Home  | News

20.04.2026

Teaser image to Research Stay at École Polytechnique

Research Stay at École Polytechnique

Viktoria Ehm – Funded by the MCML AI X-Change Program

As a PhD student at the Technical University of Munich, supervised by Daniel Cremers and affiliated with the Munich Center for Machine Learning (MCML), I recently had the opportunity to spend a research stay at École Polytechnique near Paris. During this visit, I joined the research group of Maks Ovsjanikov, whose work at the intersection of geometry processing, computer vision, and machine learning has long influenced my own research. Being able to spend time with the group in person, learn from them, and discuss ideas directly was a very valuable experience. I am very grateful to Prof. Maks Ovsjanikov and his group for their warm welcome and for the many inspiring conversations throughout my stay.

Scientific Exchange and Collaboration

My research focuses on 3D shape analysis, a field at the intersection of computer vision and computer graphics. Maks Ovsjanikov is one of the leading researchers in this area, so it was a real privilege to work with him during my visit. Our discussions and joint work centred on the role of large language models in correspondence problems. Although I had worked on correspondence problems before, approaching them through language-based foundation models opened a new perspective for me and encouraged me to explore ideas I had not considered before.

École Polytechnique’s campus is located outside the centre of Paris, making it a calm place to focus on research, while offering many opportunities for exchange with the people on campus.

One highlight was a small poster session organised for the whole building, where I had the chance to present my research and speak with researchers from a range of different backgrounds. These conversations were both enjoyable and useful, and they added a broader perspective to my visit.

Outside the University

Beyond the academic side, I also had some time to explore Paris. The city has a great atmosphere, shaped by its history, architecture, and everyday street life. Moving between focused research discussions on campus and walks along the Seine or through different neighbourhoods created a nice balance during the stay.

Being in Paris in the weeks before Christmas made the experience even more memorable. I was able to see the Christmas markets and festive lights, and to get a better sense of how the season is experienced there. I also enjoyed the food very much, from fresh baguettes and pain au chocolat in the morning to galettes later in the day.

Looking Ahead

I enjoyed my stay at Polytechnique a lot and returned with new ideas, new contacts, and a broader perspective on my PhD research.

I am thankful to both groups, especially Daniel Cremers and Maks Ovsjanikov, as well as MCML, for making this exchange possible. Experiences like this show how much international collaboration can contribute to research.

#ai-x-change #blog #cremers

Related

Link to MCML Members Receive IEEE/CVF CVPR 2026 DriveX Best Paper Award

12.06.2026

MCML Members Receive IEEE/CVF CVPR 2026 DriveX Best Paper Award

Daniel Cremers, Malaz Tamim and Johannes Meier received the IEEE/CVF CVPR DriveX Best Paper Award for 3D object detection in autonomous systems.

Read more
Link to How Should Researchers Report Their Use of LLMs?

10.06.2026

How Should Researchers Report Their Use of LLMs?

Is AI making science impossible to replicate? Stefan Feuerriegel and the MCML team introduce the GUIDE-LLM framework in Nature.

Read more
Link to Benjamin Lange: The Real Risk of AI Agents is Manipulation Through Kindness

02.06.2026

Benjamin Lange: The Real Risk of AI Agents Is Manipulation Through Kindness

MCML Junior Research Group Leader Benjamin Lange examines how trust in AI agents can itself become a source of risk.

Read more
Tiny logo
Link to MCML at CVPR 2026

02.06.2026

MCML at CVPR 2026

MCML researchers are represented with 28 papers at CVPR 2026 (26 Main, and 2 Workshops).

Read more
Link to Zeynep Akata: To Trust AI, We Need to Understand What Goes On Behind the Scenes

28.05.2026

Zeynep Akata: To Trust AI, We Need to Understand What Goes on Behind the Scenes

MCML PI Zeynep Akata explains that to trust AI, we must understand its inner workings, address foundation model bias, and make explainability central.

Read more
Back to Top