This is an outdated version published on 2024-02-29. Read the most recent version.

Speaker Attribution in German Parliamentary Debates with QLoRA-adapted Large Language Models

Authors

  • Tobias Bornheim Dept. of Medical Engineering and Technomathematics, FH Aachen University of Applied Sciences
  • Niklas Grieger Dept. of Medical Engineering and Technomathematics, Institute for Data-Driven Technologies, FH Aachen University of Applied Sciences and Department of Information and Computing Sciences, Utrecht University
  • Patrick Gustav Blaneck Dept. of Medical Engineering and Technomathematics, FH Aachen University of Applied Sciences
  • Stephan Bialonski Dept of Medical Engineering and Technomathematics, Institute for Data-Driven Technologies, FH Aachen University of Applied Sciences https://orcid.org/0000-0003-1150-8080

DOI:

https://doi.org/10.21248/jlcl.37.2024.244

Keywords:

large language models, German, speaker attribution, semantic role labeling

Abstract

The growing body of political texts opens up new opportunities for rich insights into political dynamics and ideologies but also increases the workload for manual analysis. Automated speaker attribution, which detects who said what to whom in a speech event and is closely related to semantic role labeling, is an important processing step for computational text analysis. We study the potential of the large language model family Llama 2 to automate speaker attribution in German parliamentary debates from 2017-2021. We fine-tune Llama 2 with QLoRA, an efficient training strategy, and observe our approach to achieve competitive performance in the GermEval 2023 Shared Task On Speaker Attribution in German News Articles and Parliamentary Debates. Our results shed light on the capabilities of large language models in automating speaker attribution, revealing a promising avenue for computational analysis of political discourse and the development of semantic role labeling systems.

Downloads

Published

2024-02-29

Versions

How to Cite

Bornheim, T., Grieger, N., Blaneck , P. G., & Bialonski, S. (2024). Speaker Attribution in German Parliamentary Debates with QLoRA-adapted Large Language Models. Journal for Language Technology and Computational Linguistics, 37(1), 1–13. https://doi.org/10.21248/jlcl.37.2024.244

Issue

Section

Research articles