LLMs Infer Cultural Context but Fail to Apply It When Responding

2026-06-16

AI Summary

While large language models (LLMs) can infer users’ cultural backgrounds, they struggle to proactively generate culturally adapted responses. To address this gap, this work introduces CAPRI, a multi-level dialogue dataset enriched with cultural cues, and proposes a novel evaluation framework grounded in LLMs to assess cultural reasoning capabilities alongside a quantitative cultural sensitivity metric. Systematic experiments reveal—for the first time—that without explicit step-by-step prompting, models fail to effectively apply inferred cultural knowledge (e.g., in units of measurement, temporal expressions, and numerical conventions) during response generation and exhibit a prior bias toward their country-of-origin culture. However, their adaptability improves progressively as cultural cues accumulate. CAPRI establishes a new benchmark for research on cultural adaptation in language models.

Abstract

Recent work has shown that LLMs overrepresent dominant cultures, particularly Western ones, while marginalizing others. We investigate whether this affects models' ability to generate culturally adapted responses by evaluating their use of local measurement units based on the user's perceived cultural background. We introduce Cultural and Pragmatic Response Inference (CAPRI), a dataset of conversations with varying levels of cultural cues. Experiments with state-of-the-art LLMs show that models can infer cultural background and recall relevant conventions, but often fail to utilize the information to adapt their answers to the relevant cultural conventions, unless explicitly prompted to perform the tasks sequentially. We further evaluate adaptation to the interpretation of time and quantity expressions, two subjective language grounding dimensions that are affected by culture. We find that models increasingly adapt their answers as cultural cues accumulate, but their priors are not culture-neutral, sometimes aligning with the model's country of origin. Overall, CAPRI provides a resource for future research aimed at narrowing the gap between cultural knowledge and culturally adaptive language generation.

Problem

• cultural adaptation
• large language models
• cultural bias
• pragmatic response
• cross-cultural communication

Innovation

• cultural adaptation
• large language models
• CAPRI dataset
• pragmatic inference
• cross-cultural NLP