This section brings together the main scholarly and practical outputs associated with the project. It is a central reference point for publications, conference presentations, and research resources.

Conference presentations

• González-Bustamante, B., Klamm, C., Bellens, T., & Koch, M. (2026). From Benchmark to Lessons: Guidelines from Computational Analysis of Parliamentary Speech. Presentation delivered at the RadiUnce Project Meeting, Utrecht, Netherlands. [Upon request at ocpsg@politics.ox.ac.uk]

• González-Bustamante, B., Bellens, T., Klamm, C., & Koch, M. (2026). A proof-of-concept benchmark of LLMs and fine-tuned transformers for multilingual policy agenda coding. Presentation delivered at the 8th Annual COMPTEXT Conference, Birmingham, UK. [Upon request at ocpsg@politics.ox.ac.uk]

• Klamm, C., González-Bustamante, B., Koch, M., & Bellens, T. (2026). Evaluating Large Language Models for Multilingual Policy Coding: A Transparent Benchmarking and Reporting Framework. Presentation delivered at the APSA Workshop 2026, Virtual Research Group on Aligning Computational Tools for the Political Science Research Lifecycle, Virtual. [Upon request at ocpsg@politics.ox.ac.uk]

• González-Bustamante, B., Bellens, T., Klamm, C., & Koch, M. (2026). Benchmarking LLMs and Fine-Tuned Models for Multilingual Policy Agenda Annotation. Presentation delivered at the OCPSG Research Cafe Research Cafe, Virtual. [Open PDF]

Resources

This subsection shall include documentation, benchmark repositories, datasets, codebases, and related materials.

• Release name generator. Generate a deterministic codename from a seed, with optional Semantic Versioning. https://ocpsg-benchmarking-llms.github.io/release-name-generator/.