Add pipeline tag, project links and sample usage

Hi! I'm Niels from the community science team at Hugging Face.

I've opened this PR to improve the discoverability and usability of the EvoToken-DLM adapter. Here's a summary of the changes:
- Added the `text-generation` pipeline tag to the metadata.
- Included links to the official project page and GitHub repository.
- Added a sample usage section with the inference command from the official repository to help users get started.

Please let me know if you have any questions!

README.md

@@ -1,35 +1,46 @@
 ---
-license: bsd-2-clause
-language:
-- en
 base_model:
 - GSAI-ML/LLaDA-8B-Instruct
+language:
+- en
 library_name: transformers
+license: bsd-2-clause
+pipeline_tag: text-generation
 tags:
 - DLM
 - EvoToken
 - lora
 ---
-# EvoTokenDLM LoRA adapter training from pretrained weights LLaDA-8B-Instruct
 
-Starting from the original MDLM (Masked Discrete Diffusion Language Model) LLaDA-8B-Instruct, we trained the EvoTokenDLM LoRA adapter using the **Continuous Trajectory Supervision** method.
+# EvoToken-DLM (LoRA Adapter)
 
-Our implementation replaces traditional hard binary masks with evolving soft token distributions. This allows EvoTokenDLM to facilitate a progressive transition from masked states to discrete outputs, effectively supporting revisable decoding.
+[**Project Page**](https://aim-uofa.github.io/EvoTokenDLM/) | [**GitHub**](https://github.com/aim-uofa/EvoTokenDLM) | [**Paper**](https://arxiv.org/abs/2601.07351)
 
-The method and its results are detailed in the paper: [Beyond Hard Masks: Progressive Token Evolution for Diffusion Language Models](https://arxiv.org/abs/2601.07351).
+EvoToken-DLM is a novel diffusion-based language modeling approach that replaces hard binary masks with evolving soft token distributions. While most Diffusion Language Models (DLMs) rely on hard binary masking and discrete token assignments, which can hinder the revision of early decisions, EvoToken-DLM enables a progressive transition from masked states to discrete outputs, supporting revisable decoding.
+
+This repository provides the LoRA adapter weights trained from [LLaDA-8B-Instruct](https://huggingface.co/GSAI-ML/LLaDA-8B-Instruct) using **Continuous Trajectory Supervision**.
 
 ## How to Use
 
-⚠️ **Important:** This is a LoRA adapter and requires the official EvoTokenDLM codebase for inference. 
+⚠️ **Important:** This is a LoRA adapter and requires the [official EvoTokenDLM codebase](https://github.com/aim-uofa/EvoTokenDLM) for inference. 
+
+### Sample Inference Command
 
-For detailed instructions and code, please refer to the official GitHub repository: [EvoTokenDLM GitHub Repository](https://github.com/aim-uofa/EvoTokenDLM)
+Once you have set up the environment following the instructions in the official repository, you can run progressive inference using the following command:
 
+```bash
+python generate.py  --model_path GSAI-ML/LLaDA-8B-Instruct \
+    --checkpoint_path zhongzero/EvoToken_LLaDA_Instruct_8B_Lora \
+    --prompt "Lily can run 12 kilometers per hour for 4 hours. After that, she runs 6 kilometers per hour. How many kilometers can she run in 8 hours?" \
+    --k_soft 3 \
+    --alpha_soft_mask 0.7
+```
 
 ## Citation
 
 If you find this work helpful for your research, please cite:
 
-```BibTeX
+```bibtex
 @article{zhong2026beyond,
     title={Beyond Hard Masks: Progressive Token Evolution for Diffusion Language Models},
     author={Zhong, Linhao and Wu, Linyu and Fang, Bozhen and Feng, Tianjian and Jing, Chenchen and Wang, Wen and Zhang, Jiaheng and Chen, Hao and Shen, Chunhua},