Welcome to MCSE

MCSE: Modelling for Continuous Software Engineering (previously known as ARE: Architecture-driven Requirements Engineering) has been started in February 2013 and is headed by Prof. Anne Koziolek. Since 2021, we have been a member of the KASTEL — Institute of Information Security and Dependability, and we renamed the research group to MCSE. The group is part of Software Design and Quality.

On Thursday, December 3rd, the SDQ Winter Day 2026 will take place, starting at 15:45 at the Infobau (building 50.34) in front of room 237. There, the research groups MCSE (KASTEL Koziolek), DSiS (KASTEL Reussner), and SASIS (KASTEL Mirandola) will present themselves to interested students. In addition to presentations from research and industry, there will be free mulled wine, non-alcoholic punch, and waffles. There is also the opportunity to discuss thesis topics with members of the groups.
More information about the program can be found at the SDQ Wiki Winter Day Page.
Link to SDQ WIKI Winter Day Page09.07.2026
The paper "Criminal Minds: How First-Year CS Students Plagiarize Code" has received a Distinguished Paper award of the Software Engineering Education Track at the 2026 ACM International Conference on the Foundations of Software Engineering (FSE). Congratulations to the authors, Robin Maisch, Larissa Schmid (KTH), Richard Glassey (KTH), Dominik Fuchß, Nils Niehues, Haoyu Liu, and Anne Koziolek!
In their work, the researchers have designed and conducted a study to collect and analyze the strategies and tools of first-term students at KIT and KTH when trying to obfuscate code plagiarism in a lab session. The results revealed fascinating LLM interactions, provided insights into students' conceptions about plagiarism detection, and form a valuable documentation of the current state of LLM proficiency of first-term students.
View news at FSE website
26.02.2026
At the SE26, Larissa Schmid successfully presented her dissertation "Automatic Performance Modeling and Analysis of Configurable Scientific Software and Workflows" and won the dissertation award (Dissertationspreis) of the Software Engineering Division (SWT) of the German Informatics Society (GI).
We congratulate her on this immaculate achievement!
SE26 Dissertationspreis
23.02.2026
A part of our team from SDQ had the privilege of attending and presenting at the Software Engineering conference of the Software Engineering Division (SWT) of the German Informatics Society (GI), organised in cooperation with the Swiss Informatics Society (SI) and the Austrian Computer Society (OCG).
At the conference, taking place in the stadium where the wonder of Bern happened in 1954, we had the opportunity to host a workshop about Research Data Management, presented our best papers from 2025, and Larissa Schmid successfully presented her dissertation for the Dissertationspreis.
SE26 Homepage
03.11.2025
The paper titled "HubLink: A Novel Question Answering Retrieval Approach over Knowledge Graphs" by Angelika Kaplan, Jan Keim, Marco Schneider, and Ralf Reussner was praised as exceptional contribution and awarded best paper in the full paper category at the workshop for Retrieval-Augmented Generation Enabled by Knowledge Graphs (RAGE-KG) at the ISWC 2025.
In their paper, the authors introduce HubLink, a schema-agnostic and training-free approach for Question Answering over Knowledge Graphs that uses large language models to improve scholarly search and source-aware inference. Evaluated on the Open Research Knowledge Graph, HubLink outperforms state-of-the-art baselines, particularly for complex scholarly queries.
Link to the paper
20.10.2025
We are happy to share that our paper “Same Same But Different: Preventing Refactoring Attacks on Software Plagiarism Detection” by Robin Maisch, Larissa Schmid, Timur Sağlam, and Nils Niehues has been accepted for presentation at the International Conference on Software Engineering (ICSE) 2026.
The work addresses a longstanding challenge in academic integrity: the vulnerability of source code plagiarism detection tools to refactoring-based obfuscation. The authors introduce Nocte, an extensible framework that leverages graph transformations on code property graphs to normalize code structure. This approach significantly strengthens the resilience of plagiarism detection systems against structural modifications often used to conceal copied code.
Congratulations to all authors on this excellent achievement!
ICSE26 Homepage