WorldView-Bench: A Benchmark for Evaluating Global Cultural Perspectives in Large Language Models

Abstract

Background: Large Language Models (LLMs) are predominantly trained and aligned in ways that reinforce Westerncentric epistemologies and socio-cultural norms, leading to cultural homogenization and limiting their ability to reflect global civilizational plurality. Existing benchmarking frameworks fail to adequately capture this bias, as they rely on rigid, closed-form assessments that overlook the complexity of cultural inclusivity.

Objectives: To address this cultural bias problem, we introduce WorldView-Bench, a benchmark designed to evaluate Global Cultural Inclusivity (GCI) in LLMs by analyzing their ability to accommodate diverse worldviews.

Methods: Our approach is grounded in the Multiplex Worldview proposed by Senturk et al., which distinguishes between Uniplex models, reinforcing cultural homogenization, and Multiplex models, which integrate diverse perspectives. WorldViewBench measures Cultural Polarization, the exclusion of alternative perspectives, through free-form generative evaluation rather than conventional categorical benchmarks. We implement applied multiplexity through two intervention strategies: (1) Contextually-Implemented Multiplex LLMs, where system prompts embed multiplexity principles, and (2) Multi-Agent System (MAS)-Implemented Multiplex LLMs, where multiple LLM agents representing distinct cultural perspectives collaboratively generate responses.

Results: Our results demonstrate a significant increase in Perspectives Distribution Score (PDS) entropy from 13% at baseline to 94% with MAS-Implemented Multiplex LLMs, alongside a shift toward positive sentiment (67.7%) and enhanced cultural balance.

Conclusions: The success of multiplex-aware evaluation in WorldView-Bench demonstrates that cultural bias in LLMs can be meaningfully measured and mitigated through structured worldview diversity. We expect this to pave the way for more inclusive, globally representative, and ethically aligned AI systems.