Doctoral Researcher in the Educational Technologies at DIPF
Information Center for Education
Rostocker Straße 6, 60323 Frankfurt am Main

Researchgate ScholarTwitterLinkedIn

Telephone: please contact me by E-Mail
Email: s.gombert [at]

I started to work as a doctoral researcher at the Educational Technologies group, DIPF, in 2021. Before this, I studied Linguistic and Literary Computing at TU Darmstadt. I finished with a thesis on detecting semantic uncertainty in scientific writing using pre-trained transformer language models and conditional random fields. Besides this, I worked as a teaching assistant, software developer and junior research assistant in various contexts and was active in academic self-government. When I am not busy researching, I like producing electronic music, playing the drums and hiking.
  • Artificial Intelligence in Education
  • Natural Language Processing
  • Learning Analytics
  • Feedback Generation
  • Human-Computer Interaction
  • Discourse Semantics
  • Cognitive Linguistics
  • Master of Arts in Linguistic and Literary Computing, 2021, TU Darmstadt
  • Bachelor of Arts in Digital Philology and Computer Science, 2019, TU Darmstadt
  • 2014 – Developer @ Sanofi-Aventis Deutschland GmbH
  • 2015-2017 – Developer @ Abius GmbH
  • 2015-2020 – Co-founder @ mediacollective UG
  • 2017 – IT Administrator @ Student Council FB2, TU Darmstadt
  • 2016-2021 Teaching- and Junior Research Assistant @ Corpus- and Computational Linguistics, TU Darmstadt
  • 2020 Teaching Assistant @ Ubiquitous Knowledge Processing Lab, TU Darmstadt
  • 2020-2021 Developer @ AskAlbert / studiumdigitale, Goethe Universität Frankfurt
  • Fink, A., Gombert, S., Liu, T., & Frey, A., (2024, April). The integration of automated essay scoring models using hierarchical rater models. Paper presented at the Annual Meeting of the National Council on Measurement in Education (NCME), Philadelphia, USA.
  • Menzel, L., Jivet, I., Gombert, S., Drachsler, H., Giorgashvili, T., Schmitz, M. (2024). 2nd Workshop on Highly Informative Learning Analytic (HILA), Learning Analytics and Knowledge Conference 2024, Kyoto, Japan.
  • Fink, A., Gombert, S., & Frey, A. (2023). Using natural language processing to automate the assessment of student essay content. Paper presented at the Frontier Research in Educational Measurement – FREMO 2023, Oslo, Norway.
  • Fink, A., Gombert, S., & Frey, A. (2023). Automatisierte Inhaltsanalyse von Essays mittels Transformern [Automated analysis of essay content using transformer models]. Paper presented at the 16th meeting of the Methods and Evaluation Division of the DGPs, Constance, Germany.
  • Gombert, S., & Karademir, O. (2023). KI in der Bildung: Chancen und Risiken für die Gesellschaft. @ Common Purpose.
  • Gombert, S., & Kube, D. (2023). Virtueller Kaffeetalk zum Thema ChatGPT in der Bildung. @ Digitaltag 2023.
  • Gombert, S., & Menzel, L. (2023). Hyperchalk – How to Implement Self-hosted Whiteboard Tasks. @ Seventeenth EATEL Summer School on Technology Enhanced Learning.
  • Gombert, S., Menzel, L., & Di Mitri, D. (2023). Large Language Models for Feedback Generation. @ Seventeenth EATEL Summer School on Technology Enhanced Learning.
  • Karademir, O., Gombert, S., & Di Mitri, D. (2023). Teacher Dashboard Supported by AI-generated Feedback. @ Seventeenth EATEL Summer School on Technology Enhanced Learning.
  • Di Mitri, D., Schmitz, M., Jivet, I., Gombert, S., Ahmad, A., Drachsler, H. (2023). Hi, LA! – Highly Informative Learning Analytics. @ 13th International Learning Analytics and Knowledge Conference.
  • Fink, A., Frey, A., Liu, T., & Gombert, S. (2022). Nutzung von Natural Language Processing zur automatisierten Kodierung von Essays in digital gestützten Großveranstaltungen an Hochschulen. @ 52nd Conference of the German Psychological Society (DGPs).
  • Gombert, S. (2022). Workshop: Assessing constructed responses with explainable natural language processing. @ Sixteenth EATEL Summer School on Technology Enhanced Learning.
  • Gombert, S., Bhattacharya, S., Di Mitri, D., Drachsler, H. (2022). Interpretierbare maschinelle Sprachverarbeitung für die Identifikation konzeptionellen Wissens in energiephysikbezogenen deutschsprachigen Kurzantworten. @ Symposium: Komplexität nicht nur bewältigen, sondern zu Nutze machen: Log- und Textdaten im Educational Assessment. 9. Tagung der Gesellschaft für Empirische Bildungsforschung.