Beyond prompt brittleness: Evaluating the reliability and consistency of political worldviews in LLMs
Fuente:
arXiv
Salvato in:
| Autori principali: | , , , , |
|---|---|
| Natura: | Preprint |
| Pubblicazione: |
2024
|
| Soggetti: | |
| Accesso online: | |
| Tags: |
Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
|
| _version_ | 1866929453482901504 |
|---|---|
| author | Ceron, Tanise Falk, Neele Barić, Ana Nikolaev, Dmitry Padó, Sebastian |
| author_facet | Ceron, Tanise Falk, Neele Barić, Ana Nikolaev, Dmitry Padó, Sebastian |
| contents | 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. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2402_17649 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | Beyond prompt brittleness: Evaluating the reliability and consistency of political worldviews in LLMs Ceron, Tanise Falk, Neele Barić, Ana Nikolaev, Dmitry Padó, Sebastian Computation and Language Computers and Society 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. |
| title | Beyond prompt brittleness: Evaluating the reliability and consistency of political worldviews in LLMs |
| topic | Computation and Language Computers and Society |
| url | https://arxiv.org/abs/2402.17649 |