Beyond Translation: Cross-Cultural Meme Transcreation with Vision-Language Models

TL;DR

This paper explores the challenge of adapting memes across cultures using vision-language models, introducing a large dataset and an evaluation framework to analyze the quality and limitations of current models in cross-cultural meme transcreation.

Contribution

It presents a hybrid framework and a large-scale dataset for cross-cultural meme transcreation, along with an analysis of model performance and transferability of humor and design elements.

Findings

Models perform limited cross-cultural transcreation

US-to-Chinese transcreation is higher quality than Chinese-to-US

Identifies aspects of humor and design that transfer or remain challenging

Abstract

Memes are a pervasive form of online communication, yet their cultural specificity poses significant challenges for cross-cultural adaptation. We study cross-cultural meme transcreation, a multimodal generation task that aims to preserve communicative intent and humor while adapting culture-specific references. We propose a hybrid transcreation framework based on vision-language models and introduce a large-scale bidirectional dataset of Chinese and US memes. Using both human judgments and automated evaluation, we analyze 6,315 meme pairs and assess transcreation quality across cultural directions. Our results show that current vision-language models can perform cross-cultural meme transcreation to a limited extent, but exhibit clear directional asymmetries: US-Chinese transcreation consistently achieves higher quality than Chinese-US. We further identify which aspects of humor and visual-textual design transfer across cultures and which remain challenging, and propose an evaluation framework for assessing cross-cultural multimodal generation. Our code and dataset are publicly available at https://github.com/AIM-SCU/MemeXGen.