Wednesday, September 9th, 14:15-15:15

 Generating Tests that Matter: Breaking the Boundaries of In-House Testing 

Leonardo Mariani, University of Milano-Bicocca, Italy

ABSTRACT 

Generating test cases is easy; generating test cases that truly matter is far more challenging. The core difficulty lies in producing tests that accurately reflect real‑world usage and align with stakeholder expectations. An alignment that traditional, purely in‑house testing often fails to achieve. This talk argues that field activity can play a pivotal role in bridging this gap. By establishing a continuous feedback loop between the field and the testing process, testers can ground their approaches on authentic usage patterns and evolving stakeholder needs. The talk presents this outside‑in perspective and offers insights into how such a feedback‑driven testing workflow can be implemented.

Leonardo Mariani is a Full Professor at the University of Milano-Bicocca. He holds a PhD in Computer Science, received from the same university. His research interests include software engineering, in particular software testing, program analysis, automated debugging, AI4SE, and self-adaptive systems. He has authored more than 200 papers appeared at top software engineering conferences and journals. He received the ICST MIP award for his ICST 2013 paper. He was awarded the ERC Consolidator Grant in 2015, an ERC Proof of Concept grant in 2018. He is regularly involved in the organization of major software engineering conferences. 

Homepage – https://en.unimib.it/leonardo-mariani 

Friday, September 11th, 10.45-11.45

Beyond the Pass Rate: Humans, Benchmarks, and the Limits of AI-Based Software Engineering

Gabriele Bavota, Università della Svizzera italiana (USI), Switzerland

ABSTRACT 

AI-based tools can generate code, review pull requests, and autonomously address software issues. Their progress is often described through two indicators: higher benchmark scores and increased developer productivity. Yet these indicators provide only a partial view of what success means in AI-based software engineering. In this talk, I revisit the current narrative surrounding AI for software engineering through empirical evidence on both its technical and human dimensions. I discuss how we assess the capabilities of AI coding tools, what may be hidden behind apparently strong performance, and how their use can reshape developers' work and behavior. 

Gabriele Bavota is a full professor and Vice Dean at the Faculty of Informatics of the Università della Svizzera italiana (USI), Switzerland, where he is part of the Software Institute and he leads the SEART research group. He received the PhD in Computer Science from the University of Salerno, Italy, in 2013. His research interests include AI for software engineering, software maintenance and evolution, code quality, mining software repositories, and empirical software engineering. On these topics, he authored over 200 papers appeared in international journals and conferences and has received 15 best/distinguished paper awards. He is the recipient of the 2018 ACM Sigsoft Early Career Researcher Award for outstanding contributions in the area of software engineering as an early career investigator and has been the principal investigator of the DEVINTA ERC Starting Grant.

Homepage – https://www.inf.usi.ch/faculty/bavota/

SEDES KEYNOTE

Wednesday, September 9th, 12.00-13.00

Show me your data. How to (not) visualize research data in SE

António Lopes

ABSTRACT 

The usage of figures to represent data or concepts in scientific articles is a common practice. The use of visualization can improve retrieval time, leverage high information density and provide information in context.

Unfortunately, when used inappropriately, such advantages disappear, leading to added confusion.

This talk will describe how figures are used in SE articles and will focus on how quantitative data is represented, the most common types of plots, and their quality.

A general critical theory of visualization quality (rooted in Tufte's integrity principles) will be presented, which can be used as a guideline for crafting and improving your own plots.

Marco Torchiano is full professor of Software Engineering at the Control and Computer Engineering Dept. of Politecnico di Torino, Italy; he has been post-doctoral research fellow at Norwegian University of Science and Technology (NTNU), Norway. He received an MSc and a PhD in Computer Engineering from Politecnico di Torino. He is Senior Member of the IEEE, member of the software engineering committee of UNINFO (part of ISO/IEC JTC 1), faculty fellow of the Nexa Center on Internet & Society. He is author or co-author of over 200 research papers published in international journals and conferences, of the book ‘Software Development—Case studies in Java’ from Addison-Wesley, and co-editor of the book ‘Developing Services for the Wireless Internet’ from Springer. He has been a visiting professor at Polytechnique Montréal studying software energy consumption. His current research interests are: UI testing methods, open-data quality, and software modeling notations. The methodological approach he adopts is that of empirical software engineering.

Homepage – https://www.polito.it/en/staff?p=marco.torchiano