Prof. Ali Sunyaev, Vice President of the TUM Campus Heilbronn, welcomed the participants and provided an overview of the offerings at Campus Heilbronn and emphasized the proximity of researchers to both small and medium-sized enterprises and international partners, highlighting the synergies that result. Furthermore, he introduced the TUM Heilbronn Data Science Center (HDSC) and explained opportunities for HDSC-MDSI collaborations in tandem projects.

MDSI Best Paper Awards and PhD Fellowships

This year brought two innovations: for the first time, the MDSI Best Paper Award was presented in three categories to showcase the diversity of research under the MDSI umbrella: Societal Impact, Foundational Impact, and Applications. 
The awards carried a prize of €500 per category, and all winners received an official MDSI certificate.
Prof. Stephan Günnemann and Prof. Sunyaev honored Friederike Jungmann and Paul Hager from the groups of MDSI Core Member Prof. Marcus Makowski and MDSI Director Prof. Daniel Rückert at TUM Klinikum rechts der Isar in the Societal Impact category. Their publication critically assessed the performance of large language models (LLMs) such as ChatGPT in clinical decision-making.
Yawar Siddiqui, together with a team led by MDSI Core Members Prof. Angela Dai and Prof. Matthias Niessner, published MeshGPT, a novel approach for generating triangle meshes. Their work won in the Foundational Impact category.
The best publication in the Applications category was submitted by Fabian Pfitzner, MDSI Advisory Board Member Dr.-Ing. Alexander Braun, and MDSI Core Member Prof. André Borrmann. They presented a data-driven pipeline that transforms raw construction site data into structured, usable information.

There was also a fundamental change in the MDSI PhD Fellowships: starting with this year’s fellows, full or 50% funding is available for a three years, providing students with the opportunity to complete their doctoral studies within the funding period.
Daniel Herbst will work on “Foundations of Deep Neural Networks” under the supervision of Prof. Stefanie Jegelka, Tatu Linnala will pursue “AI-based Materials Science” with Prof. Patrick Rinke, and Katharina Schmid will focus on “Visual Computing” in Prof. Matthias Niessner’s group.

“With our funding, we create space for outstanding research and provide targeted support to young talent so their ideas can grow and make an impact.,” said Günnemann.

Current Research Highlights in Keynotes and Science Pitches

In their scientific keynotes, the MDSI Core Members Prof. Sandra Hirche and Prof. Gjergji Kasneci provided fascinating insights into their current research.
Prof. Hirche presented a system that enables patients with limited mobility to regain significant freedom through the use of exoskeletons.
In the MDSI focus topic “AI in Finance”, Prof. Kasneci bridges research and application in the financial sector. AI solutions are being developed for automated trading, risk management, fraud detection, and sustainability. In his keynote, Kasneci introduced new approaches in human-centered AI with a focus on data-driven processes in finance, aiming to promote the responsible use of AI in education and society.

A new format at this year’s assembly was the Science Pitches, short highlight presentations showcasing individual projects.
Prof. Jana Giceva introduced “AnyBlox – A Framework for Self-Decoding Datasets”, which won the Best Research Paper Award at the Conference on Very Large Data Bases (VLDB). Prof. Pierluigi D’Acunto offered a deep dive into “Machine Learning-Enhanced Computational Structural Design” for bridges and roof structures. AMC Director Prof. Julija Zavadlav demonstrated how machine learning can be applied to molecular-level material design, for example, predicting the self-assembly of peptide chains with 100% accuracy, enabling preselection for experimental approaches. Prof. Felix Dietrich pitched on “Physics-Enhanced Machine Learning”, addressing challenges such as the unreliability of ML algorithms and presenting new methods to overcome them.

MDSI Annual Report

Prof. Stephan Günnemann presented the institute’s annual report. In addition to scientific highlights, he emphasized interdisciplinary focus topics and kickstarter projects, successful industry partnerships, the doctoral program, which features numerous workshops on data management, entrepreneurship, and science communication, the TUM Research Data Hub, an update on the new building, as well as MDSI’s extensive outreach activities. These included the new website, the finals of the Bundeswettbewerb Informatik, the Open Day with the Mouse, the Women in Data Science Conference, the Munich Career Fair, and the MOPTAS Autumn School. Günnemann underscored MDSI’s role as a platform for exchange between science, industry, and society.

Panel Discussion and Poster Session

In the subsequent plenary session, Prof. Nikolaus Adams, Prof. Donna Ankerst, Prof. André Borrmann, and Prof. Jana Giceva discussed the further development of MDSI programs. Key topics included supporting early-career researchers, strengthening networking within TUM, and increasing the visibility of the MDSI Core Members’ research.

At the MDSI offices, around 50 researchers from MDSI and the Konrad Zuse School of Excellence in Reliable AI (relAI) presented their projects in a poster session. The wide range of topics, from basic research to practical applications, highlighted the breadth of activities at the institute and provided an opportunity for lively exchange among the scientists.

Further information on the MDSI Best Paper Award 2025 can be found here.