Pragmatics Meets Culture: Culturally-adapted Artwork Description Generation and Evaluation

TL;DR

This paper introduces culturally-adapted artwork description generation, evaluating models' cultural competence and demonstrating that pragmatic speaker models improve comprehension and helpfulness for diverse cultural audiences.

Contribution

It proposes a new task for culturally-adapted art description generation and a framework for evaluating cultural competence using grounded question answering.

Findings

Base models are only marginally adequate for cultural adaptation.

Pragmatic speaker models improve listener comprehension by up to 8.2%.

Human evaluations show increased helpfulness with higher pragmatic competence.

Abstract

Language models are known to exhibit various forms of cultural bias in decision-making tasks, yet much less is known about their degree of cultural familiarity in open-ended text generation tasks. In this paper, we introduce the task of culturally-adapted art description generation, where models describe artworks for audiences from different cultural groups who vary in their familiarity with the cultural symbols and narratives embedded in the artwork. To evaluate cultural competence in this pragmatic generation task, we propose a framework based on culturally grounded question answering. We find that base models are only marginally adequate for this task, but, through a pragmatic speaker model, we can improve simulated listener comprehension by up to 8.2%. A human study further confirms that the model with higher pragmatic competence is rated as more helpful for comprehension by 8.0%.