BLEnD: A Benchmark for LLMs on Everyday Knowledge in Diverse Cultures and Languages

TL;DR

BLEnD is a comprehensive benchmark designed to evaluate large language models' knowledge of everyday cultural facts across diverse regions and languages, highlighting disparities in model performance based on cultural and linguistic representation.

Contribution

This paper introduces BLEnD, a new benchmark with 52,600 questions across 16 countries and 13 languages, addressing the gap in evaluating LLMs' cultural knowledge beyond high-resource languages.

Findings

LLMs perform better on cultures with high online representation.

Performance varies significantly between languages and cultures.

Models perform better in English for low-resource languages.

Abstract

Large language models (LLMs) often lack culture-specific knowledge of daily life, especially across diverse regions and non-English languages. Existing benchmarks for evaluating LLMs' cultural sensitivities are limited to a single language or collected from online sources such as Wikipedia, which do not reflect the mundane everyday lifestyles of diverse regions. That is, information about the food people eat for their birthday celebrations, spices they typically use, musical instruments youngsters play, or the sports they practice in school is common cultural knowledge but uncommon in easily collected online sources, especially for underrepresented cultures. To address this issue, we introduce BLEnD, a hand-crafted benchmark designed to evaluate LLMs' everyday knowledge across diverse cultures and languages. BLEnD comprises 52.6k question-answer pairs from 16 countries/regions, in 13 different languages, including low-resource ones such as Amharic, Assamese, Azerbaijani, Hausa, and Sundanese. We construct the benchmark to include two formats of questions: short-answer and multiple-choice. We show that LLMs perform better for cultures that are highly represented online, with a maximum 57.34% difference in GPT-4, the best-performing model, in the short-answer format. For cultures represented by mid-to-high-resource languages, LLMs perform better in their local languages, but for cultures represented by low-resource languages, LLMs perform better in English than the local languages. We make our dataset publicly available at: https://github.com/nlee0212/BLEnD.