2026 ASEE Annual Conference & Exposition

Grass: Automatically Generating Reviewer-friendly Evidence Reports from Plagiarism Detection Tools

Presented at CIT Technical Session 2: Assessment, Evaluation, and Academic Integrity.

Addressing student programming plagiarism presents a dual challenge: initial code similarity detection and subsequent formal case reporting. While tools like MOSS are effective at the first step, instructors face a significant, time-consuming manual effort in verifying cases, identifying collusion groups, and translating complex technical evidence into a defensible report for a non-computing academic integrity board. This manual preparation often requires undoing "disguising" techniques used by students and summarizing results for a lay audience. This paper describes experiences with GRASS (Generating Plagiarism Reports for Programming Assignments Using Grass), a standalone processing system designed to automate this critical reporting workflow. GRASS works by feeding submissions into a similarity detection tool like MOSS and then analyzing the raw similarity results to identify multi-student collusion groups, processing the source code to remove obfuscatory differences, and generating comprehensive case reports tailored specifically for non-programmer reviewers. These reports include a written summary of the findings, multiple descriptive visualizations, and side-by-side code comparisons that highlight essential structural similarities. The primary objective of GRASS is to eliminate the manual effort required after a potential violation is detected. By providing automatically generated, high-quality, and easily digestible evidence reports, GRASS significantly reduces the workload on course staff, encourages consistent adherence to academic integrity policies, and streamlines the path from initial detection to disciplinary review. This paper describes the design of GRASS, how it operates, and our experiences using it over multiple programming courses.

Authors

Jennifer Alexandra Thompson Virginia Polytechnic Institute and State University
Prof. Stephen H Edwards http://orcid.org/https://0000-0002-5162-9314 Virginia Polytechnic Institute and State University [biography]

Stephen H. Edwards is a Professor and the Associate Department Head for Undergraduate Studies in the Department of Computer Science at Virginia Tech, where he has been teaching since 1996. He received his B.S. in electrical engineering from Caltech, and M.S. and Ph.D. in computer and information science from The Ohio State University. His research interests include computer science education, software testing, software engineering, and programming languages. He is the project lead for Web-CAT, the most widely used open-source automated grading system in the world. Web-CAT is known for allowing instructors to grade students based on how well they test their own code. In addition, his research group has produced a number of other open-source tools used in classrooms at many other institutions. Currently, he is researching innovative methods for giving feedback to students as they work on assignments to provide a more welcoming experience for students, recognizing the effort they put in and the accomplishments they make as they work on solutions. The goals of his research are to strengthen growth mindset beliefs while encouraging deliberate practice, self-checking, and skill improvement as students work.
Dr. Bob Edmison Virginia Polytechnic Institute and State University [biography]

Dr. Edmison is a Collegiate Associate Professor in the Department of Computer Science at Virginia Tech. His research focuses on alternative grading practices, accessibility, and the tools to support them.

Note

The full paper will be available to logged in and registered conference attendees once the conference starts on June 21, 2026, and to all visitors after the conference ends on June 24, 2026

« View session

For those interested in:

2 Year Institution
computer science
information technology
undergraduate