Name: Lamarr Scientific Forum
Start: 2025-09-03T08:30:00+02:00
End: 2025-09-04T18:00:00+02:00
Location: Hörsaalgebäude, Campus Poppelsdorf, Universität Bonn

Lamarr Scientific Forum

from Wednesday, September 3, 2025 (8:30 AM) to Thursday, September 4, 2025 (6:00 PM)

Monday, September 1, 2025
Tuesday, September 2, 2025
Wednesday, September 3, 2025

8:30 AM Arrival & Registration
Arrival & Registration
8:30 AM - 9:30 AM

9:30 AM "Lamarr Today and Tomorrow"
"Lamarr Today and Tomorrow"
9:30 AM - 10:15 AM
Room: Lecture Hall 2 (ground floor)
10:15 AM Session 1
Session 1
10:15 AM - 12:45 PM
Room: Lecture Hall 2 (ground floor)
Contributions

10:15 AM Area Presentation: Hybrid ML - Ramsés Sánchez (Lamarr institute, University of Bonn) Paul Roetzer

10:45 AM Area Presentation: Planning & Logistics - Anike Murrenhoff Nilah Nair

11:15 AM Area Presentation: Resource-Aware Machine Learning - Jian-Jia Chen Sebastian Buschjäger (Lamarr Institute for ML and AI, TU Dortmund)

11:45 AM Area Presentation: Physics - Wolfgang Rhode (TU Dortmund) Pascal Gutjahr (TU Dortmund University)

12:15 PM Area Presentation: Trustworthy AI
12:45 PM Lunch
Lunch
12:45 PM - 2:00 PM
Room: Foyer (ground floor)
2:00 PM Session 1
Session 1
2:00 PM - 3:30 PM
Room: Open Space (first floor)
Contributions

2:00 PM VISOR: VIsual Seizure Onset detection peRsonalized for epilepsy patients - Uttam Kumar

2:00 PM Mitigating Emerging Misalignment with Training Regularization - David Kaczér (University of Bonn)

2:00 PM Stretch-Compose: Semantic-Geometric Reasoning based Open Vocabulary Search and Retrieval of Objects in Dynamic Environments - Rohit Menon

2:00 PM Hybrid Quantum-Classical Multi-Agent Pathfinding - Loong Kuan Lee (Fraunhofer IAIS)

2:00 PM Cluster-based prediction of chatter vibrations in milling operations - Florian Wöste

2:00 PM Towards Causal Understanding in Oncology: Structural Causal Modeling with Missing Data - Michael Kamp Osman Mian (The Lamarr Institute for Machine Learning and Artificial Intelligence)

2:00 PM Fast, Registration-Free, and Lesion- Robust Tractography Parcellation using a Transformer Model - Justus Bisten (b-it and Institute for Computer Science II & Department for Neuroradiology, University Hospital Bonn)

2:00 PM On the Limitations of Language Targeted Pruning - Simon Kurz

2:00 PM Completing the Puzzle of Unseen Perception in Grasping and Navigation - Anas Gouda (TU Dortmund)

2:00 PM Multi-Hop Reasoning for Question Answering with Hyperbolic Representations - Akbar Karimi (University of Bonn)

2:00 PM Interaction-Aware and Domain-Invariant Representation Learning for Inclusive Flavour Tagging - Quentin Führing

2:00 PM Towards a Dataset of Realistic 3D Intralogistics Scenes for AI Applications - Anike Murrenhoff

2:00 PM Context-Based Meta Reinforcement Learning for Robust and Adaptable Peg-in-Hole Assembly Tasks - Ahmed Shokry

2:00 PM Property Testing of Curve Similarity - Marena Richter

2:00 PM Load Balancing Neurons: Controlling Firing Rates Improves Plasticity in Continual Learning - Jan Robine (TU Dortmund)

2:00 PM Neurodivergence in AI Research: Traits, Stress, and Well-being - Nicolo' Brandizzi (Fraunhofer IAIS)

2:00 PM Rule vs. SHAP: Complementary Tools for Understanding and Verifying ML Models - Bahavathy Kathirgamanathan (Fraunhofer IAIS)

2:00 PM Deep Learning for Real-time Classification of Astronomical Radio Signals: Evaluating the Applicability of Synthetic Data for Training - Andrei Kazantsev (Max Planck Institute for Radio Astronomy)

2:00 PM Long-Horizon Flare Forecasting in Blazars Time Series - Amal Saadallah (Lamarr Institute-TU Dortmund)

2:00 PM GRaMPa: Subword Regularisation by Skewing Uniform Segmentation Distributions with an Efficient Path-counting Markov Model - David Kaczér (University of Bonn)

2:00 PM Diff2SAT: SAT Instance Generation via Graph Diffusion - Lukas Schneider (LS8 TU Dortmund)

2:00 PM Flatness is Necessary, Neural Collapse is Not: Rethinking Generalization via Grokking - Ting Han

2:00 PM Adversarial Perturbations Improve Generalization of Confidence Prediction in Medical Image Segmentation - Jonathan Lennartz

2:00 PM Synthetic Data for Multi-Camera Multi-Object Tracking in Logistics - Christian Pionzewski (Fraunhofer IML)

2:00 PM Zero-shot Imputation with Foundation Inference Models for Dynamical Systems - Patrick Seifner (University of Bonn)

2:15 PM AI-based sensor layout for predicting thermal deformations of CFRP machine tools - Felix Finkeldey
3:30 PM Grab your coffee and head to the next session
Grab your coffee and head to the next session
3:30 PM - 3:45 PM
Room: Transfer from first to ground floor

3:45 PM Spearheading European, open-source Foundation Models
Spearheading European, open-source Foundation Models
3:45 PM - 5:15 PM
Room: Lecture Hall 2 (ground floor)

5:15 PM Closing of day one
Closing of day one
5:15 PM - 5:30 PM
Room: Lecture Hall 2 (ground floor) The Lamarr Scientific Forum is rounding off the first day with a closing and all information needed on dinner plans.

5:30 PM [Time for Transfer and Check-In]
[Time for Transfer and Check-In]
5:30 PM - 7:00 PM

7:00 PM Dinner
Dinner
7:00 PM - 10:00 PM
Thursday, September 4, 2025

9:00 AM Welcome
Welcome
9:00 AM - 9:15 AM
Room: Lecture Hall 2 (ground floor) We begin program day number 2 with a short look back at the previous day and ahead at today's program.
9:15 AM Session 2
Session 2
9:15 AM - 11:45 AM
Room: Lecture Hall 2 (ground floor)
Contributions

9:15 AM Area Presentation: NLP - Lucie Flek (University of Bonn) Akbar Karimi (University of Bonn)

9:45 AM Area Presentation: Human-Centered Systems - Natalia Andrienko (Fraunhofer Institute IAIS)

10:15 AM Area Presentation: AI in Life Sciences and Health - Jürgen Bajorath Elena Xerxa (University of Bonn) Andrea Mastropietro (University of Bonn)

10:45 AM Area Presentation: Embodied AI - Julian Eßer Sven Behnke (Universität Bonn)

11:15 AM Area Presentation: Industry and Production - Felix Finkeldey
11:45 AM Lunch
Lunch
11:45 AM - 1:00 PM
Room: Foyer (ground floor)
1:00 PM Session 2
Session 2
1:00 PM - 2:15 PM
Room: Open Space (first floor)
Contributions

1:00 PM Anytime YOLO - Early exits for interruptable object detection - Daniel Kuhse

1:00 PM Sustainable Model Selection via Meta-Learning - Raphael Fischer (TU Dortmund University)

1:00 PM CINEMETRIC: A Framework for Multi-Perspective Dialogue Evaluation Using Role-Playing of Large Language Models - Vahid Sadiri Javadi (University of Bonn)

1:00 PM Efficient Manipulation-Enhanced Semantic Mapping With Uncertainty-Informed Action Selection - Nils Dengler (University of Bonn)

1:00 PM AI and Astroparticle Physics with Neutrinos - Tim Ruhe (TU Dortmund)

1:00 PM Little Is Enough: A Privacy-Preserving Framework for Healthcare Models - Amr Abourayya (Trustworthy AI)

1:00 PM Constructive Empiricism for Explainable AI - Sebastian Müller (University of Bonn)

1:00 PM Reduction of data labeling effort by stability evaluation using drift detection techniques for milling operations - Shubham Gupta

1:00 PM Adaption of process monitoring motivated by a combination of rejection ensembles - Jim Bergmann

1:00 PM Predicting machining stability with a quantum regression model - Sascha Mücke (TU Dortmund)

1:00 PM Hierarchical Vector Quantization for Unsupervised Action Segmentation - Federico Spurio (University of Bonn)

1:00 PM AALF: Almost Always Linear Forecasting - Matthias Jakobs (TU Dortmund)

1:00 PM First Investigation of Deep Learning for Intraoperative Gauze Segmentation in Minimally Invasive Abdominal Surgery - Priya Priya (Fraunhofer IAIS, University of Bonn)

1:00 PM MCBench: A Benchmark Suite for Monte Carlo Sampling Algorithms - Zeyu Ding (TU Dortmund)

1:00 PM Simplified random forest models and topic modelling for human-understandable explanation of compound potency predictions - Tiago Janela

1:00 PM LLM Value Alignment - Shangrui Nie (Bonn-Aachen International Center for Information Technology (b-it))

1:00 PM Synthetic Data from Physics Simulations: A Viable Alternative for Pallet Activity Recognition? - Alexander Krooß (Fraunhofer IML)

1:00 PM State-Transition-Aware Anomaly Detection Under Concept Drifts - Shubham Gupta Daniil Kaminskyi (TU Dortmund University)

1:00 PM Deep Learning Applications in Radio Interferometry - Kevin Schmitz

1:00 PM Towards Human Motion Generation for Industrial Simulations - Nilah Nair

1:00 PM JSON is all you need (for visual symbolic planning) - Sami Azirar

1:00 PM The Anatomy of Evidence: An Investigation Into Explainable ICD Coding - Katharina Beckh

1:00 PM Jailbreaking LLMs Without Gradients or Priors: Effective and Transferable Attacks - Zhakshylyk Nurlanov (Learning and Optimisation for Visual Computing Group, University of Bonn)

1:00 PM Chi: Symmetry Understanding of 3D Shapes via Chirality Disentanglement - Weikang Wang (Learning and Optimisation for Visual Computing Group, University of Bonn) Tobias Weißberg (Learning and Optimisation for Visual Computing Group, University of Bonn)

1:00 PM Pairwise-TD: A Bellman Operator for Relative Value Learning - Marc Höftmann
2:15 PM Grab your coffee and head to the next session
Grab your coffee and head to the next session
2:15 PM - 2:30 PM
Room: Transfer from first to ground floor
2:30 PM Introduction
Introduction
2:30 PM - 3:30 PM
Room: Lecture Hall 2 (ground floor)
Contributions

2:30 PM Research Center Trustworthy Data Science and Security (RC Trust) - Emmanuel Müller

2:45 PM PhenoRob - Robotics and Phenotyping for Sustainable Crop Production - Cyrill Stachniss

3:00 PM Bonn Sustainable AI Lab at IWE

3:15 PM Robotics Institute Germany (RGI) - Maren Bennewitz
3:30 PM Closing of the Event
Closing of the Event
3:30 PM - 3:45 PM
Room: Lecture Hall 2 (ground floor)