diff --git a/recommendations/README.md b/recommendations/README.md
index a41cd7dd8..badf3d141 100644
--- a/recommendations/README.md
+++ b/recommendations/README.md
@@ -1 +1,27 @@
-Run the recommendations code to produce a set of scores and recs. 
+Run the recommendations code to produce a set of scores and recs.
+The model learns a `1024 dimensional embedding` corresponding to each paper based on the abstract data.
+
+## Setup & Usage
+
+#### Install git-lfs, required to download the saved model later.
+```bash
+#### Linux
+apt-get install git-lfs
+
+#### MacOS
+brew install git-lfs
+```
+
+```bash
+git lfs install
+rm -fr iclr.github.io
+git clone https://github.com/ICLR/iclr.github.io
+pip install -r iclr.github.io/requirements.txt
+pip install -r iclr.github.io/recommendations/requirements.txt
+```
+
+To create the embeddings for the papers using the abstract information, run the following python script.
+
+```bash
+python cache_paper_embeddings.py
+```
\ No newline at end of file
diff --git a/recommendations/cache_paper_embeddings.py b/recommendations/cache_paper_embeddings.py
index 4b294e24a..8b96491c5 100644
--- a/recommendations/cache_paper_embeddings.py
+++ b/recommendations/cache_paper_embeddings.py
@@ -8,7 +8,7 @@
 from models import load_model
 import torch
 
-accepted_submissions = pickle.load(open("../cached_or.pkl", "br"))
+accepted_submissions = pickle.load(open("../data/pkl/cached_or.pkl", "br"))
 
 # Load the model
 abstracts = []