Update README.md

README.md

@@ -15,6 +15,8 @@ tags:
 # VisualQuality-R1-v1
 This is a demo version of VisualQuality-R1 which is trained on the combination of KADID-10K, TID2013, and KONIQ-10K. The base model of VisualQuality-R1 is Qwen2.5-VL-7B-Instruct.
 
+Paper link: [arXiv](https://arxiv.org/abs/2505.14460)
+
 ## Quick Start
 ```python
 from transformers import Qwen2_5_VLForConditionalGeneration, AutoTokenizer, AutoProcessor