SERG Seminar

In this edition of our weekly SERG seminar we will have two speakers:

Amir Deljouyi will present his ICSE’25 paper on “Leveraging Large Language Models for Enhancing the Understandability of Generated Unit Tests”. It introduces UTGen—a tool designed to improve the understandability of automatically generated unit test cases. While tools like EvoSuite can achieve high test coverage, they often generate tests that are difficult for software engineers to understand.

UTGen addresses these challenges by enhancing test data context, improving identifier naming, and adding descriptive comments to make the tests more interpretable. In a controlled experiment with 32 participants from both academia and industry, we examined how these enhancements influence a software engineer’s ability to fix bugs—a task that mirrors real-world development challenges where the understandability of test cases is crucial. Our findings showed that engineers using UTGen-enhanced test cases fixed 33% more bugs and completed tasks 20% faster than those using standard test cases. Participant feedback further validated that clearer test names, variable names, and data significantly improved the bug-fixing process.

You? The second spot is still free! If you have a complimentary (not too similar) topic you’d like to present in our seminar, please poke Caro!

If you are interested to give a 15 min + 10 talk or host a 25 min discussion session in one of our next meetings, please contact Carolin Brandt via Mattermost or email.