Add library_name and pipeline_tag metadata

README.md

@@ -1,18 +1,21 @@
 ---
-license: apache-2.0
+base_model:
+- Qwen/Qwen2.5-VL-7B-Instruct
 datasets:
 - inclusionAI/ZoomBench
 - inclusionAI/ZwZ-RL-VQA
 language:
 - en
-base_model:
-- Qwen/Qwen2.5-VL-7B-Instruct
+license: apache-2.0
+library_name: transformers
+pipeline_tag: image-text-to-text
 ---
+
 # ZwZ-7B
 
 <div align="center">
 
-📃 [Paper](https://arxiv.org/pdf/2602.11858) | 🏠 [Project](https://github.com/inclusionAI/Zooming-without-Zooming) | 🤗 [Collection](https://huggingface.co/collections/inclusionAI/zooming-without-zooming)
+📃 [Paper](https://huggingface.co/papers/2602.11858) | 🏠 [Project](https://github.com/inclusionAI/Zooming-without-Zooming) | 🤗 [Collection](https://huggingface.co/collections/inclusionAI/zooming-without-zooming)
 
 </div>
 
@@ -37,7 +40,7 @@ Traditional "Thinking-with-Images" methods zoom into regions of interest during
 ### Installation
 
 ```bash
-pip install transformers accelerate torch
+pip install transformers accelerate torch qwen-vl-utils
 ```
 
 ### Inference
@@ -105,7 +108,7 @@ We introduce [ZoomBench](https://huggingface.co/datasets/inclusionAI/ZoomBench),
 
 ### Benchmark Results
 
-ZwZ-7B achieves state-of-the-art performance among open-source models on fine-grained perception benchmarks. Please refer to the [paper](https://arxiv.org/pdf/YOUR_ARXIV_ID) for detailed results.
+ZwZ-7B achieves state-of-the-art performance among open-source models on fine-grained perception benchmarks. Please refer to the [paper](https://huggingface.co/papers/2602.11858) for detailed results.
 
 ## Limitations