diff --git a/README.md b/README.md
index f2450e0..38426cd 100644
--- a/README.md
+++ b/README.md
@@ -1,27 +1,27 @@
-###<a href="http://arxiv.org/abs/1605.05395">Learning Deep Representations of Fine-grained Visual Descriptions</a>
+### <a href="http://arxiv.org/abs/1605.05395">Learning Deep Representations of Fine-grained Visual Descriptions</a>
 Scott Reed, Zeynep Akata, Honglak Lee, Bernt Schiele
 
 <img src="images/description_embedding.jpg" width="500px" height="250px"/>
 
-#####How to train a char-CNN-RNN model:
+##### How to train a char-CNN-RNN model:
 
 1. Download the [birds](https://drive.google.com/open?id=0B0ywwgffWnLLZW9uVHNjb2JmNlE)
  and [flowers](https://drive.google.com/open?id=0B0ywwgffWnLLcms2WWJQRFNSWXM) data.
 2. Modify the training script (e.g. `train_cub_hybrid.sh` for birds) to point to your data directory.
 3. Run the training script: `./train_cub_hybrid.sh`
 
-#####How to evaluate:
+##### How to evaluate:
 
 1. Train a model (see above).
 2. Modify the eval bash script (e.g. `eval_cub_cls.sh` for birds) to point to your saved checkpoint.
 3. Run the eval script: `./eval_cub_cls.sh`
 
-#####Pretrained models:
+##### Pretrained models:
 
 * [Char-CNN-RNN for birds](https://drive.google.com/open?id=0B0ywwgffWnLLYUNVWVV5Sm1xcWc)
 * [Char-CNN-RNN for flowers](https://drive.google.com/open?id=0B0ywwgffWnLLV205RXF4Y2hFY1E)
 
-#####Citation
+##### Citation
 
 If you find this work useful, please cite as follows: