LLMs show limited ability to emulate subgroup cultural values via persona prompts, with fine-tuning providing gains that come with widened fairness disparities.
WVS Respon- dents
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.CY 1years
2026 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
Can Persona-Prompted LLMs Emulate Subgroup Values? An Empirical Analysis of Generalisability and Fairness in Cultural Alignment
LLMs show limited ability to emulate subgroup cultural values via persona prompts, with fine-tuning providing gains that come with widened fairness disparities.