Update README.md

README.md

@@ -25,4 +25,19 @@ size_categories:
 The expanded XTD-11 dataset, now including Arabic, enhances the original XTD collection. This dataset introduces a 1,000-image multi-lingual MSCOCO2014 caption to test multimodel in zeroshot image or text retrieval in 11 Languages.
 
 
-# Dataset
+## Dataset Details
+
+
+## Citation [optional]
+
+
+@misc{aggarwal2020zeroshot,
+      title={Towards Zero-shot Cross-lingual Image Retrieval},
+      author={Pranav Aggarwal and Ajinkya Kale},
+      year={2020},
+      eprint={2012.05107},
+      archivePrefix={arXiv},
+      primaryClass={cs.CL}
+}
+
+