pretrained("openbmb/UltraX")
+tokenizer = AutoTokenizer.from_pretrained("openbmb/UltraX")
 ```
 
 ## ❤️ Acknowledgements
@@ -154,4 +155,6 @@ If you find our work useful, please consider citing:
 
 ## 📜 License
 
-This project is licensed under the [Apache 2.0](./LICENSE) license.
+This project is licensed under the [Apache 2.0](https://huggingface.co/openbmb/UltraX-0.6B-Preview/blob/main/LICENSE) license.
+
+**No unauthorized unchanged redistribution:** Without prior written permission from the original authors (or this organization), any institution, organization, or third-party platform is strictly prohibited from directly reposting, mirroring, re-hosting, or commercially repackaging and republishing any artifacts of this project in any form.

README_ZH.md

@@ -1,14 +1,15 @@
+<table width="100%" border="0" cellspacing="0" cellpadding="0">
+<tr>
+<td valign="middle">
+
 # UltraX：基于自适应程序化编辑的大规模预训练数据精炼
 
-<p align="center">
-  <a href="https://huggingface.co/collections/openbmb/ultradata">
-    <img src="https://raw.githubusercontent.com/OpenBMB/UltraX/main/assets/ultradata-logo.png" alt="OpenBMB UltraData" width="350"/>
-  </a>
-  <br/>
-  <a href="https://huggingface.co/collections/openbmb/ultradata">
-    <img src="https://raw.githubusercontent.com/OpenBMB/UltraX/main/assets/ultrax-logo.png" alt="UltraX Logo" width="350"/>
-  </a>
-</p>
+</td>
+<td width="280" align="right" style="padding-top: 14px;" valign="top">
+<a href="https://huggingface.co/collections/openbmb/ultradata"><img src="https://huggingface.co/openbmb/UltraX-0.6B-Preview/resolve/main/assets/ultradata-logo.png" width="250"/></a>
+</td>
+</tr>
+</table>
 
 <p align="center">
 <a href="https://arxiv.org/abs/2607.08646">📜 论文</a> |
@@ -19,7 +20,7 @@
 </p>
 
 <p align="center">
-<a href="README.md">English</a> |
+<a href="https://huggingface.co/openbmb/UltraX-0.6B-Preview/blob/main/README.md">English</a> |
 中文
 </p>
 
@@ -37,8 +38,7 @@
 
 ## 📢 最新动态
 
-- **[2026.07.13]** UltraX 代码库、精炼模型和精炼数据集现已在 GitHub 与 Hugging Face 上发布。🚀🚀🚀
-- **[2025.07.10]** UltraX 技术报告已发布于 [arXiv](https://arxiv.org/abs/2607.08646)。🔥🔥🔥
+- **[2026.07]** 🎉 UltraX 代码、精炼模型和精炼数据集正式开源。
 
 ## 💡 亮点
 
@@ -49,7 +49,7 @@
 ## 🔬 流程概览
 
 <div align="center">
-  <img src="https://raw.githubusercontent.com/OpenBMB/UltraX/main/assets/ultrax_pipeline.png" width="800"/>
+  <img src="https://huggingface.co/openbmb/UltraX-0.6B-Preview/resolve/main/assets/ultrax_pipeline.png" width="800"/>
 </div>
 
 ## 🔍 模型详情
@@ -77,11 +77,11 @@
 在 FineWeb 上的性能对比（20B tokens，1B MiniCPM，10 个基准，零样本）：
 
 <div align="center">
-  <img src="https://raw.githubusercontent.com/OpenBMB/UltraX/main/assets/results.png" width="900"/>
+  <img src="https://huggingface.co/openbmb/UltraX-0.6B-Preview/resolve/main/assets/results.png" width="900"/>
 </div>
 
 <div align="center">
-  <img src="https://raw.githubusercontent.com/OpenBMB/UltraX/main/assets/fineweb_token_curve.png" alt="FineWeb Token 曲线" width="450"/>
+  <img src="https://huggingface.co/openbmb/UltraX-0.6B-Preview/resolve/main/assets/fineweb_token_curve.png" alt="FineWeb Token 曲线" width="450"/>
   <p><i>FineWeb 在不同训练 token 预算下的平均下游性能。</i></p>
 </div>
 
@@ -114,8 +114,8 @@ remove_lines(12, 15)
 ```python
 from transformers import AutoModelForCausalLM, AutoTokenizer
 
-model = AutoModelForCausalLM.from_pretrained("openbmb/UltraX-0.6B-Preview")
-tokenizer = AutoTokenizer.from_pretrained("openbmb/UltraX-0.6B-Preview")
+model = AutoModelForCausalLM.from_pretrained("openbmb/UltraX")
+tokenizer = AutoTokenizer.from_pretrained("openbmb/UltraX")
 ```
 
 ## ❤️ 致谢
@@ -142,4 +142,6 @@ tokenizer = AutoTokenizer.from_pretrained("openbmb/UltraX-0.6B-Preview")
 
 ## 📜 许可证
 
-本项目基于 [Apache 2.0](./LICENSE) 许可证发布。
+本项目基于 [Apache 2.0](https://huggingface.co/openbmb/UltraX-0.6B-Preview/blob/main/LICENSE) 许可证发布。
+
+**禁止不加处理的二次发布：** 未经原作者（或本机构）书面明确授权，任何其他机构、组织或第三方平台不得以任何形式对本项目成果进行直接转载、复制、托管、镜像克隆或商业化包装再发布。