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

arXiv: https://arxiv.org/abs/2604.02557 · HTML: https://mlsanity.com/paper/2604.02557/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 38 of its 56 references that are in the map.
- First posted on arXiv 2 Apr 2026.

## Builds on (most cited first, 30 of 38)

- [LLaVA-OneVision: Easy Visual Task Transfer](https://mlsanity.com/paper/2408.03326/) (2024)
- [SimCSE: Simple Contrastive Learning of Sentence Embeddings](https://mlsanity.com/paper/2104.08821/) (2021)
- [Qwen-Image Technical Report](https://mlsanity.com/paper/2508.02324/) (2025)
- [Deal or No Deal? End-to-End Learning for Negotiation Dialogues](https://mlsanity.com/paper/1706.05125/) (2017)
- [Visually Grounded Reasoning across Languages and Cultures](https://mlsanity.com/paper/2109.13238/) (2021)
- [Cultural Bias and Cultural Alignment of Large Language Models](https://mlsanity.com/paper/2311.14096/) (2023)
- [Survey of Cultural Awareness in Language Models: Text and Beyond](https://mlsanity.com/paper/2411.00860/) (2024)
- [Crossmodal-3600: A Massively Multilingual Multimodal Evaluation Dataset](https://mlsanity.com/paper/2205.12522/) (2022)
- [How to Read Paintings: Semantic Art Understanding with Multi-Modal Retrieval](https://mlsanity.com/paper/1810.09617/) (2018)
- [A fine-grained comparison of pragmatic language understanding in humans and language models](https://mlsanity.com/paper/2212.06801/) (2022)
- [Reasoning About Pragmatics with Neural Listeners and Speakers](https://mlsanity.com/paper/1604.00562/) (2016)
- [Broaden the Vision: Geo-Diverse Visual Commonsense Reasoning](https://mlsanity.com/paper/2109.06860/) (2021)
- [Unified Pragmatic Models for Generating and Following Instructions](https://mlsanity.com/paper/1711.04987/) (2017)
- [Which of These Best Describes Multiple Choice Evaluation with LLMs? A) Forced B) Flawed C) Fixable D) All of the Above](https://mlsanity.com/paper/2502.14127/) (2025)
- [On the Evaluation of Vision-and-Language Navigation Instructions](https://mlsanity.com/paper/2101.10504/) (2021)
- [Pragmatically Informative Image Captioning with Character-Level Inference](https://mlsanity.com/paper/1804.05417/) (2018)
- [A Dataset and Baselines for Visual Question Answering on Art](https://mlsanity.com/paper/2008.12520/) (2020)
- [An image speaks a thousand words, but can everyone listen? On image transcreation for cultural relevance](https://mlsanity.com/paper/2404.01247/) (2024)
- [Pragmatics in the Era of Large Language Models: A Survey on Datasets, Evaluation, Opportunities and Challenges](https://mlsanity.com/paper/2502.12378/) (2025)
- [Culturally Aware and Adapted NLP: A Taxonomy and a Survey of the State of the Art](https://mlsanity.com/paper/2406.03930/) (2024)
- [FoodieQA: A Multimodal Dataset for Fine-Grained Understanding of Chinese Food Culture](https://mlsanity.com/paper/2406.11030/) (2024)
- [FOAM: A Follower-aware Speaker Model For Vision-and-Language Navigation](https://mlsanity.com/paper/2206.04294/) (2022)
- [Multiple LLM Agents Debate for Equitable Cultural Alignment](https://mlsanity.com/paper/2505.24671/) (2025)
- [Modeling Cross-Cultural Pragmatic Inference with Codenames Duet](https://mlsanity.com/paper/2306.02475/) (2023)
- [No Culture Left Behind: ArtELingo-28, a Benchmark of WikiArt with Captions in 28 Languages](https://mlsanity.com/paper/2411.03769/) (2024)
- [ArtELingo: A Million Emotion Annotations of WikiArt with Emphasis on Diversity over Language and Culture](https://mlsanity.com/paper/2211.10780/) (2022)
- [Biased Tales: Cultural and Topic Bias in Generating Children's Stories](https://mlsanity.com/paper/2509.07908/) (2025)
- [Learning to Mediate Disparities Towards Pragmatic Communication](https://mlsanity.com/paper/2203.13685/) (2022)
- [Can MLLMs Understand the Deep Implication Behind Chinese Images?](https://mlsanity.com/paper/2410.13854/) (2024)
- [Beyond Words: Exploring Cultural Value Sensitivity in Multimodal Models](https://mlsanity.com/paper/2502.14906/) (2025)

## Abstract (arXiv)

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%.