Beyond Prompt Brittleness: Evaluating the Reliability and Consistency of Political Worldviews in LLMs
Transactions of the Association for Computational Linguistics · 04.11.2024 · 12 · 1378-1400
DOI: 10.1162/tacl_a_00710 ↗Kurz gefasst
Fachartikel in der Zeitschrift Transactions of the Association for Computational Linguistics, online erschienen am 04.11.2024. Verfasst von Tanise Ceron und 4 weiteren Autor:innen. Die Arbeit wurde bisher 18-mal zitiert (Crossref) und nennt 59 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0).
Abstract (Originalsprache)
Due to the widespread use of large language models (LLMs), we need to understand whether they embed a specific “worldview” and what these views reflect. Recent studies report that, prompted with political questionnaires, LLMs show left-liberal leanings (Feng et al., 2023; Motoki et al., 2024). However, it is as yet unclear whether these leanings are reliable (robust to prompt variations) and whether the leaning is consistent across policies and political leaning. We propose a series of tests which assess the reliability and consistency of LLMs’ stances on political statements based on a dataset of voting-advice questionnaires collected from seven EU countries and annotated for policy issues. We study LLMs ranging in size from 7B to 70B parameters and find that their reliability increases with parameter count. Larger models show overall stronger alignment with left-leaning parties but differ among policy programs: They show a (left-wing) positive stance towards environment protection, social welfare state, and liberal society but also (right-wing) law and order, with no consistent preferences in the areas of foreign policy and migration.
Zitieren
APA
Ceron, T., Falk, N., Barić, A., Nikolaev, D., & Padó, S. (2024). Beyond Prompt Brittleness: Evaluating the Reliability and Consistency of Political Worldviews in LLMs. Transactions of the Association for Computational Linguistics, 12, 1378-1400. https://doi.org/10.1162/tacl_a_00710
Vancouver
Ceron T, Falk N, Barić A, Nikolaev D, Padó S. Beyond Prompt Brittleness: Evaluating the Reliability and Consistency of Political Worldviews in LLMs. Transactions of the Association for Computational Linguistics. 2024;12:1378-1400. doi:10.1162/tacl_a_00710
BibTeX
@article{Ceron2024,
author = {Ceron, Tanise and Falk, Neele and Barić, Ana and Nikolaev, Dmitry and Padó, Sebastian},
title = {Beyond Prompt Brittleness: Evaluating the Reliability and Consistency of Political Worldviews in LLMs},
journal = {Transactions of the Association for Computational Linguistics},
year = {2024},
volume = {12},
pages = {1378--1400},
doi = {10.1162/tacl_a_00710}
}
EUVIMED