WorldValuesBench: A Large-Scale Benchmark Dataset for Multi-Cultural Value Awareness of Language Models

TL;DR

WorldValuesBench is a large-scale, diverse benchmark dataset derived from the World Values Survey, designed to evaluate and improve language models' understanding of multi-cultural human values across various demographic contexts.

Contribution

The paper introduces WorldValuesBench, the first large-scale dataset for multi-cultural value prediction, enabling systematic study of language models' awareness of diverse human values.

Findings

Strong models struggle with value prediction accuracy.

GPT-3.5 Turbo approaches human answer distribution on some questions.

The dataset reveals significant challenges in multi-cultural value understanding.

Abstract

The awareness of multi-cultural human values is critical to the ability of language models (LMs) to generate safe and personalized responses. However, this awareness of LMs has been insufficiently studied, since the computer science community lacks access to the large-scale real-world data about multi-cultural values. In this paper, we present WorldValuesBench, a globally diverse, large-scale benchmark dataset for the multi-cultural value prediction task, which requires a model to generate a rating response to a value question based on demographic contexts. Our dataset is derived from an influential social science project, World Values Survey (WVS), that has collected answers to hundreds of value questions (e.g., social, economic, ethical) from 94,728 participants worldwide. We have constructed more than 20 million examples of the type "(demographic attributes, value question) $\rightarrow$ answer" from the WVS responses. We perform a case study using our dataset and show that the task is challenging for strong open and closed-source models. On merely $11.1\%$, $25.0\%$, $72.2\%$, and $75.0\%$ of the questions, Alpaca-7B, Vicuna-7B-v1.5, Mixtral-8x7B-Instruct-v0.1, and GPT-3.5 Turbo can respectively achieve $<0.2$ Wasserstein 1-distance from the human normalized answer distributions. WorldValuesBench opens up new research avenues in studying limitations and opportunities in multi-cultural value awareness of LMs.