What question did this study set out to answer?

This research aims to analyze the representational equality of large language models in simulating human opinions across different countries.

Representational Equality in Cross-country Value Simulation: A Systematic Analysis of Large Language Models

View Full Paper

Ask AI

Bookmark

Authors

Systematic analysis investigates simulation accuracy across countries, highlighting inequalities in opinion representation.

This research aims to analyze the representational equality of large language models in simulating human opinions across different countries.
Investigated simulation accuracy across 59 countries using large language models.
Compared two intervention pathways: contextual adaptation and parametric modification.
Analyzed the impact of language prompting and preference data on simulation accuracy and equality.
Found substantial inequality in simulation accuracy across countries, favoring wealthier nations.
Contextual adaptation improves accuracy but is model-dependent; additional information often enhances both accuracy and equality.
Parametric modification shows uneven improvements in accuracy across language groups, with human-annotated data outperforming AI-annotated data.

Jian et al. (2026) studied this question.

View Full Paper

Ask AI

Bookmark

View Full Paper

Ask AI

Bookmark

Authors

Systematic analysis investigates simulation accuracy across countries, highlighting inequalities in opinion representation.

This research aims to analyze the representational equality of large language models in simulating human opinions across different countries.
Investigated simulation accuracy across 59 countries using large language models.
Compared two intervention pathways: contextual adaptation and parametric modification.
Analyzed the impact of language prompting and preference data on simulation accuracy and equality.
Found substantial inequality in simulation accuracy across countries, favoring wealthier nations.
Contextual adaptation improves accuracy but is model-dependent; additional information often enhances both accuracy and equality.
Parametric modification shows uneven improvements in accuracy across language groups, with human-annotated data outperforming AI-annotated data.

Jian et al. (2026) studied this question.

View Full Paper

Ask AI

Bookmark