Tailored Emotional LLM-Supporter: Enhancing Cultural Sensitivity

2025-08-11

AI Summary

Current large language models (LLMs) exhibit insufficient cultural sensitivity in cross-cultural emotional support, primarily due to scarce high-quality multicultural annotated data and a lack of systematic adaptation methodologies. Method: We introduce CultureCare—the first fine-grained, dual-annotated dataset jointly labeling emotional and cultural dimensions across four cultures (Chinese, American, Japanese, and Indian). We propose four culture-adaptation strategies integrating parameter-efficient fine-tuning and prompt engineering, moving beyond simplistic role-playing paradigms. Evaluation employs a tripartite framework involving LLM-based adjudication, culturally embedded human annotators, and clinical psychologists. Results: Culturally adapted models significantly outperform anonymized online peer responses in both cultural sensitivity and empathic quality. Furthermore, they demonstrate promising potential for enhancing cultural competence in clinical training contexts.

Technology Category

• Machine Learning: Large Multimodal Models (LMMs)
• Natural Language Processing: (Large) Language Models
• Cognitive Modeling & Cognitive Systems: Adaptive Behavior

Application Category

• Search and Retrieval-Augmented AI: Multilingual and cross-lingual Web search
• User Modeling, Personalization and Recommendation: Large Language Models (LLM) for user modeling and recommendation
• Economics, Online Markets and Human Computation: Humans versus LLMs for data annotation and labeling

Abstract

Large language models (LLMs) show promise in offering emotional support and generating empathetic responses for individuals in distress, but their ability to deliver culturally sensitive support remains underexplored due to lack of resources. In this work, we introduce CultureCare, the first dataset designed for this task, spanning four cultures and including 1729 distress messages, 1523 cultural signals, and 1041 support strategies with fine-grained emotional and cultural annotations. Leveraging CultureCare, we (i) develop and test four adaptation strategies for guiding three state-of-the-art LLMs toward culturally sensitive responses; (ii) conduct comprehensive evaluations using LLM judges, in-culture human annotators, and clinical psychologists; (iii) show that adapted LLMs outperform anonymous online peer responses, and that simple cultural role-play is insufficient for cultural sensitivity; and (iv) explore the application of LLMs in clinical training, where experts highlight their potential in fostering cultural competence in future therapists.

Problem

• Enhancing cultural sensitivity in LLM emotional support responses
• Developing a dataset for culturally sensitive emotional support
• Evaluating LLM adaptation strategies for cultural sensitivity

Innovation

• Develop CultureCare dataset for cultural sensitivity
• Test four adaptation strategies on LLMs
• Evaluate with LLM judges and human experts