tceron/probvaa

Citation text from the card

@article{ceron-etal-2024-beyond,
    title = "Beyond Prompt Brittleness: Evaluating the Reliability and Consistency of Political Worldviews in {LLM}s",
    author = "Ceron, Tanise  and
      Falk, Neele  and
      Bari{\'c}, Ana  and
      Nikolaev, Dmitry  and
      Pad{\'o}, Sebastian",
    journal = "Transactions of the Association for Computational Linguistics",
    volume = "12",
    year = "2024",
    address = "Cambridge, MA",
    publisher = "MIT Press",
    url = "https://aclanthology.org/2024.tacl-1.76/",
    doi = "10.1162/tacl_a_00710",
    pages = "1378--1400",
    abstract = "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 diff

All datasets marked Permissive or share-alike | JSON