Mobile apps are now indispensable for people’s daily life. To ensure the app quality, automated GUI testing is widely explored for locating bugs. However, there are many issues with current GUI testing tools including low activity coverage, excessive overhead, and missing issues of app usability (e.g., GUI aesthetics or animation) and accessibility (e.g., to the aged and disabled like the blind). The emergence of powerful Large Language Models (LLM) brings an opportunity to overcome these GUI testing issues. In this talk, he is going to introduce his latest works on different aspects of mobile app testing such as boosting GUI testing coverage, knowledge-guided scenario exploratory testing, and multi-agent testing, by leveraging LLM methods. In addition to the academic publications mentioned above, he will also briefly share potential future directions in this field.

I am a professor at Technical University of Munich, Germany. My primary research interests are in Software Engineering, Applied Data Analysis, Deep Learning, and Human-Computer Interaction. I am especially concerned with applying data analysis to automated software development and mining software repositories.