Aligning Large Language Models with Diverse Political Viewpoints

Dominik Stammbach; Philine Widmer; Eunjung Cho; Caglar Gulcehre; Elliott Ash

EMNLP 2024main3 citations

Aligning Large Language Models with Diverse Political Viewpoints

Dominik Stammbach, Philine Widmer, Eunjung Cho, Caglar Gulcehre, Elliott Ash

Abstract

Large language models such as ChatGPT exhibit striking political biases. If users query them about political information, they often take a normative stance. To overcome this, we align LLMs with diverse political viewpoints from 100,000 comments written by candidates running for national parliament in Switzerland. Models aligned with this data can generate more accurate political viewpoints from Swiss parties, compared to commercial models such as ChatGPT. We also propose a procedure to generate balanced overviews summarizing multiple viewpoints using such models. The replication package contains all code and data.

BibTeX

@inproceedings{stammbach-etal-2024-aligning,
    title = "Aligning Large Language Models with Diverse Political Viewpoints",
    author = "Stammbach, Dominik  and
      Widmer, Philine  and
      Cho, Eunjung  and
      Gulcehre, Caglar  and
      Ash, Elliott",
    editor = "Al-Onaizan, Yaser  and
      Bansal, Mohit  and
      Chen, Yun-Nung",
    booktitle = "Proceedings of the 2024 Conference on Empirical Methods in Natural Language Processing",
    month = nov,
    year = "2024",
    address = "Miami, Florida, USA",
    publisher = "Association for Computational Linguistics",
    url = "https://aclanthology.org/2024.emnlp-main.412/",
    doi = "10.18653/v1/2024.emnlp-main.412",
    pages = "7257--7267"
}