update readme

README.md

@@ -25,7 +25,7 @@ tags:
     <a href="https://github.com/SapienzaNLP/zebra"><img src="https://img.shields.io/badge/GitHub-Repo-121013?logo=github&logoColor=white"></a> &nbsp; &nbsp;
 </div>
 
-A retrieval augmentation framework for zero-shot commonsense question answering with LLMs. 
+<div align="center"> A retrieval augmentation framework for zero-shot commonsense question answering with LLMs. </div>
 
 ## 🛠️ Installation
 
@@ -48,12 +48,11 @@ pip install -e .
 ## 🚀 Quick Start
 
 ZEBRA is a plug-and-play retrieval augmentation framework for **Commonsense Question Answering**. \
-It is composed of two pipeline stages: *knowledge generation* and *informed reasoning*. \
-The knowledge generation step is responsible for:
-- given a question, retrieving relevant examples of question-knowledge pairs from a large collection
-- prompting a LLM to generate useful explanations for the given input question by leveraging the relationships between the retrieved question-knowledge pairs.
+It is composed of three pipeline stages: *example retrieval*, *knowledge generation* and *informed reasoning*.
 
-The informed reasoning step is responsible for prompting a LLM for the question answering task by taking advantage of the previously generated explanations.
+- Example retrieval: given a question, we retrieve relevant examples of question-knowledge pairs from a large collection
+- Knowledge generation: we prompt an LLM to generate useful explanations for the given input question by leveraging the relationships in the retrieved question-knowledge pairs.
+- Informed reasoning: we prompt the same LLM for the question answering task by taking advantage of the previously generated explanations.
 
 Here is an example of how to use ZEBRA for question answering: