Martin Eberlein

I am a doctoral researcher and Ph.D. candidate in the Software Engineering group of Prof. Lars Grunske at Humboldt-Universität zu Berlin, Germany. I previously collaborated with Prof. Andreas Zeller’s research group at the CISPA Helmholtz Center for Information Security in Saarbrücken.

My research focuses on automated software engineering, with an emphasis on software testing, vulnerability detection, and fuzzing. Currently, I am part of the EMPEROR project, which seeks to automatically generate explanations for program behaviors, particularly program failures. By leveraging predictive and generative models, we aim to uncover and refine relationships among input features, enhancing our understanding of how and why software behaves as it does.

news

Aug 1, 2024	Our Tool FixKit has been accepted at ASE 2024. FixKit is a comprehensive collection of program repair approaches for Python, allowing researchers to easily compare and explore different repair methods on a unified set of techniques. Be sure to check out our pre-print!
Jul 10, 2024	I am honored to have been awarded the prestigious Black Shirt by my research group at Humboldt-Universität Berlin for my contributions to the field of software engineering! The Blackshirt tradition originates from the Nebraska University Football Team, where it is given to exceptional defense players for outstanding performances.
May 4, 2024	I co-founded InputLab, a startup that generates test data for various formats, from electronic invoices to communication between government authorities, covering all input features. We secured €900k in initial funding from the German Federal Ministry of Education and Research! Visit us at inputlab.net.

selected publications

Which Inputs Trigger My Patch?

Martin Eberlein, Moeketsi Raselimo, and Lars Grunske

In In Proceedings of the 6th International Workshop on Automated Program Repair, 2025

Bib

@inproceedings{10.1145/3691620.3695347,
  author = {Eberlein, Martin and Raselimo, Moeketsi and Grunske, Lars},
  title = {Which Inputs Trigger My Patch?},
  year = {2025},
  booktitle = {In Proceedings of the 6th International Workshop on Automated Program Repair},
  series = {APR'25},
}

Semantic Debugging

Martin Eberlein, Marius Smytzek, Dominic Steinhöfel, and 2 more authors

In Proceedings of the 31nd ACM International Conference on the Foundations of Software Engineering, ESEC/FSE 2023, 2023

Bib

@inproceedings{DBLP:conf/sigsoft/EberleinSSGZ23,
  author = {Eberlein, Martin and Smytzek, Marius and Steinh{\"{o}}fel, Dominic and Grunske, Lars and Zeller, Andreas},
  title = {Semantic Debugging},
  booktitle = {Proceedings of the 31nd {ACM} International Conference on
                    the Foundations of Software Engineering, {ESEC/FSE} 2023},
  year = {2023},
}

Evolutionary Grammar-Based Fuzzing

Martin Eberlein, Yannic Noller, Thomas Vogel, and 1 more author

In 12th International Symposium, SSBSE 2020, 2020

Abs Bib

Many works have recently proposed the use of Large Language Model (LLM) based agents for performing ’repository level’ tasks, loosely defined as a set of tasks whose scopes are greater than a single file. This has led to speculation that the orchestration of these repository-level tasks could lead to software engineering agents capable of performing almost independently of human intervention. However, of the suite of tasks that would need to be performed by this autonomous software engineering agent, we argue that one important task is missing, which is to fulfil project level dependency by installing other repositories. To investigate the feasibility of this repository level installation task, we introduce a benchmark of of repository installation tasks curated from 40 open source Python projects, which includes a ground truth installation process for each target repository. Further, we propose INSTALLAMATIC, an agent whose goal is to perform and verify the installation of a given repository by searching for relevant instructions from documentations in the repository. We evaluate our agent using this benchmark, and report that 55% of the studied repositories can be automatically installed by our agent. Through the results of our empirical evaluation, we identify the common causes for our agent’s inability to install a repository, discuss the challenges faced in the design and implementation of such an agent and consider the implications that such an agent could have for developers.
@inproceedings{DBLP:conf/ssbse/EberleinN0G20, title = {Evolutionary Grammar-Based Fuzzing}, author = {Eberlein, Martin and Noller, Yannic and Vogel, Thomas and Grunske, Lars}, year = {2020}, booktitle = {12th International Symposium, {SSBSE} 2020}, }