Simon Münker

Research Fellow at Trier University
Mail LinkedIn GitHub Google Scholar ORCID

I started my career as a media designer and front-end web developer (IHK), focusing on WordPress and reactive JavaScript while working on projects of various scales in finance, education, and health. During this time (2016), voice assistants and text-centric user interfaces began to rise. Thus, I decided to pivot my career and actively participate in the research and development of language processing techniques and generative text-based AI. I pursued an undergraduate in computational linguistics (2017) to complement my programming skills with a linguistic foundation. I continued my education with a master’s in natural language processing (2021), specializing in text analysis and generation with large language models. Afterward (2023), I joined the Department of Computational Linguistics at Trier University as a research fellow, focusing on language models as synthetic online social media users in the scope of an EU-funded research project.

Experience

Work

Oct. 2023 - Present | Research Fellow (Trier University): Implementing and maintaining Python-based modules for synthetic social media users through language models and the evaluation of discourse metrics for the EU-founded project Twin of Online Social Networks (TWON). Researching the ability of language models to mimic human behavior in conversation in a semantic and philosophical dimension. Managing graduate student research groups in applied computational linguistics subjects. Planning and conducting the lecture and exercise natural language processing.

Mar. 2022 - Sep. 2023 | Research Assistant (Trier University, Part-Time): Implemented and co-authored “InvBERT: Reconstructing Text from Contextualized Word Embeddings by inverting the BERT pipeline”, published in the Conference of Computational Literary Studies. Implemented and maintained a pipeline to scrape social media posts and match them with related news articles as part of the Social Sentiment in Times of Crises (SOSEC) project founded by the Alfred Landecker Foundation. Planned and conducted the exercise sessions for the lecture algorithmic methods: a Python introduction.

Apr. 2019 - Sep. 2021 | Course Assistant (University of Düsseldorf, Part-Time): Provided academic support for multiple advanced mathematics and computer science courses, including probability theory, machine learning, and formal language. Facilitated learning by developing supplementary teaching materials, conducting exercise sessions, and providing individualized student guidance. Assisted professors in grading assignments, preparing course content, and managing classroom logistics.

Feb. 2020 - Mar. 2020 | Intern (University of Düsseldorf): Conducted research on multiword-expression (MWE) identification by (re-)implementing and extending the Vine tool using PyTorch deep learning framework. Developed a natural language processing pipeline that improved the accuracy of MWE detection through deep learning techniques.

Oct. 2017 - Sep. 2019 | Frontend Web Developer (DSC-Medien, Part-Time): Developed responsive web applications using modern JavaScript frameworks, with a focus on React and Vue.js. Implemented responsive design principles and optimized web interfaces for multiple devices and browsers. Collaborated with design and backend teams to ensure seamless integration of frontend components and adherence to project specifications.

Aug. 2016 - Sep. 2017 | Frontend Web Developer (DSC-Medien): Created and maintained responsive web solutions using HTML5, CSS3, and JavaScript. Developed custom WordPress themes and plugins to meet client-specific requirements. Implemented SEO best practices and performance optimization techniques to improve website loading speeds and user experience.

Aug. 2013 - Jul. 2016 | Trainee (SIXWASNINE – Agentur für Kommunikation & Werbung): Gained comprehensive experience in communication and advertising, supporting creative team in developing marketing strategies and visual communication materials. Assisted in client presentations, brand identity design, and social media content creation. Developed skills in graphic design, copywriting, and project management within a dynamic creative agency environment.

Education

2021 - 2023 | Master of Science - MS, Natural Language Processing (Trier University): Grade: 1.2 (very good) | Developed expertise in Python-based NLP methodologies, language model interpretation, and deep learning techniques. Conducted in-depth research on semantic analysis and language model reconstruction. Active member of CoDiPho (student council Computational Linguistics & Digital Humanities).

2017 - 2021 | Bachelor of Arts - BA, Computational Linguistics (University of Düsseldorf): Grade: 1.4 (very good) | Developed comprehensive skills in mathematical foundations, machine learning, and natural language processing. Implemented advanced projects in computational linguistic downstream tasks and machine learning applications. Gained practical experience in research methodologies including presenting and academic writing.

2013 - 2016 | Mediengestalter Digital und Print, Graphic Design (IHK Mittlerer Niederrhein): Grade: 91/100 (good) | Acquired foundational skills in digital media design, web development, and visual communication. Developed expertise in frontend technologies, graphic design principles, and creative problem-solving approaches.

Honors & awards

Best Master’s Degree 2023, in Natural Language Processing Program Issued by Fach Computerlinguistik & Digital Humanities · Jan 2024

Best Bachelor’s Degree 2021, in Computational Linguistics Program Issued by FörderLinK e.V. - Förderkreis der Allgemeinen Linguistik an der Heinrich-Heine-Universität Düsseldorf · Jan 2022

Skills

Natural Language Processing, Large Language Models, Agent-based Modeling, API Development, Machine Learning, Communication & Presentations

Programming & Tools: Python (Poetry, Pydantic, Pandas, FastAPI, scikit-learn, Pytorch, Hugging Face), JavaScript (Astro, Svelte, Vue), R (Tidyverse)

Miscellaneous: Academic Research, University Teaching

Activities

Research

  • Kugler, K., Münker, S., Höhmann, J., Rettinger, A. (2024). InvBERT: Reconstructing Text from Contextualized Word Embeddings by inverting the BERT pipeline. Journal of Computational Literary Studies, 2(1), Article 1. https://doi.org/10.48694/jcls.3572
  • Münker, S. (2023). Can Large Language Models replace human annotation for text classification? A comparison to prompt-based approaches on German Twitter data [Poster]. Patterns Poster-Session, University Trier
  • Münker, S. (2024). On the authenticity of generated OSN Content: Advances in Agent-based LLM prompting for persuasive posts [Presentation]. Patterns Early Career Forum, University Trier
  • Heseltine, M., Münker, S., Stolwijk, S., Trilling, D., Oschatz, C. (2024). Generative User Content for Social Media Platforms: Comparing LLM Effectiveness and Approaches [Poster]. Etmaal 2024, Rotterdam
  • Münker, S. (2024). LLMs values towards digital Privacy: Exploring synthetic social media content from generated Agents [Poster]. Dagstuhl Seminar: Generative AI for Knowledge Engineering https://www.dagstuhl.de/24143
  • Sittar A., Münker S., Gucek A., Grobelnik M. (2024). Simulating debates on social media with digital twin – TWON [Poster]. European Data Science Day at KDD 2024, Barcelona
  • Münker, S., Kugler, K., Rettinger, A. (2024). Zero-shot prompt-based classification: topic labeling in times of foundation models in German Tweets. arXiv preprint arXiv:2406.18239
  • Münker, S. (2024). **Towards” Differential AI Psychology” and in-context Value-driven Statement Alignment with Moral Foundations Theory. arXiv preprint arXiv:2408.11415
  • Heseltine M., Münker S., Stolwijk S. (2024). Generative User Content for Social Media: LLM Effectiveness and Approaches. [Poster] American Political Science Association: Annual Meeting, Philadelphia

Teaching

  • Research Case Study: Aligning Language Models to the behavior of social media users [Student Research Group]. Winter Term 2023, Trier University
  • Algorithmische Methoden: Introduction to Python programming for text analysis [Exercise]. Winter Term 2023, Trier University
  • Natural Language Processing: Fundamentals of modern language processing with generative AI [Lecture & Exercise]. Summer Term 2024, Trier University
  • Anwendung der KI & CL - KI == Nutzer?: Critical Analysis of generative AI in social sciences [Guest Lecture]. Summer Term 2024, Trier University
  • Einführung in die Sprachwissenschaft: Fundamentals of contemporary (computational) linguistics [Lecture]. Winter Term 2024, Trier University

Projects/Packages

  • cltrier_{nlp, lib}: Common libraries for department internal (Trier University, Computational Linguistics) teaching and research [GitHub | GitHub]
  • cltrier_annotation: Simple web-based annotation & tool [GitHub | Demo]
  • TWON-{Agents, Ranker, Metrics}: API-based services for the EU-funded research project TWON (Twin of Online Social Networks) [GitHub | GitHub | GitHub]