Lessons Without Borders? Evaluating Cultural Alignment of LLMs Using Multilingual Story Moral Generation

TL;DR

This paper introduces a multilingual story moral generation task to evaluate cultural alignment in LLMs, revealing models' strengths in semantic similarity and preferences but limitations in cultural diversity.

Contribution

It presents a new dataset and evaluation framework for assessing how well language models capture cultural and moral diversity across languages.

Findings

Models like GPT-4o and Gemini produce morals similar to humans and preferred by evaluators.

Models show less cross-linguistic variation and focus on shared values.

Contemporary models approximate central moral tendencies but lack cultural diversity.

Abstract

Stories are key to transmitting values across cultures, but their interpretation varies across linguistic and cultural contexts. Thus, we introduce multilingual story moral generation as a novel culturally grounded evaluation task. Using a new dataset of human-written story morals collected across 14 language-culture pairs, we compare model outputs with human interpretations via semantic similarity, a human preference survey, and value categorization. We show that frontier models such as GPT-4o and Gemini generate story morals that are semantically similar to human responses and preferred by human evaluators. However, their outputs exhibit markedly less cross-linguistic variation and concentrate on a narrower set of widely shared values. These findings suggest that while contemporary models can approximate central tendencies of human moral interpretation, they struggle to reproduce the diversity that characterizes human narrative understanding. By framing narrative interpretation as an evaluative task, this work introduces a new approach to studying cultural alignment in language models beyond static benchmarks or knowledge-based tests.