Conference Agenda
Overview and details of the sessions of this conference. Please select a date or location to show only sessions at that day or location. Please select a single session for detailed view (with abstracts and downloads if available).
|
Daily Overview |
| Session | ||
Poster Session I
| ||
| Session Abstract | ||
|
Discover cutting-edge AI research and innovations at the Poster Session. Connect with authors, ask questions, and engage in lively discussions that spark collaboration and new ideas. | ||
    | ||
ID: 358
/ Poster No. # 9: 001
Modalities: Image Methods: Foundation Models, Other Application Domain: Health Interpretable Representations for Hematology 1Helmholtz Munich, Germany; 2Medical Department for Hematology and Oncology, Technical University Munich ID: 255
/ Poster No. # 9: 002
Modalities: Graphs, Multimodal Data Methods: Graph Neural Networks Application Domain: Health Analysis of multi-omics data using graph neural networks identifies novel schizophrenia-associated genes suitable as drug targets 1Helmholtz Munich, Germany; 2Boehringer Ingelheim Pharma, Germany; 3Technical University of Munich, Germany ID: 1218
/ Poster No. # 9: 003
Modalities: Audio, Image, Tabular Data, Text, Time Series, Video Methods: Other Application Domain: Core Machine Learning MLflow pilot service for Helmholtz researchers Karlsruhe Institute of Technology (KIT), Germany ID: 1273
/ Poster No. # 9: 004
Modalities: Image Methods: Foundation Models Application Domain: Health Towards Robust Foundation Models for Digital Pathology 1Berlin Institute for the Foundations of Learning and Data (BIFOLD), Berlin, Germany; 2Machine Learning Group, Technische Universität Berlin, Berlin, Germany; 3Aignostics, Berlin, Germany; 4The Netherlands Cancer Institute Amsterdam (NKI), Antoni van Leeuwenhoek Hospital (AvL), Amsterdam, Netherlands; 5Institute of Pathology, Ludwig Maximilian University, Munich, Germany; 6German Cancer Research Center, Heidelberg, and German Cancer Consortium, Munich, Germany; 7Institute of Pathology, Charité Universitätsmedizin, Berlin, Germany; 8Department of Artificial Intelligence, Korea University, Seoul, Korea; 9Max-Planck Institute for Informatics, Saarbrücken, Germany ID: 1253
/ Poster No. # 9: 005
Modalities: Image, Simulation Data, Text, Video Methods: Generative Models Application Domain: Information, Matter Sailing Past Syntax: A Human-in-the-Loop Framework for Safe Generative AI in Science Centre de Physique des Particules de Marseille, France ID: 1331
/ Poster No. # 9: 006
Modalities: Audio, Graphs, Image, Multimodal Data, Simulation Data, Tabular Data, Text, Time Series, Video Methods: Other Application Domain: Core Machine Learning, Information The Helmholtz Model Zoo: Enabling AI Model Sharing and Inference in the Helmholtz Cloud Deutsches Elektronen-Synchroton DESY, Germany ID: 1230
/ Poster No. # 9: 007
Modalities: Simulation Data, Text Methods: Foundation Models, Generative Models Application Domain: Core Machine Learning Improving Reliability of LLM-Based Robotic Task Planning Through Domain Adaptation and Benchmarking ARENA2036, Germany ID: 1184
/ Poster No. # 9: 009
Modalities: Image, Multimodal Data Methods: Generative Models, Uncertainty Quantification Application Domain: Health Uncertainty-Guided Generation of Dark-Field Radiographs 1School of Computation, Information and Technology, Technical University of Munich, 85748 Garching, Germany; 2Munich Center for Machine Learning (MCML), Munich, Germany; 3Institute of Machine Learning in Biomedical Imaging, Helmholtz Munich, 85764 Neuherberg, Germany; 4Department of Physics, School of Natural Sciences, Technical University of Munich, 85748 Garching, Germany; 5Munich Institute of Biomedical Engineering, Technical University of Munich, 85748 Garching, Germany; 6Institute for Diagnostic and Interventional Radiology, School of Medicine and Health, TUM University Hospital Klinikum rechts der Isar, Technical University of Munich, 81675 Munich, Germany; 7Institute for Advanced Study, Technical University of Munich, 85748 Garching, Germany; 8School of Biomedical Engineering & Imaging Sciences, King’s College, London, UK ID: 1167
/ Poster No. # 9: 010
Modalities: Image, Video Methods: Generative Models Application Domain: Health, Information SoraCT: Unconditioned 3D CT Synthesis via Video Diffusion Transformers 1Friedrich-Alexander University Erlangen-Nürnberg, Germany; 2Department of Computing, Imperial College London, UK ID: 1394
/ Poster No. # 9: 011
Modalities: Image Methods: Foundation Models Application Domain: Health A 3D Foundation Model for Generalizable Biological Structure Segmentation in Tissue Clearing Images (DISCO-CAR: whole-body mapping of engineered cells and diseases at single-cell resolution) 1Institute for Stroke and Dementia Research, Klinikum der Universität München, Ludwig-Maximilians University Munich, Munich, Germany; 2Institute for Intelligent Biotechnologies (iBIO), Helmholtz Center Munich, Neuherberg, Germany; 3Faculty of Medicine, Ludwig-Maximilians University Munich, Munich, Germany; 4Munich Cluster for Systems Neurology (SyNergy), Munich, Germany; 5Munich Medical Research School (MMRS), Munich, Germany; 6Deep Piction GmbH, Munich, Germany; 7School of Medicine, Koç University, İstanbul, Turkey; 8TUM School of Computation, Information and Technology, Technical University of Munich, Munich, Germany; 9German Center for Infection Research (DZIF), Partner Site Munich, Munich, Germany; 10Institute for Medical Microbiology, Immunology and Hygiene, Technische Universität München (TUM), Munich, Germany; 11Focus Group 'Clinical Cell Processing and Purification', Institute for Advanced Study, TUM, Munich, Germany ID: 295
/ Poster No. # 9: 012
Modalities: Graphs, Tabular Data, Text Methods: Agentic AI, Foundation Models, Uncertainty Quantification Application Domain: Aeronautics, Space & Transport, Information M²S³OM-graph: A Hybrid Deterministic-LLM Pipeline for Automated Metadata Crosswalk Extraction and Graph-Grounded Interoperability Deutsches Zentrum für Luft- und Raumfahrt e. V. (DLR), Germany ID: 370
/ Poster No. # 9: 013
Modalities: Image Methods: Foundation Models, Uncertainty Quantification Application Domain: Earth & Environment EO Foundation Models for Pre-Training Dataset Uncertainty Analysis German Aerospace Center, Remote Sensing Technology Institute, Germany ID: 160
/ Poster No. # 9: 014
Modalities: Simulation Data, Other Methods: Physics-informed Machine Learning Application Domain: Health, Information FrustrAI-Seq: Scaling Local Energetic Frustration to the Protein Sequence Space 1Helmholtz Munich, Germany; 2Technical University of Munich, Germany; 3Barcelona Supercomputing Center, Spain; 4University College London, United Kingdom ID: 199
/ Poster No. # 9: 015
Modalities: Image Methods: Generative Models Application Domain: Information muBRAND: Multi-Branch Diffusion for Flexible Generation and Inpainting of Image Series 1Forschungszentrum Jülich, Germany; 2Helmholtz AI, Jülich, Germany; 3C. und O. Vogt Insitut, Universitätsklinikum Düsseldorf, Düsseldorf, Germany; 4Institute of Computational Visualistics, Koblenz University, Koblenz, Germany ID: 367
/ Poster No. # 9: 016
Modalities: Graphs, Image, Multimodal Data, Time Series, Other Methods: Reinforcement Learning, Probabilistic Methods, Uncertainty Quantification, Other Application Domain: Health, Information Multimodal Analysis of Organoid Network Dynamics Using Multielectrode Arrays and Calcium Imaging Helmholtz Munich, Germany ID: 206
/ Poster No. # 9: 017
Modalities: Multimodal Data, Simulation Data Methods: Physics-informed Machine Learning Application Domain: Energy, Matter Towards Multimodal Inference of Electron Bunch Structures using Gradient-Based Reconstruction with Differentiable Detector Models 1Helmholtz-Zentrum Dresden-Rossendorf, Germany; 2Ludwig-Maximilians-Universität München, Germany; 3Shanghai Jiao Tong University, China; 4Center for Advanced Systems Understanding, Germany; 5Technische Universität Dresden, Germany; 6Technische Universität Chemitz, Germany ID: 262
/ Poster No. # 9: 018
Modalities: Image Methods: Other Application Domain: Health NURBSeg: bridging the gap between medical imaging and CAD via end-to-end parametric segmentation 1German Aerospace Center (DLR), Germany; 2German Cancer Research Center (DKFZ), Germany ID: 216
/ Poster No. # 9: 019
Modalities: Other Methods: Generative Models Application Domain: Health From prediction to generation of antigen-specific, full-length T-cell receptors 1Helmholtz Munich, Germany; 2Technical University of Munich, Germany ID: 261
/ Poster No. # 9: 020
Modalities: Graphs, Tabular Data, Text, Time Series Methods: Agentic AI, Probabilistic Methods, Other Application Domain: Energy, Matter Provenance-Aware Knowledge Graphs for Reusable PEFC and PEWE Experiments 1Institute of Energy Technologies (IET-4), Electrochemical Process Engineering, Forschungszentrum Jülich GmbH, 52425 Jülich, Germany; 2Theory and Computation of Energy Materials (IET-3), Institute of Energy Technologies, Forschungszentrum Jülich GmbH, 52425 Jülich, Germany; 3Chair for Theory and Computation of Energy Materials, Faculty of Georesources and Materials Engineering, RWTH Aachen University, 52062 Aachen, Germany; 4Data Analytics, Access and Applications, Scientific Computing Center (SCC), Karlsruher Institut für Technologie, 76344 Eggenstein-Leopoldshafen, Germany; 5Helmholtz AI, Karlsruher Institut für Technologie, 76344 Eggenstein-Leopoldshafen, Germany ID: 327
/ Poster No. # 9: 021
Modalities: Multimodal Data Methods: Foundation Models, Generative Models, Graph Neural Networks Application Domain: Core Machine Learning, Health Deep Learning Modeling of RNA Structure Perturbations Induced by Small Molecules 1Computational Health Center, Helmholtz Munich, Munich, Germany; 2School of Computation, Information and Technology, Technical University of Munich, Munich, Germany; 3Department of Molecular Genetics, Groningen Biomolecular Sciences and Biotechnology Institute (GBB), University of Groningen, Groningen, The Netherlands; 4Equal contributions ID: 189
/ Poster No. # 9: 022
Modalities: Multimodal Data, Simulation Data, Other Methods: Probabilistic Methods, Other Application Domain: Health T-time: Multimodal Manifold Learning for Clonally Constrained Trajectory Inference 1Helmholtz Munich, Germany; 2Mila – Quebec AI Institute, Montréal, Canada; 3Department of Mathematics and Statistics, Université de Montréal ID: 248
/ Poster No. # 9: 023
Modalities: Text, Other Methods: Foundation Models, Generative Models, Reinforcement Learning Application Domain: Health End-To-End Protein Engineering With Protein Language Models 1Institute of Computational Biology, Computational Health Center, Helmholtz Munich, Germany; 2Department of Bioinformatics and Computational Biology , Technical University of Munich, Germany ID: 161
/ Poster No. # 9: 024
Modalities: Image Methods: Other Application Domain: Health AI-assisted Labeling and its Pitfalls: A Case Study in Electron Microscopy Segmentation 1Helmholtz AI, Helmholtz Center Munich, Germany; 2Institute of Toxicology and Environmental Hygiene, TUM School of Medicine and Health, Technical University of Munich, Germany; 3Institute of Molecular Toxicology and Pharmacology, Helmholtz Center Munich, Germany ID: 323
/ Poster No. # 9: 025
Modalities: Image, Multimodal Data, Time Series Methods: Foundation Models Application Domain: Core Machine Learning, Earth & Environment Spatial and Temporal Evaluation of Extreme Precipitation in ML-Based Global Weather Forecasts Forschungszentrum Jülich, Germany ID: 311
/ Poster No. # 9: 026
Modalities: Audio, Text, Time Series, Video Methods: Other Application Domain: Core Machine Learning, Information Recurrent Neural Networks and Tensor Ring Decomposition KU Leuven, Belgium ID: 326
/ Poster No. # 9: 027
Modalities: Graphs, Tabular Data, Text Methods: Foundation Models, Graph Neural Networks, Other Application Domain: Health AI and Hierarchical Clustering Techniques for Accurate Patient Stratification 1Hochschule München, Campus for Health and Engineering, Munich, Germany; 2PerMediQ, Stuttgart, Germany; 3University of Stuttgart, HLRS - High Performance Computing Center, Stuttgart, Germany; 4Klinikum Stuttgart, Stuttgart Cancer Center - Tumorzentrum Eva Mayr-Stihl DE, Stuttgart, Germany; 5University Hospital Regensburg, Department of Internal Medicine I, Regensburg, Germany ID: 362
/ Poster No. # 9: 028
Modalities: Text, Other Methods: Agentic AI, Graph Neural Networks, Other Application Domain: Core Machine Learning, Aeronautics, Space & Transport, Energy, Earth & Environment, Information, Matter SciTracer - the reliable and logically provable autonomous scientific discovery. Unaffiliated, Ukraine ID: 300
/ Poster No. # 9: 029
Modalities: Image Methods: Other Application Domain: Earth & Environment Benchmarking for Microscopic Biodiversity Screening (AIMBIS-project) 1Helmholtz-Centre for Environmental Research GmbH (UFZ), Germany; 2German Centre for integrative Biodiversity Research e.V. Halle-Jena-Leipzig (iDiv), Germany; 3Helmholtz-Centre for Polar and Marine Research – AWI, Department/Section Polar Terrestrial Environmental Systems; 4Helmholtz-Centre for Polar and Marine Research – AWI, Department/Section Polar Biological Oceanography ID: 245
/ Poster No. # 9: 030
Modalities: Simulation Data, Tabular Data, Text, Other Methods: Generative Models, Probabilistic Methods, Other Application Domain: Core Machine Learning, Matter MolecuLA: Learning Molecules As A Language For Generative And Interpretable Chemistry 1CASUS/Helmholtz-Zentrum Dresden-Rossendorf, Germany; 2University of Wrocław, Poland ID: 342
/ Poster No. # 9: 031
Modalities: Simulation Data Methods: Generative Models, Other Application Domain: Health Construct-valid and adaptive benchmarks for single-cell RNA-seq via a hierarchical, streaming simulator 1Chair for Clinical Bioinformatics, Saarland University, Germany; 2Helmholtz Institute for Pharmaceutical Research Saarland ID: 344
/ Poster No. # 9: 032
Modalities: Image, Multimodal Data Methods: Generative Models, Probabilistic Methods, Uncertainty Quantification Application Domain: Core Machine Learning, Health From T-FAKE to TFAN: Reusable Thermal Face Alignment for AI in Scientific Imaging 1Chair for Clinical Bioinformatics, Saarland University, Germany; 2Technical University of Berlin, Germany; 3MPI for Informatics, SIC, Germany; 4Helmholtz Institute for Pharmaceutical Research, Germany; 5Equal author contribution ID: 201
/ Poster No. # 9: 033
Modalities: Image, Other Methods: Probabilistic Methods, Other Application Domain: Core Machine Learning From Feature Learning to Neural Collapse: A Data-Dependent Kernel View 1Helmholtz Munich, Germany; 2Technical University of Munich, TUM School of Computation, Information and Technology; 3Faculty of Mathematics, Informatics and Mechanics, University of Warsaw ID: 354
/ Poster No. # 9: 035
Modalities: Multimodal Data, Time Series, Video Methods: Other Application Domain: Core Machine Learning Do world models model the world? Helmholtz Munich, Germany ID: 173
/ Poster No. # 9: 036
Modalities: Image Methods: Foundation Models Application Domain: Core Machine Learning, Health Editing the Foundations: Task Arithmetic for 3D Medical Segmentation Foundation Models 1Helmholtz Munich, Germany; 2Technical University of Munich, Germany; 3King's College London, UK ID: 285
/ Poster No. # 9: 037
Modalities: Image, Multimodal Data Methods: Foundation Models, Other Application Domain: Health Spatial transcriptomics-informed inference of cell types and states from nuclear-stained whole slide images. 1Institute of Computational Biology (ICB), Helmholtz Zentrum München, Munich, Germany; 2Institute for Stroke and Dementia Research (ISD), Klinikum der Universität München, Munich, Germany; 3German Center for Neurodegenerative Diseases (DZNE), Munich, Germany; 4Munich Cluster of Systems Neurology (SyNergy), Munich, Germany; 5Institute of Neuronal Cell Biology, Technical University Munich, Munich, Germany ID: 277
/ Poster No. # 9: 038
Modalities: Image Methods: Uncertainty Quantification Application Domain: Health Uncertainty Quantification for Medical Image Segmentation HZDR, Germany ID: 281
/ Poster No. # 9: 039
Modalities: Simulation Data, Text Methods: Agentic AI, Other Application Domain: Core Machine Learning Fast and Few-Shot: Equipping Scientific LLM Agents with Meta-Learned Surrogates 1Institute for Advanced Simulations – Materials Data Science and Informatics (IAS-9), Forschungszentrum Jülich GmbH, 52425 Jülich, Germany; 2Chair of Materials Data Science and Materials Informatics, Faculty 5 – Georesources and Materials Engineering, RWTH Aachen University, 52056 Aachen, Germany ID: 397
/ Poster No. # 9: 040
scooby: modeling multimodal genomic profiles from DNA sequence at single-cell resolution. 1Technical University of Munich, School of Computation, Information and Technology, Munich, Germany; 2Helmholtz Center Munich, Computational Health Center, Neuherberg, Germany; 3Technical University of Munich, Munich Center for Machine Learning, Munich, Germany; 4Technical University of Munich, Institute of Human Genetics, School of Medicine, Munich, Germany ID: 197
/ Poster No. # 9: 042
Modalities: Image, Multimodal Data, Time Series Methods: Physics-informed Machine Learning, Uncertainty Quantification, Other Application Domain: Earth & Environment Geospatial Causal Inference for Drought Driver Analysis DLR, Germany ID: 283
/ Poster No. # 9: 043
Modalities: Graphs, Multimodal Data, Simulation Data, Text Methods: Generative Models, Graph Neural Networks, Other Application Domain: Core Machine Learning, Health When Protein Dynamics Matter: Integrating Molecular Dynamics into Protein Foundation Models jülich forschungszentrum, Germany ID: 314
/ Poster No. # 9: 044
Modalities: Image, Multimodal Data Methods: Probabilistic Methods, Uncertainty Quantification, Other Application Domain: Health Toward Automated MRI-Based Screening of Cortical Superficial Siderosis Using Multimodal 3D nnU-Net 1Institute for Stroke and Dementia Research(ISD), University Hospital, LMU Munich,Munich, Bavaria, Germany; 2Institute of Computational Biology (ICB), Helmholtz Munich, Neuherberg, Germany ID: 321
/ Poster No. # 9: 045
Modalities: Multimodal Data, Tabular Data Methods: Generative Models, Probabilistic Methods, Uncertainty Quantification Application Domain: Core Machine Learning, Health Cross-View Latent Integration via Nonparametric Gamma Shrinkage Factor Analysis 1Institute of AI for Health, Helmholtz Munich; 2Faculty of Mathematics, Informatics and Mechanics, University of Warsaw; 3Oncology Data Science, Merck Healthcare KGaA ID: 209
/ Poster No. # 9: 046
Modalities: Image Methods: Generative Models, Probabilistic Methods Application Domain: Health Physics-Guided Diffusion for Neonatal Ultra-Low-Field MRI enhancement 1University of Bonn, University Hospital Bonn, Clinic for Diagnostic and Interventional Radiology, Germany; 2University of Bonn, University Hospital Bonn, Department of Experimental Neonatology, Germany; 3Technical University of Munich, School of Computation, Information and Technology, Germany; 4Laboratoire Traitement du Signal et de l’Image, Inserm UMR 1099, Universit´ e de Rennes, France ID: 286
/ Poster No. # 9: 047
Modalities: Image Methods: Foundation Models Application Domain: Health Atlas 2 - Foundation Models for Clinical Deployment 1Aignostics, Germany; 2Department of Laboratory Medicine and Pathology, Mayo Clinic, Rochester, MN, US; 3Department of Radiology, Mayo Clinic, Rochester MN, US; 4Department of Information Technology, Mayo Clinic, Rochester MN, US; 5Mayo Clinic, Rochester MN, US; 6Digital Pathology, Mayo Clinic, Rochester MN, US; 7Machine Learning Group, Technische Universität Berlin, Germany; 8BIFOLD – Berlin Institute for the Foundations of Learning and Data, Germany; 9Department of Artificial Intelligence, Korea University, Republic of Korea; 10Max-Planck Institute for Informatics, Germany; 11German Cancer Research Center (DKFZ) & German Cancer Consortium (DKTK), Berlin & Munich Partner Sites, Germany; 12Institute of Pathology, Ludwig-Maximilians-Universität München, Germany; 13Institute of Pathology, Charité – Universitätsmedizin Berlin, Germany; 14Bavarian Cancer Research Center (BZKF), Germany; 15Helmholtz Munich, Germany; 16Technical University Munich, Germany ID: 219
/ Poster No. # 9: 048
Modalities: Audio, Image, Multimodal Data, Simulation Data, Text, Time Series, Video Methods: Agentic AI, Foundation Models, Generative Models Application Domain: Core Machine Learning EOSC-ARENA: Towards an Open and Ethical GenAI Infrastructure for European Research 1Scientific Computing Center (SCC), Karlsruhe Institute of Technology (KIT), Eggenstein-Leopoldshafen, 76344,Germany; 2Research Data Alliance Association - RAD, Belgium; 3Instituto de Física de Cantabria (IFCA), CSIC-UC, Avda. los Castros s/n, Santander, 39006, Cantabria, Spain; 4Instituto de Instrumentación para Imagen Molecular (I3M), Centro Mixto CSIC — Universitat Politècnica de València, Camino de Vera s/n, 46022 Valencia, España; 5Ethniko Kentro Erevanas Kai Technologikis Anaptyxis - CERTH, Greece; 6nstitute of Informatics, Slovak Academy of Sciences (IISAS), Dúbravská cesta 9, Bratislava, 84507, Slovakia; 7Instytut Chemii Bioorganicznej Polskiej Akademii Nauk - PSNC, Poland ID: 121
/ Poster No. # 9: 049
Modalities: Graphs, Multimodal Data Methods: Graph Neural Networks Application Domain: Health Multimodal graph-based fusion learning for Alzheimer’s disease prediction 1Modular High-Performance Computing and Artificial Intelligence, German Center for Neurodegenerative Diseases (DZNE), Bonn, North Rhine-Westphalia, Germany; 2Systems Medicine, German Center for Neurodegenerative Diseases (DZNE), Bonn, North Rhine-Westphalia, Germany; 3PRECISE Platform for Genomics and Epigenomics at DZNE and University of Bonn, Bonn, North Rhine-Westphalia, Germany; 4Life and Medical Sciences (LIMES) Institute, Genomics & Immunoregulation, Bonn, North Rhine-Westphalia, Germany ID: 178
/ Poster No. # 9: 050
Modalities: Text Methods: Generative Models Application Domain: Information Can Large Language Models Grade Short Free-Text Exam Answers? An Evaluation on Real German Student Data Forschungszentrum Jülich, Germany ID: 202
/ Poster No. # 9: 051
Modalities: Time Series Methods: Other Application Domain: Health Pseudo-Anomaly Augmented One-Class Autoencoder and Supervised Deep Learning for Murine ECG Anomaly Detection 1Helmholtz AI, Helmholtz Center Munich, Germany; 2Mary Lyon Centre at MRC Harwell, Harwell Campus, United Kingdom; 3Integrative Physiology/Advanced Technology Cores, Baylor College of Medicine, Texas, United States; 4Institute of Experimental Genetics, Helmholtz Center Munich, German Research Center for Environmental Health, Germany ID: 349
/ Poster No. # 9: 052
Modalities: Text Methods: Other Application Domain: Energy, Earth & Environment A Physics-Based Machine Learning Workflow for Urban-Scale Residential Heat-Demand Estimation Chair of Methods for Model-based Development in Computational Engineering, RWTH Aachen University, Germany ID: 305
/ Poster No. # 9: 053
Modalities: Multimodal Data, Tabular Data, Other Methods: Probabilistic Methods Application Domain: Health scDigital-Karyotypes: a genomic representation to learn shared variants 1Max Delbrück Center for Molecular Medicine, Berlin, Germany; 2Helmholtz AI, Helmholtz Munich, Germany; 3Department of Nephrology and Hypertension, Hannover Medical School, Hannover, Germany; 4Charité – Universitätsmedizin Berlin, Germany ID: 139
/ Poster No. # 9: 054
Modalities: Image Methods: Foundation Models, Probabilistic Methods, Other Application Domain: Health Semi-Supervised Federated Learning for Medical Image Classification Ozyegin University, Turkey (Türkiye) ID: 181
/ Poster No. # 9: 055
Modalities: Time Series Methods: Other Application Domain: Health ECG Template-Based Deep Learning for Amplified P-wave Duration Prediction 1Karlsruhe Institute of Technology, Germany; 2University Heart Center Freiburg - Bad Krozingen, Germany; 3Lucerne Cantonal Hospital, Switzerland; 4Justus Liebig University Giessen, Germany ID: 155
/ Poster No. # 9: 056
Modalities: Simulation Data Methods: Other Application Domain: Aeronautics, Space & Transport Reduction of Outflow Boundary Influence on Aerodynamic Performance Using Neural Networks 1Jülich Supercomputing Centre (JSC), Forschungszentrum Jülich, Wilhelm-Johnen-Straße 52428 Jülich, Germany; 2School of Mechanical and Mining Engineering, The University of Queensland, St Lucia 4072, Queensland, Australia; 3Chair of Fluid Mechanics, University of Siegen, Paul-Bonatz-Straße 9-11, Siegen-Weidenau 57076, Germany; 4Institute of Technology, Resource and Energy-efficient Engineering (TREE), Bonn-Rhein-Sieg University of Applied Sciences, Grantham-Allee 20, Sankt Augustin 53757, Germany; 5Fraunhofer Institute for Algorithms and Scientific Computing (SCAI), Schloss Birlinghoven, Sankt Augustin 53754, Germany; 6Institute of Aeronautics and Applied Mechanics, Warsaw University of Technology, Nowowiejska 24, Warszawa 00-665, Poland ID: 162
/ Poster No. # 9: 057
Modalities: Tabular Data, Text, Other Methods: Agentic AI Application Domain: Health Rolv.io - Multi-Agent AI For Healthcare And Research 1Max Delbrück Center (MDC), Germany; 2Hacettepe University, Turkey ID: 264
/ Poster No. # 9: 058
Modalities: Tabular Data Methods: Other Application Domain: Energy Understanding Green Technology Adoption in Germany: A Cluster-Based Analysis of Household-Scale Photovoltaics and Battery Electric Vehicles 1Institute of Climate and Energy Systems: Energy Systems Engineering (ICE-1), Forschungszentrum Jülich, Germany; 2Helmholtz AI, Helmholtz Munich, Germany; 3Institute of Climate and Energy Systems: Jülich Systems Analysis (ICE-2), Forschungszentrum Jülich, Germany; 4School of Business and Economics, RWTH Aachen University, Germany; 5Institute for Theoretical Physics, University of Cologne, Germany ID: 269
/ Poster No. # 9: 059
Modalities: Image Methods: Foundation Models Application Domain: Core Machine Learning From Learned Representation To Localization: Towards Semantic Integration Of Microscopic Data In The Human Brain 1Institute of Neuroscience and Medicine (INM-1), Research Centre Jülich, Germany; 2Helmholtz AI, Research Centre Jülich, Germany; 3Computer Vision, Institute for Computational Visualistics, University of Koblenz, Germany; 4C. & O. Vogt Institute of Brain Research, University Hospital Düsseldorf, Germany; 5Department of Physics, University of Wuppertal, Germany ID: 1375
/ Poster No. # 9: 060
Modalities: Image Methods: Foundation Models, Other Application Domain: Earth & Environment AI-assisted Classification of Polar Phytoplankton’s Functional Biodiversity using Multispectral Imaging Flow Cytometry 1Alfred Wegener Institute, Germany; 2University of Bremen, Germany ID: 293
/ Poster No. # 9: 062
Modalities: Image Methods: Uncertainty Quantification Application Domain: Health LEVERAGING MULTI-RATER ANNOTATIONS TO CALIBRATE OBJECT DETECTORS IN MICROSOPY IMAGING Helmholtz AI, Helmholtz Zentrum Muenchen, Germany ID: 276
/ Poster No. # 9: 063
Modalities: Image, Text Methods: Foundation Models, Generative Models Application Domain: Core Machine Learning, Health CytoDiff: AI-Driven Cytomorphology Image Synthesis for Medical Diagnostics Helmholtz Munich, Germany ID: 143
/ Poster No. # 9: 064
Modalities: Time Series Methods: Probabilistic Methods Application Domain: Earth & Environment DeepElbe: Multi-Horizon Oxygen Forecasting and Hypoxia Early Warning Using Statistical and Deep Learning Methods Helmholtz-Zentrum Hereon, Germany ID: 395
/ Poster No. # 9: 065
Cancer-label-free Tumor Segmentation in Whole-mouse Light Sheet Microscopy Images Using Vision-language Models 1Institute for Stroke and Dementia Research, Klinikum der Universität München, Ludwig-Maximilians University Munich, Munich, Germany; 2Institute for Intelligent Biotechnologies (iBIO), Helmholtz Center Munich, Neuherberg, Germany; 3Faculty of Medicine, Ludwig-Maximilians University Munich, Munich, Germany; 4Munich Cluster for Systems Neurology (SyNergy), Munich, Germany; 5Munich Medical Research School (MMRS), Munich, Germany; 6Graduate School of Neuroscience (GSN), Munich, Germany; 7Deep Piction GmbH, Munich, Germany; 8School of Medicine, Koç University, İstanbul, Turkey ID: 186
/ Poster No. # 9: 066
Modalities: Graphs, Image, Multimodal Data, Simulation Data, Text Methods: Generative Models, Physics-informed Machine Learning Application Domain: Energy, Matter Physics-Informed Reconstruction of Electron-Bunch Shapes from CTR Spectra using an Untrained Neural Network 1Helmholtz-Zentrum Dresden-Rossendorf, Germany; 2Technische Universität Dresden, Germany; 3Center for Advanced Systems Understanding, Germany; 4Technische Universität Chemnitz, Germany ID: 308
/ Poster No. # 9: 067
Modalities: Text Methods: Generative Models Application Domain: Core Machine Learning Modeling Long Contexts with Hierarchical Compression Transformers Forschungszentrum Jülich, Germany ID: 340
/ Poster No. # 9: 068
Modalities: Simulation Data Methods: Generative Models Application Domain: Earth & Environment Towards high resolution air quality forecasts for Europe using WGAN-based statistical downscaling with EURAD-IM data 1Forschungszentrum Jülich, Germany; 2Quarda Energy GmbH, Germany ID: 267
/ Poster No. # 9: 069
Modalities: Tabular Data Methods: Other Application Domain: Health Data-Driven Identification of Possible Transmission Sites for Carbapenem-Resistant Gram-Negative Bacteria 1Helmholtz AI, Helmholtz Munich, Germany; 2Hessian State Office for Health and Care (HLfGP), Germany; 3Helmholtz Centre for Infection Research, Germany ID: 336
/ Poster No. # 9: 070
Modalities: Multimodal Data Methods: Other Application Domain: Information Topological Data Analysis of Single-Cell Multimodal Data 1Biomedical Center (BMC), Physiological Chemistry, Faculty of Medicine, LMU Munich, Munich, Planegg-Martinsried, Germany; 2TUM, Germany; 3Institute of Computational Biology, Computational Health Center, Helmholtz Zentrum München, German Research Center for Environmental Health, Neuherberg, Germany; 4Hospital del Mar Research Institute (HMRIB), Barcelona, Spain ID: 383
/ Poster No. # 9: 071
Are Soft Labels Worth the Effort? Helmholtz, Germany ID: 237
/ Poster No. # 9: 072
Modalities: Tabular Data Methods: Agentic AI Application Domain: Information Building a Local AI Agent Assisting with Sensor Observation to Archives Framework (O2A) 1Alfred-Wegener-Institute for Polar and Marine Research, Data Science, Bremerhaven, Bremen, Germany; 2Alfred-Wegener-Institute for Polar and Marine Research, Software Engineering, Bremerhaven, Bremen, Germany ID: 142
/ Poster No. # 9: 073
Modalities: Text Methods: Agentic AI Application Domain: Earth & Environment SEA2LAND Navigator-GPT 1Hereon, Germany; 2HELCOM; 3GERICS; 4Tallinn University ID: 359
/ Poster No. # 9: 074
Modalities: Multimodal Data Methods: Foundation Models, Generative Models Application Domain: Core Machine Learning, Health Messanger RNA Untranslated Region Design through Transformers 1Helmholtz Munich, Institute of Computational Biology, Germany; 2Faculty of Mathematics, Informatics and Mechanics, University of Warsaw ID: 210
/ Poster No. # 9: 075
Modalities: Image, Multimodal Data, Text Methods: Agentic AI, Foundation Models Application Domain: Core Machine Learning, Health Semantic Equilibrium Decoding for Open-Ended Medical Visual Question Answering 1FAU Erlangen-Nürnberg, Germany; 2Imperial College London, UK; 3UCL, UK ID: 310
/ Poster No. # 9: 076
Modalities: Tabular Data Methods: Other Application Domain: Health Single Cell PBMC Analysis Shows Cell Type Specific Changes On Inflammation In Both Depressed And Non-Depressed Individuals 1Helmholtz AI, Helmholtz Zentrum München, Germany; 2Max Planck Institute of Psychiatry, München, Germany ID: 334
/ Poster No. # 9: 077
Modalities: Other Methods: Other Application Domain: Aeronautics, Space & Transport, Earth & Environment Advancing Sea Ice Surface Classification by Self-Supervised Contrastive Learning for Radar Altimetry 1Section for Sea Ice Physics, Alfred Wegener Institute, Germany; 2Earth Observation Center, German Aerospace Center, Germany; 3Institute of Software Technology, German Aerospace Center, Germany; 4Earth and Environmental Engineering, Columbia University, USA; 5Center for Industrial Mathematics, Bremen University, Germany ID: 316
/ Poster No. # 9: 078
Modalities: Simulation Data Methods: Physics-informed Machine Learning Application Domain: Matter Analytic model-informed representations for sample-efficient learning of scattering patterns Forschungszentrum Jülich GmbH, Germany ID: 138
/ Poster No. # 9: 079
Modalities: Tabular Data Methods: Physics-informed Machine Learning Application Domain: Information Quantum Kernel Alignment for Clinical Data Classification 1German Cancer Research Center, Germany; 2Heidelberg University; 3Fraunhofer IKS, LMU Munich ID: 232
/ Poster No. # 9: 080
Modalities: Tabular Data, Text, Other Methods: Agentic AI, Physics-informed Machine Learning, Other Application Domain: Core Machine Learning, Earth & Environment Syxplain: An Agentic Framework for Explainable Scientific Hypothesis Generation from Data 1HUN-REN, Hungary; 2Obuda University, Hungary ID: 357
/ Poster No. # 9: 081
Modalities: Text Methods: Agentic AI Application Domain: Matter Jülich Neutron Agent (JüNA): Vitess AI Agent Jülich Centre for Neutron Science (JCNS), FZ Jülich, Germany ID: 134
/ Poster No. # 9: 082
Modalities: Simulation Data Methods: Physics-informed Machine Learning Application Domain: Matter NEOPIC: HPC Strategies for FNO-DSE in Kinetic Plasma Simulation 1Forschungszentrum Jülich, Germany; 2Helmholtz Zentrum Dresden-Rossendorf, Germany ID: 292
/ Poster No. # 9: 083
Modalities: Graphs, Image, Multimodal Data, Simulation Data, Text, Time Series Methods: Foundation Models Application Domain: Core Machine Learning, Energy Diagnosing and Fixing Training Bottlenecks: A Case Study of a Foundation Model 1Jülich Supercomputing Centre, Forschungszentrum Jülich, 52428 Jülich, Germany; 2ECMWF, 53175 Bonn, Germany ID: 141
/ Poster No. # 9: 084
Modalities: Tabular Data Methods: Foundation Models, Uncertainty Quantification Application Domain: Health Advancing Explainability as a Differentiator in Machine Learning prediction models for Alzheimer’s Disease 1Modular HPC and AI, German Center for Neurodegenerative Diseases (DZNE), Bonn, Germany; 2Systems Medicine, German Center for Neurodegenerative Diseases (DZNE) e.V., 53127 Bonn, Germany; 3PRECISE Platform for Genomics and Epigenomics at DZNE and University of Bonn, 53127 Bonn, Germany ID: 212
/ Poster No. # 9: 085
Modalities: Video Methods: Physics-informed Machine Learning, Other Application Domain: Aeronautics, Space & Transport Improving Pedestrian Detection through Temporal Movement Estimations German Aerospace Center (DLR), Germany ID: 338
/ Poster No. # 9: 086
Modalities: Image, Text Methods: Other Application Domain: Core Machine Learning Neural Networks as Varieties Forschungszentrum Jülich, Germany ID: 176
/ Poster No. # 9: 087
Modalities: Multimodal Data, Simulation Data, Time Series Methods: Physics-informed Machine Learning Application Domain: Energy, Earth & Environment Surrogate Models for Severe Nuclear Accidents 1Helmholtz AI, Germany; 2Karlsruhe Institute of Technology, Germany ID: 270
/ Poster No. # 9: 088
Modalities: Image, Multimodal Data, Text, Video Methods: Foundation Models, Generative Models Application Domain: Core Machine Learning Training-free Uncertainty Guidance for Complex Visual Tasks with MLLMs 1Helmholtz Zentrum München, Germany; 2Technical University of Munich, Germany; 3Munich Center for Machine Learning, Germany; 4Télécom Paris, France; 5Google, Switzerland ID: 192
/ Poster No. # 9: 089
Modalities: Other Methods: Probabilistic Methods, Other Application Domain: Earth & Environment Helixer – a tool for ab initio gene calling combining Deep Learning and a Hidden Markov Model 1Forschungszentrum Jülich, Germany; 2Heinrich Heine Univsersität Düsseldorf, Germany; 3Cluster of Excellence on Plant Sciences (CEPLAS), Germany; 4Recursion, Valence Labs, Canada ID: 287
/ Poster No. # 9: 090
Modalities: Simulation Data, Tabular Data Methods: Probabilistic Methods, Uncertainty Quantification Application Domain: Core Machine Learning metabeta - A fast neural model for Bayesian mixed-effects regression Human-Centered AI, Helmholtz Munich, Germany ID: 236
/ Poster No. # 9: 091
Modalities: Other Methods: Probabilistic Methods Application Domain: Health Meta-Learning for Phenotypic Drug Discovery with Transformer Neural Processes 1Institute of AI for Health, Helmholtz Munich; 2School of Computation, Information and Technology, Technical University of Munich; 3Helmholtz AI; 4Munich Center for Machine Learning; 5Faculty of Mathematics, Informatics and Mechanics, University of Warsaw; 6Department of Computer Science and Artificial Intelligence, University of Technology Nuremberg ID: 266
/ Poster No. # 9: 092
Modalities: Graphs, Multimodal Data, Text Methods: Agentic AI, Generative Models Application Domain: Energy Agentic Data Extraction and Dynamic Knowledge Graph Construction for Hydrogen Technology Research 1Theory and Computation of Energy Materials (IET-3), Institute of Energy Technologies, Forschungszentrum Jülich GmbH, 52425 Jülich, Germany; 2Centre for Advanced Simulation and Analytics (CASA), Simulation and Data Science Lab for Energy Materials (SDL-EM), Forschungszentrum Jülich GmbH, 52425 Jülich, Germany; 3Chair for Theory and Computation of Energy Materials, Faculty of Georesources and Materials Engineering, RWTH Aachen University, Germany ID: 385
/ Poster No. # 9: 093
Diagnosing Shortcuts in Traffic Video Question Answering Datasets 1Technical University of Munich, Germany; 2Helmholtz Zentrum München, Germany; 3Munich Center for Machine Learning, Germany ID: 172
/ Poster No. # 9: 094
Modalities: Image, Other Methods: Other Application Domain: Core Machine Learning, Health DREAMS: Preserving both Local and Global Structure in Dimensionality Reduction 1Hertie Institute for AI in Brain Health, Germany; 2University of Tübingen, Germany ID: 227
/ Poster No. # 9: 095
Modalities: Multimodal Data, Simulation Data, Tabular Data, Other Methods: Probabilistic Methods Application Domain: Core Machine Learning, Health Factor dependencies and their associations with covariates in multimodal factor analysis 1University of Warsaw, Poland; 2Helmholtz Center Munich, Germany ID: 352
/ Poster No. # 9: 097
Modalities: Image, Multimodal Data Methods: Generative Models, Other Application Domain: Health Discovering Interpretable Visual Concepts in Monkey Visual Cortex Using Sparse Autoencoders 1Institute of Computational Biology, Computational Health Center, Helmholtz Munich, Germany; 2Department of Experimental Psychology, University of Oxford, UK ID: 196
/ Poster No. # 9: 098
Modalities: Image, Time Series Methods: Foundation Models, Other Application Domain: Earth & Environment Incremental learning for continuous forest disturbance mapping using AlphaEarth embeddings GFZ Helmholtz Centre for Geosciences, Germany ID: 194
/ Poster No. # 9: 099
Modalities: Multimodal Data, Tabular Data, Time Series Methods: Other Application Domain: Health Multimodal Machine Learning for Postoperative Atrial Fibrillation Prediction 1Karlsruhe Institute of Technology, Germany; 2Medical Faculty Mannheim, Germany; 3University Hospital Heidelberg, Germany ID: 256
/ Poster No. # 9: 100
Modalities: Graphs, Multimodal Data, Tabular Data, Text, Time Series Methods: Agentic AI, Foundation Models, Generative Models Application Domain: Earth & Environment FrevaGPT: An LLM-Powered Assistant for Reproducible Climate Analysis 1GFZ Helmholtz Centre for Geosciences, Germany; 2German Climate Computing Center DKRZ, Germany; 3Universität Hamburg, Germany; 4Climate Service Center Germany (GERICS), Germany ID: 272
/ Poster No. # 9: 101
Modalities: Image Methods: Foundation Models Application Domain: Core Machine Learning, Health Mind The Gap: Continuous Magnification Sampling For Pathology Foundation Models 1Technical University Berlin, Germany; 2Berlin Institute for the Foundations of Learning and Data; 3Aignostics Gmbh; 4Charité, Universitätsmedizin Berlin ID: 153
/ Poster No. # 9: 102
Modalities: Text Methods: Agentic AI Application Domain: Information Grounding Large Language Models in Formal Solvers via Unified Planning: Towards High-Accuracy Natural Language Planning Hybrid Methods in Artificial Intelligence and Machine Learning, University of Rostock, Germany ID: 171
/ Poster No. # 9: 103
Modalities: Image, Multimodal Data, Video Methods: Other Application Domain: Core Machine Learning MLArray: Unifying Machine Learning-Optimized Array Storage and Imaging Metadata German Cancer Research Center (DKFZ), Germany ID: 379
/ Poster No. # 9: 104
Post-training makes large language models less human-like Helmholtz Munich, Germany ID: 180
/ Poster No. # 9: 106
Modalities: Image, Multimodal Data, Text Methods: Foundation Models, Generative Models Application Domain: Aeronautics, Space & Transport, Earth & Environment A Domain-Specialized Multimodal Foundation Model for Interpreting Sentinel-2 True Color Imagery in Flood Disaster Response DLR, Germany ID: 363
/ Poster No. # 9: 107
Modalities: Image Methods: Foundation Models Application Domain: Health A Hematology Foundation Model Enables Blood Cancer Screening from Peripheral Blood Smears 1Helmholtz Munich, Germany; 2Munich Leukemia Laboratory ID: 313
/ Poster No. # 9: 108
Modalities: Simulation Data Methods: Generative Models, Graph Neural Networks, Physics-informed Machine Learning, Other Application Domain: Energy, Matter Surrogate Modeling of Scale-Bridging Dislocation-Driven Plasticity through Latent Dynamics and Microstructure Descriptors 1Institute for Advanced Simulations – Materials Data Science and Informatics (IAS-9) Forschungszentrum Jülich GmbH 52425 Jülich, Germany; 2Chair of Materials Data Science and Materials Informatics, Faculty 5 Georesources and Materials Engineering, RWTH Aachen University 52056 Aachen, Germany ID: 148
/ Poster No. # 9: 110
Modalities: Multimodal Data, Simulation Data, Time Series Methods: Foundation Models, Generative Models, Uncertainty Quantification Application Domain: Earth & Environment NVIDIA Earth-2: Open Models, Accessible Hardware, and the Future of AI for Weather and Climate NVIDIA Corporation ID: 156
/ Poster No. # 9: 111
Modalities: Multimodal Data, Simulation Data, Tabular Data Methods: Physics-informed Machine Learning, Probabilistic Methods Application Domain: Health Amortized Bayesian Multilevel Model of Bacterial Cell-Cycle with BayesFlow 1Helmholtz Institute for RNA-based Infection Research (HIRI), Helmholtz Centre for Infection Research (HZI), Würzburg, Germany; 2Department of Biology, University of Toronto, Mississauga, ON, Canada; 3Department of Cell and Systems Biology, University of Toronto, Toronto, ON, Canada ID: 320
/ Poster No. # 9: 112
Modalities: Image Methods: Other Application Domain: Information, Matter Exploring the Effectiveness of Pretraining Task Alignment for Self-Supervised Learning in a Low-Data Regime 1Institute for Advanced Simulations — Materials Data Science and Informatics (IAS-9), Forschungszentrum Jülich GmbH, 52425 Jülich, Germany; 2Chair of Materials Data Science and Materials Informatics, Faculty 5 — Georesources and Materials Engineering, RWTH Aachen University, 52056 Aachen, Germany ID: 190
/ Poster No. # 9: 113
Modalities: Image Methods: Probabilistic Methods, Uncertainty Quantification Application Domain: Health, Matter Taming Multimodal Posteriors over Rotations: A Geometric Approach to Bayesian Alignment in Cryo-EM and Beyond 1Helmholtz Institute Jena, Germany; 2GSI Helmholtzzentrum für Schwerionenforschung, Germany; 3Friedrich Schiller University Jena, Germany; 4Fraunhofer Institute for Applied Optics and Precision Engineering, Germany; 5University Hospital Jena, Germany; 6Max Plank Institute for Multidisciplinary Sciences, Germany ID: 125
/ Poster No. # 9: 114
Modalities: Other Methods: Generative Models Application Domain: Core Machine Learning, Health Sample Efficient Generative Molecular Optimization with Joint Self-Improvement 1Helmholtz Munich, Germany; 2Technical University of Munich, TUM School of Computation, Information and Technology; 3Technical University of Munich, Campus Straubing for Biotechnology and Sustainability; 4University of Applied Sciences Weihenstephan Triesdorf; 5Faculty of Mathematics, Informatics and Mechanics, University of Warsaw ID: 324
/ Poster No. # 9: 115
Modalities: Multimodal Data, Text Methods: Agentic AI, Generative Models Application Domain: Health Mind the Gap – From expectation to reality in agentic AI 1Helmholtz Munich, Germany; 2German Center for Diabetes Research, Germany ID: 1401
/ Poster No. # 9: 116
Modalities: Multimodal Data, Text, Other Methods: Foundation Models Application Domain: Health Genomic language model based prediction of transcription start sites in twelve yeast species 1Computation Molecular Medicine, School of Computation, Information and Technology, TU Munich, Germany; 2Computational Health Center, Helmholtz Center Munich, Germany | ||