diff --git a/README.md b/README.md
index 50a35e6..07cb80f 100644
--- a/README.md
+++ b/README.md
@@ -53,6 +53,10 @@ These can also be created on the fly during training, with no need to read files
   - `demo.py`: contains a demo script.
 
 ## To cite
+**The paper is under review.**
+
+For now you can cite this Master's Thesis, which led to the development of what is now _DeepNetSlice_:
+
 ```
 @article{pasquali2023deep,
   title={Deep Reinforcement Learning for Network Slice Placement},
diff --git a/requirements.txt b/requirements.txt
index 64c1446..646bec1 100644
--- a/requirements.txt
+++ b/requirements.txt
@@ -1,8 +1,11 @@
-numpy~=1.23.4
+wheel<=0.38.4
+setuptools~=64.0
 gym~=0.21.0
+numpy~=1.23.4
 networkx~=2.8.7
 torch~=1.12.1
 stable-baselines3~=1.7.0
+sb3-contrib~=1.7.0
 torch-geometric~=2.1.0
 tensorboard~=2.10.0
 wandb~=0.13.4