CulturePark：提升大语言模型的跨文化理解能力
CulturePark: Boosting Cross-cultural Understanding in Large Language Models
AI总结：
本文提出CulturePark多智能体通信框架，通过LLM模拟跨文化对话生成41,000个文化样本微调八个文化专用模型，在内容审核、文化对齐和文化教育任务上超越GPT-4，证明文化包容性数据对减少文化偏见和推动AI民主化至关重要。
摘要：
文化偏见在许多大语言模型（LLMs）中普遍存在，这在很大程度上是由于缺乏能够代表不同文化的数据。通常，文化数据集和基准的构建方式要么是从现有数据集中提取子集，要么是从维基百科和社交媒体等平台聚合内容。然而，这些方法高度依赖真实世界数据和人工标注，导致成本高昂且难以扩展。受社会交流认知理论的启发，本文提出了CulturePark，一个由LLM驱动的多智能体通信框架，用于文化数据收集。CulturePark利用基于LLM的智能体扮演不同文化中的角色，模拟跨文化人类交流。它生成高质量的跨文化对话，涵盖人类的信念、规范和习俗。利用CulturePark，我们生成了41,000个文化样本，用于微调八个特定文化的LLM。我们在三个下游任务上评估了这些模型：内容审核、文化对齐和文化教育。结果表明，在内容审核方面，我们基于GPT-3.5的模型在数据集上达到或超过了GPT-4的表现。在文化对齐方面，我们的模型在Hofstede的VSM 13框架上超越了GPT-4。此外，在对人类参与者的文化教育中，与GPT-4相比，我们的模型在学习效果和用户体验方面均表现出更优的结果。CulturePark被证明是解决文化偏见和推动AI民主化的重要一步，凸显了文化包容性数据在模型训练中的关键作用。代码已发布在https://github.com/Scarelette/CulturePark。
英文摘要：
Cultural bias is pervasive in many large language models (LLMs), largely due to the deficiency of data representative of different cultures. Typically, cultural datasets and benchmarks are constructed either by extracting subsets of existing datasets or by aggregating from platforms such as Wikipedia and social media. However, these approaches are highly dependent on real-world data and human annotations, making them costly and difficult to scale. Inspired by cognitive theories on social communication, this paper introduces CulturePark, an LLM-powered multi-agent communication framework for cultural data collection. CulturePark simulates cross-cultural human communication with LLM-based agents playing roles in different cultures. It generates high-quality cross-cultural dialogues encapsulating human beliefs, norms, and customs. Using CulturePark, we generated 41,000 cultural samples to fine-tune eight culture-specific LLMs. We evaluated these models across three downstream tasks: content moderation, cultural alignment, and cultural education. Results show that for content moderation, our GPT-3.5-based models either match or outperform GPT-4 on datasets. Regarding cultural alignment, our models surpass GPT-4 on Hofstede's VSM 13 framework. Furthermore, for cultural education of human participants, our models demonstrate superior outcomes in both learning efficacy and user experience compared to GPT-4. CulturePark proves an important step in addressing cultural bias and advancing the democratization of AI, highlighting the critical role of culturally inclusive data in model training. Code is released at https://github.com/Scarelette/CulturePark.
机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。
