@inproceedings{7002, abstract = {Selecting an appropriate semantic segmentation model for a given application domain remains a challenging and time-consuming task for practitioners and researchers. This paper presents an interactive, web-based platform that enables side-by-side visual comparison of multiple neural network segmentation models applied to identical images. The system integrates three transformer-based segmentation models: a face-parsing network producing 19 semantic classes, a SegFormer-B3 clothing segmentation model with 18 classes, and a Mask2Former model for general-purpose scene segmentation spanning 150 ADE20K categories. Key contributions include side-by-side evaluation of model outputs across multiple architectures and image categories, with real-time segment highlighting and a scalable inference caching system that enables model comparisons without requiring repeated graphics processing unit (GPU) computation. The platform organizes a curated dataset of images under a hierarchical category taxonomy, supporting structured evaluation across demographic and contextual variables. As a practical use case, the system is applied within the ADRIAN project to assist in verifying identity consistency across images through segmentation-based analysis. The platform thus contributes a specialized artificial intelligence (AI) tool for systematic segmentation and object detection model evaluation within media analysis pipelines, where selecting appropriate models is a recurring challenge across tasks from identity verification to content moderation. It is available under https://github.com/vika-v-v/neural-networks-for-image-segmentation and designed to lower the barrier for comparative model evaluation in applied computer vision workflows.}, author = {Vovchenko, Viktoriia and Schultenkämper, Sergej and Bäumer, Frederik}, booktitle = {2026 The Second International Conference on AI-based Media Innovation}, keywords = {segmentation, model comparison, face parsing}, location = {Nizza, Frankreich}, publisher = {IARIA}, title = {{A Web-Based Platform for Interactive Comparison of Neural Network Image Segmentation Models}}, year = {2026}, } @inproceedings{6998, abstract = {Instruction-guided image editing models have made photorealistic facial manipulation accessible through natural language prompts, substantially lowering the barrier to misuse in the cybersecurity landscape. These systems facilitate attacks such as identity theft, phishing, misinformation, and social engineering, while also challenging existing forensic methods that were not designed for prompt-driven edits. However, current facial forensics datasets do not cover this manipulation paradigm. We present ADRIAN InstructFace-Edit, a dataset of 36,000 edited facial image pairs at 1024x1024 resolution generated by four state-of-the-art instruction-guided image editing models (FLUX.1-Kontext-dev, FLUX.2-dev, Qwen-Image-Edit-2511, and Step1X-Edit-v1.2) across six semantically distinct edit types, of which 18,768 are annotated as high quality. The full dataset, including all pairs with quality labels, is made publicly available. To improve data quality, we design a multi-stage filtering pipeline that integrates structural, semantic, and multimodal large language model–based quality checks, with thresholds derived empirically from a human-annotated subset of 1,500 images. Because open-weight vision-language models perform poorly on VIEScore evaluation out of the box, we fine-tune a LoRA adapter for Qwen3-VL-32B-Instruct that improves within +/-1 Semantic Consistency accuracy from near-random to 73.2%. We will publicly release the dataset, generation pipeline, and LoRA adapter to support future research on training, evaluating, and stress-testing detectors for instruction-guided facial manipulation.}, author = {Vovchenko, Viktoriia and Barberi, Vincenzo and Schultenkämper, Sergej and Bäumer, Frederik}, booktitle = {2026 The First International Conference on Security and Cybersecurity in the AI and Digital Context}, keywords = {Deepfakes, Image forensics, Diffusion models, Computer vision}, location = {Porto, Portugal}, title = {{ADRIAN InstructFace-Edit - Towards Robust Detection of AI-Manipulated Face Images}}, year = {2026}, } @inproceedings{7001, author = {Schultenkämper, Sergej and Vovchenko, Viktoriia and Bäumer, Frederik}, booktitle = {2025 19th International Conference on Semantic Computing (ICSC)}, location = {Laguna Hills, CA, USA}, pages = {259--264}, publisher = {IEEE}, title = {{Enhancing Social Media Summarization with Instruction Tuned Large Language Models and Adaptive Summary Ranking}}, doi = {10.1109/ICSC64641.2025.00045}, year = {2025}, } @inproceedings{6182, author = {Bäumer, Frederik and Brandt-Pook, Hans and Huppert, Christian and Šir, Yves and Stahlkopf, Alexander and Sang, Florian and Ullrich, Nadja}, issn = {2943-3509}, location = {Bielefeld}, number = {2025}, title = {{Für eine inklusive digitale Zukunft: Ableismus-sensible Chatbots gestalten}}, doi = {10.60802/sidas.2025.2}, volume = {2}, year = {2025}, } @inproceedings{6084, author = {Bourdick, Raphael and Bäumer, Frederik}, booktitle = {Jahrestagung der Gesellschaft für Informatik}, location = {Potsdam}, title = {{Beyond Q&A: Entwicklung eines agentenbasierten Retrieval-Augmented-Generation-Systems bei der imos AG}}, year = {2025}, } @inproceedings{5406, author = {Schultenkämper, Sergej and Bäumer, Frederik}, booktitle = {The 13th IEEE International Workshop on Semantic Computing for Social Networking: from user information to social knowledge and ethical AI}, location = {Laguna Hills, CA}, title = {{Multimodal Analysis of Modern Phishing Attacks: Towards an Evolving Dataset of Real-World Data}}, doi = {10.1109/ICSC64641.2025.00046}, year = {2025}, } @inproceedings{5407, author = {Schultenkämper, Sergej and Vovchenko, Viktoriia and Bäumer, Frederik}, booktitle = {The 13th IEEE International Workshop on Semantic Computing for Social Networking: from user information to social knowledge and ethical AI}, location = {Laguna Hills, CA}, title = {{Enhancing Social Media Summarization with Instruction Tuned Large Language Models and Adaptive Summary Ranking}}, doi = {10.1109/ICSC64641.2025.00045}, year = {2025}, } @article{5148, author = {Bäumer, Frederik Simon and Schultenkämper, Sergej and Geierhos, Michaela and Lee, Yeong Su}, issn = {2661-8907}, journal = {SN Computer Science}, number = {8}, publisher = {Springer Science and Business Media LLC}, title = {{Mirroring Privacy Risks with Digital Twins: When Pieces of Personal Data Suddenly Fit Together}}, doi = {10.1007/s42979-024-03413-z}, volume = {5}, year = {2024}, } @inproceedings{5049, author = {Bäumer, Frederik S. and Brandt-Pook, Hans and Matutat, André and Maoro, Falk and Pelkmann, David and Schultenkämper, Sergej}, booktitle = {INFORMATIK 2024 Lock-in or log out? Wie digitale Souveränität gelingt}, editor = {Klein, Maike and Krupka, Daniel and Winter, Cornelia and Gergeleit, Martin and Martin, Ludger}, location = {Wiesbaden}, publisher = {Köllen Druck+Verlag GmbH}, title = {{Lektionen und Anwendungsfälle aus der Implementierung von Retrieval-Augmented-Generation-Systemen}}, volume = {352}, year = {2024}, } @inproceedings{4453, author = {Schultenkämper, Sergej and Bäumer, Frederik S.}, booktitle = {Communications in Computer and Information Science}, location = {Maastricht, Niederlande}, title = {{Structured Knowledge Extraction for Digital Twins: Leveraging LLMs to Analyze Tweets}}, year = {2024}, } @article{4454, author = {Schultenkämper, Sergej and Bäumer, Frederik S.}, issn = {1942-2636}, journal = {International Journal On Advances in Security}, number = {1+2}, pages = {1--10}, publisher = {IARIA XPS Press}, title = {{Pixels versus Privacy: Leveraging Vision-Language Models for Sensitive Information Extraction}}, volume = {17}, year = {2024}, } @inproceedings{3297, author = {Schultenkämper, Sergej and Bäumer, Frederik}, booktitle = {The Fifteenth International Conference on Pervasive Patterns and Applications (PATTERNS 2023). Special Track DaMIA: Data Mining in Industrial Applications of Digital Twins.}, issn = {2308-3557}, pages = {2--5}, title = {{Protecting Your Online Privacy: Insights on Digital Twins and Threat Detection}}, year = {2023}, } @inproceedings{3629, author = {Sirvend, Bastian and Stahlkopf, Alexander and Bäumer, Frederik}, booktitle = {INFORMATIK 2023}, editor = {Klein, Maike and Krupka, Daniel and Winter, Cornelia and Wohlgemuth, Volker}, isbn = {978-3-88579-731-9}, issn = {1617-5468}, location = {Berlin}, publisher = {Köllen Druck+Verlag GmbH}, title = {{Synergien zwischen KI und menschlicher Expertise: Online-Sichtbarkeitsoptimierung durch Text-Transformation}}, doi = {10.57720/3629}, volume = {337}, year = {2023}, } @inproceedings{3606, author = {Schultenkämper, Sergej and Bäumer, Frederik}, booktitle = {Information and Software Technologies. 29th International Conference, ICIST 2023, Kaunas, Lithuania, October 12–14, 2023, Proceedings}, location = {Kaunas, Lithuania}, publisher = {Springer International Publishing}, title = {{Privacy Risks in German Patient Forums: A NER-based Approach to Enrich Digital Twins}}, doi = {10.1007/978-3-031-48981-5_9}, year = {2023}, } @inproceedings{3612, author = {Schultenkämper, Sergej and Bäumer, Frederik and Bellgrau, Benjamin and Lee, Yeong Su and Geierhos, Michaela}, booktitle = {27th International Conference on Enterprise Design, Operations Computing (EDOC 2023): First Workshop on the Modelling and Implementation of Digital Twins for Complex Systems)}, location = {Groningen, Niederlande}, title = {{From Digital Tracks to Digital Twins: On the Path to Cross-Platform Profile Linking}}, doi = {10.1007/978-3-031-54712-6_10}, year = {2023}, } @inproceedings{3614, author = {Schultenkämper, Sergej and Bäumer, Frederik and Geierhos, Michaela and Lee, Yeong Su}, booktitle = {Proceedings of the Thirteenth International Conference on Social Media Technologies, Communication, and Informatics (SOTICS 2023)}, isbn = {978-1-68558-103-9}, issn = {2326-9294}, location = {Valencia, Spanien}, title = {{From Unstructured Data to Digital Twins: From Tweets to Structured Knowledge}}, year = {2023}, } @inproceedings{3631, author = {Bäumer, Frederik and Brandt-Pook, Hans and Maoro, Falk and Schultenkämper, Sergej and Stecker, Benjamin}, booktitle = {INFORMATIK 2023}, editor = {Klein, Maike and Krupka, Daniel and Winter, Cornelia and Wohlgemuth, Volker}, isbn = {978-3-88579-731-9}, issn = {1617-5468}, location = {Berlin}, publisher = {Köllen Druck+Verlag GmbH}, title = {{Von der Theorie zur Praxis: Erfahrungen bei der akademischen Begleitung von KI-Projekten in KMUs}}, doi = {10.57720/3631}, volume = {337}, year = {2023}, } @inproceedings{3296, author = {Schultenkämper, Sergej and Bäumer, Frederik}, booktitle = {ICIW 2023 : The Eighteenth International Conference on Internet and Web Applications and Services}, isbn = {2308-3972}, issn = {2308-3972}, location = {Nizza}, pages = {2--6}, title = {{Looking for a Needle in a Haystack: How Can Vision-Language Understanding Help to Identify Privacy-Threatening Images on the Web}}, year = {2023}, } @inbook{3247, author = {Bäumer, Frederik and Chen, Wei-Fan and Geierhos, Michaela and Kersting, Joschka and Wachsmuth, Henning}, booktitle = {On-The-Fly Computing - Individualized IT-services in dynamic markets}, title = {{Dialogue-Based Requirement Compensation and Style-Adjusted Data-To-Text Generation}}, doi = {10.5281/zenodo.8068456}, year = {2023}, } @inproceedings{2716, author = {Denisov, Sergej and Bäumer, Frederik and Geierhos, Michaela}, booktitle = {Proceedings of the Fourteenth International Conference on Pervasive Patterns and Applications (PATTERNS 2022): Special Track AI-DRSWA: Maturing Artificial Intelligence - Data Science for Real-World Applications.}, title = {{Track Me If You Can: Insights into Profile Interlinking on Social Networks}}, year = {2022}, }