arXiv Open Access 2024

Aligning Large Language Models with Diverse Political Viewpoints

Dominik Stammbach Philine Widmer Eunjung Cho Caglar Gulcehre Elliott Ash

Lihat Sumber

Abstrak

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.

Topik & Kata Kunci

cs.CL

Penulis (5)

Dominik Stammbach

Philine Widmer

Eunjung Cho

Caglar Gulcehre

Elliott Ash

Format Sitasi

APA MLA BibTeX

Stammbach, D., Widmer, P., Cho, E., Gulcehre, C., Ash, E. (2024). Aligning Large Language Models with Diverse Political Viewpoints. https://arxiv.org/abs/2406.14155

Akses Cepat

Lihat di Sumber

Informasi Jurnal

Tahun Terbit: 2024
Bahasa: en
Sumber Database: arXiv
Akses: Open Access ✓