Add pipeline_tag, library_name and paper reference

README.md

@@ -1,34 +1,35 @@
 ---
-license: apache-2.0
+base_model:
+- Qwen/Qwen2.5-VL-3B-Instruct
 datasets:
 - GUI-Libra/GUI-Libra-81K-RL
 - GUI-Libra/GUI-Libra-81K-SFT
 language:
 - en
-base_model:
-- Qwen/Qwen2.5-VL-3B-Instruct
+license: apache-2.0
+library_name: transformers
+pipeline_tag: image-text-to-text
 tags:
 - VLM
 - GUI
 - agent
 ---
 
-# Introduction
-
-The models from paper "GUI-Libra: Training Native GUI Agents to Reason and Act with Action-aware Supervision and Partially Verifiable RL".
+# GUI-Libra-3B
 
+[**Project Page**](https://gui-libra.github.io) | [**Paper**](https://huggingface.co/papers/2602.22190) | [**GitHub**](https://github.com/GUI-Libra/GUI-Libra)
 
-**GitHub:** https://github.com/GUI-Libra/GUI-Libra
-**Website:** https://GUI-Libra.github.io  
+GUI-Libra is a post-training framework that turns open-source VLMs into strong native GUI agents—models that see a screenshot, think step-by-step, and output an executable action, all within a single forward pass.
 
+This model is fine-tuned from [Qwen2.5-VL-3B-Instruct](https://huggingface.co/Qwen/Qwen2.5-VL-3B-Instruct) using action-aware SFT and conservative reinforcement learning (GRPO). It addresses challenges such as action-grounding alignment and partial verifiability in GUI navigation tasks.
 
 # Usage
 ## 1) Start an OpenAI-compatible vLLM server
 
 ```bash
 pip install -U vllm
-vllm serve GUI-Libra/GUI-Libra-4B --port 8000 --api-key token-abc123
-````
+vllm serve GUI-Libra/GUI-Libra-3B --port 8000 --api-key token-abc123
+```
 
 * Endpoint: `http://localhost:8000/v1`
 * The `api_key` here must match `--api-key`.
@@ -79,11 +80,17 @@ action_type: Scroll, action_target: None, value: "up" | "down" | "left" | "right
 
 task_desc = 'Go to Amazon.com and buy a math book'
 prev_txt = ''
-question_description = '''Please generate the next move according to the UI screenshot {}, instruction and previous actions.\n\nInstruction: {}\n\nInteraction History: {}\n'''
+question_description = '''Please generate the next move according to the UI screenshot {}, instruction and previous actions.
+
+Instruction: {}
+
+Interaction History: {}
+'''
 img_size_string = '(original image size {}x{})'.format(img_size[0], img_size[1])
 query = question_description.format(img_size_string, task_desc, prev_txt)
 
-query = query + '\n' + '''The response should be structured in the following format:
+query = query + '
+' + '''The response should be structured in the following format:
 <think>Your step-by-step thought process here...</think>
 <answer>
 {
@@ -111,19 +118,18 @@ resp = client.chat.completions.create(
 print(resp.choices[0].message.content)
 ```
 
-Run:
-
-```bash
-python minimal_infer.py
-```
-
----
-
-## Notes
-
-* Replace `screen.png` with your own screenshot file.
-* If you hit OOM or slowdowns, reduce image size or run fewer concurrent requests.
-* The example assumes your vLLM server is running locally on port `8000`.
-
+## Citation
 
+If you find GUI-Libra useful for your research, please cite:
 
+```bibtex
+@misc{yang2026guilibratrainingnativegui,
+      title={GUI-Libra: Training Native GUI Agents to Reason and Act with Action-aware Supervision and Partially Verifiable RL}, 
+      author={Rui Yang and Qianhui Wu and Zhaoyang Wang and Hanyang Chen and Ke Yang and Hao Cheng and Huaxiu Yao and Baoling Peng and Huan Zhang and Jianfeng Gao and Tong Zhang},
+      year={2026},
+      eprint={2602.22190},
+      archivePrefix={arXiv},
+      primaryClass={cs.LG},
+      url={https://arxiv.org/abs/2602.22190}, 
+}
+```