10.06.2026
As Large Language Models become increasingly integrated into scientific workflows, our researchers have helped establish GUIDE-LLM, a reporting standard designed to enhance transparency, comparability, and reproducibility in AI-assisted research.
Large Language Models (LLMs) are becoming an increasingly important tool in scientific research. However, differences in prompts, model versions, and configurations can significantly affect results, making transparency and reproducibility a growing challenge.
Professor Stefan Feuerriegel and Professor Barbara Plank, together with junior researchers Kerstin Forster, Dominique Geissler, Abdurahman Maarouf, and Sebastian Maier, and in collaboration with international partners, have developed GUIDE-LLM, a framework for transparent reporting of LLM use in research. Their work has been published in Nature Human Behaviour.
Based on input from 80 experts across disciplines, GUIDE-LLM provides a 14-item checklist that helps researchers document how and why LLMs were used, which models and prompts were employed, and how outputs were validated. The framework aims to improve transparency, comparability, and reproducibility in AI-assisted research.
As LLMs become standard tools in science, GUIDE-LLM offers an important step toward ensuring that research remains rigorous, trustworthy, and reproducible.
Your can read the article here
Feuerriegel, S., Barrie, C., Crockett, M. J., Globig, L. K., McLoughlin, K. L., Mirea, D.-M., Spirling, A., Yang, D., ..., Rathje, S., & Ribeiro, M. H. (2026). A reporting checklist for LLMs in behavioural science. In: Nature Human Behaviour. DOI: 10.1038/s41562-026-02492-7.