ROUGE 2.0 is an easy to use evaluation toolkit for Automatic Summarization tasks. It uses the ROUGE system of metrics which works by comparing an automatically produced summary or translation against a set of reference summaries (typically human-produced). ROUGE is one of the standard ways to compute effectiveness of auto generated summaries. To understand how ROUGE works you can read this article.


The latest version of ROUGE 2.0 supports the following:

This is an open-source project, so if you want to implement other measures, please read the documentation on how to contribute to this project.

Quick Start


To cite the ROUGE 2.0 package, please use the following:

  title={ROUGE 2.0: Updated and Improved Measures for Evaluation of Summarization Tasks},
  author={Ganesan, Kavita},

You can download the ROUGE 2.0 report from arXiv.


To build on the source code of ROUGE 2.0, clone the repository and import the project as a Maven project by selecting the pom.xml file in the root directory from Eclipse, IntelliJ or other IDE’s. Next, add the jars under the lib directory to your build path for the code to compile. Note that you can also install the jars in the lib directory by creating a local maven repository for those jar files. All you would need to do after that is to add the local dependencies to your pom.xml.

Reporting Issues & Feature Requests

When reporting issues please use “Github Issues” and include as much detail as possible about your operating system and java version. Whenever possible, please also include a brief, self-contained code example that demonstrates the problem. Visuals are appreciated!

If you need new features that would be useful to the community in general, please open up an issue and label it as feature-request.

Contributing to ROUGE 2.0

Thanks for your interest in contributing code to ROUGE2.0! To contribute code to ROUGE2.0, please create a new branch and submit a pull request early on with your ideas and planned changes or enhancement. Once you get the thumbs up, you can use your usual workflow for development. Your changes will be reviewed and finally accepted for inclusion. Please use meaningful commit messages that intuitively explain what you are trying to achieve.

If you are proposing new evaluation measures, your changes will be included once your paper is accepted at a conference or journal.


Please check this license information.