Add paper links, Figure 1 and citation

.gitattributes

@@ -1 +1,2 @@
 *.safetensors filter=lfs diff=lfs merge=lfs -text
+assets/figure1.png filter=lfs diff=lfs merge=lfs -text

README.md

@@ -13,11 +13,13 @@ tags:
 
 # ARLoop-1.2B
 
+📚 [Paper](https://arxiv.org/abs/2611.05234) · 💻 [Code](https://github.com/OSU-NLP-Group/ARLoop)
+
 ARLoop is a fully-open pretrained looped transformer with attention residuals. This release
 contains the checkpoints of the 1.2B, 500B-token training run.
 It is a **base pretrained model**, intended for text completion, which can be used for mid-training and post-training.
 
-Training code: [OSU-NLP-Group/ARLoop](https://github.com/OSU-NLP-Group/ARLoop).
+![Figure 1: ARLoop benchmark results](assets/figure1.png)
 
 ## Load and generate
 
@@ -97,3 +99,17 @@ where 10000 denotes the number of optimization steps.
 | 238,417 (final) | 499,996,688,384 | [main](https://huggingface.co/osunlp/ARLoop-1.2B/tree/main) / [step-238417](https://huggingface.co/osunlp/ARLoop-1.2B/tree/step-238417) |
 
 All 23 scheduled intermediate checkpoints from 10k through 230k are available.
+
+## Citation
+
+```bibtex
+@misc{yao2026arloop,
+  title={{ARLoop}: Fully-Open Pre-trained Looped Transformers with Attention Residuals},
+  author={Yuekun Yao and Yupei Du and Dawei Zhu and Huan Sun},
+  year={2026},
+  eprint={2611.05234},
+  archivePrefix={arXiv},
+  primaryClass={cs.CL},
+  url={https://arxiv.org/abs/2611.05234}
+}
+```