CulturalBench: a Robust, Diverse and Challenging Benchmark on Measuring the (Lack of) Cultural Knowledge of LLMs

• URL: http://arxiv.org/abs/2410.02677v1
• Date: Thu, 3 Oct 2024 17:04:31 GMT
• Title: CulturalBench: a Robust, Diverse and Challenging Benchmark on Measuring the (Lack of) Cultural Knowledge of LLMs
• Authors: Yu Ying Chiu, Liwei Jiang, Bill Yuchen Lin, Chan Young Park, Shuyue Stella Li, Sahithya Ravi, Mehar Bhatia, Maria Antoniak, Yulia Tsvetkov, Vered Shwartz, Yejin Choi,
• Abstract summary: We introduce CulturalBench: a set of 1,227 human-written and human-verified questions for assessing cultural knowledge. We evaluate models on two setups: CulturalBench-Easy and CulturalBench-Hard which share the same questions but asked differently. Compared to human performance (92.6% accuracy), CulturalBench-Hard is more challenging for frontier LLMs with the best performing model (GPT-4o) at only 61.5% and the worst (Llama3-8b) at 21.4%.
• Score: 75.82306181299153
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: To make large language models (LLMs) more helpful across diverse cultures, it is essential to have effective cultural knowledge benchmarks to measure and track our progress. Effective benchmarks need to be robust, diverse, and challenging. We introduce CulturalBench: a set of 1,227 human-written and human-verified questions for effectively assessing LLMs' cultural knowledge, covering 45 global regions including the underrepresented ones like Bangladesh, Zimbabwe, and Peru. Questions - each verified by five independent annotators - span 17 diverse topics ranging from food preferences to greeting etiquettes. We evaluate models on two setups: CulturalBench-Easy and CulturalBench-Hard which share the same questions but asked differently. We find that LLMs are sensitive to such difference in setups (e.g., GPT-4o with 27.3% difference). Compared to human performance (92.6% accuracy), CulturalBench-Hard is more challenging for frontier LLMs with the best performing model (GPT-4o) at only 61.5% and the worst (Llama3-8b) at 21.4%. Moreover, we find that LLMs often struggle with tricky questions that have multiple correct answers (e.g., What utensils do the Chinese usually use?), revealing a tendency to converge to a single answer. Our results also indicate that OpenAI GPT-4o substantially outperform other proprietary and open source models in questions related to all but one region (Oceania). Nonetheless, all models consistently underperform on questions related to South America and the Middle East.

Related papers

• Self-Pluralising Culture Alignment for Large Language Models [36.689491885394034]
  We propose CultureSPA, a framework that allows large language models to align to pluralistic cultures. By comparing culture-aware/unaware outputs, we are able to detect and collect culture-related instances. Extensive experiments demonstrate that CultureSPA significantly improves the alignment of LLMs to diverse cultures without compromising general abilities.
  arXiv  Detail & Related papers  (2024-10-16T19:06:08Z)
• Cultural Value Differences of LLMs: Prompt, Language, and Model Size [35.176429953825924]
  Our study aims to identify behavior patterns in cultural values exhibited by large language models (LLMs) The studied variants include question ordering, prompting language, and model size. Our experiments reveal that query language and model size of LLM are the main factors resulting in cultural value differences.
  arXiv  Detail & Related papers  (2024-06-17T12:35:33Z)
• BLEnD: A Benchmark for LLMs on Everyday Knowledge in Diverse Cultures and Languages [39.17279399722437]
  Large language models (LLMs) often lack culture-specific knowledge of daily life, especially across diverse regions and non-English languages. We introduce BLEnD, a hand-crafted benchmark designed to evaluate LLMs' everyday knowledge across diverse cultures and languages. We construct the benchmark to include two formats of questions: short-answer and multiple-choice.
  arXiv  Detail & Related papers  (2024-06-14T11:48:54Z)
• CulturePark: Boosting Cross-cultural Understanding in Large Language Models [63.452948673344395]
  This paper introduces CulturePark, an LLM-powered multi-agent communication framework for cultural data collection. It generates high-quality cross-cultural dialogues encapsulating human beliefs, norms, and customs. We evaluate these models across three downstream tasks: content moderation, cultural alignment, and cultural education.
  arXiv  Detail & Related papers  (2024-05-24T01:49:02Z)
• Understanding the Capabilities and Limitations of Large Language Models for Cultural Commonsense [98.09670425244462]
  Large language models (LLMs) have demonstrated substantial commonsense understanding. This paper examines the capabilities and limitations of several state-of-the-art LLMs in the context of cultural commonsense tasks.
  arXiv  Detail & Related papers  (2024-05-07T20:28:34Z)
• Does Mapo Tofu Contain Coffee? Probing LLMs for Food-related Cultural Knowledge [47.57055368312541]
  We introduce FmLAMA, a multilingual dataset centered on food-related cultural facts and variations in food practices. We analyze LLMs across various architectures and configurations, evaluating their performance in both monolingual and multilingual settings.
  arXiv  Detail & Related papers  (2024-04-10T08:49:27Z)
• CulturalTeaming: AI-Assisted Interactive Red-Teaming for Challenging LLMs' (Lack of) Multicultural Knowledge [69.82940934994333]
  We introduce CulturalTeaming, an interactive red-teaming system that leverages human-AI collaboration to build challenging evaluation dataset. Our study reveals that CulturalTeaming's various modes of AI assistance support annotators in creating cultural questions. CULTURALBENCH-V0.1 is a compact yet high-quality evaluation dataset with users' red-teaming attempts.
  arXiv  Detail & Related papers  (2024-04-10T00:25:09Z)
• Not All Countries Celebrate Thanksgiving: On the Cultural Dominance in Large Language Models [89.94270049334479]
  This paper identifies a cultural dominance issue within large language models (LLMs) LLMs often provide inappropriate English-culture-related answers that are not relevant to the expected culture when users ask in non-English languages.
  arXiv  Detail & Related papers  (2023-10-19T05:38:23Z)
• Cultural Alignment in Large Language Models: An Explanatory Analysis Based on Hofstede's Cultural Dimensions [10.415002561977655]
  This research proposes a Cultural Alignment Test (Hoftede's CAT) to quantify cultural alignment using Hofstede's cultural dimension framework. We quantitatively evaluate large language models (LLMs) against the cultural dimensions of regions like the United States, China, and Arab countries. Our results quantify the cultural alignment of LLMs and reveal the difference between LLMs in explanatory cultural dimensions.
  arXiv  Detail & Related papers  (2023-08-25T14:50:13Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.