Researchers from MBZUAI, University of Washington, and other institutions presented studies at EMNLP 2024 exploring how LLMs represent cultures. A survey analyzed dozens of recent studies on LLMs and culture and proposes a new framework for future research. The survey found that there is no widely accepted definition of 'culture' in NLP, making it challenging to interpret how models represent culture through language. Why it matters: This highlights a key gap in the field and emphasizes the need for a more rigorous and consistent understanding of culture in AI, especially as LLMs become more globally integrated.
MBZUAI researchers presented two studies at NAACL 2025 concerning how LLMs understand cultural differences, with one study winning the SAC award. One study, titled "Reading between the lines: Can LLMs identify cross-cultural communication gaps," assesses GPT-4o's ability to identify cultural references in Goodreads book reviews. The researchers created a benchmark dataset using annotations from 50 evaluators across different cultures to measure the LLM's ability to identify culture-specific items (CSIs). Why it matters: Improving LLMs' cross-cultural understanding is crucial for ensuring these models can be used effectively and equitably across diverse global contexts.
A new paper from MBZUAI introduces JEEM, a benchmark dataset for evaluating vision-language models on their understanding of images grounded in four Arabic-speaking societies (Jordan, UAE, Egypt, and Morocco) and their ability to use local dialects. The dataset comprises 2,178 images and 10,890 question-answer pairs reflecting everyday life and culturally specific scenes. Evaluation of several Arabic-capable models (Maya, PALO, Peacock, AIN, AyaV) and GPT-4o revealed that while models can generate fluent language, they struggle with genuine understanding, consistency, and relevance, especially when cultural context is important. Why it matters: This research highlights the challenges of building AI systems that can truly understand and interact with diverse cultures, emphasizing the need for culturally grounded datasets and evaluation metrics.
MBZUAI researchers found that only 5.7% of music in existing datasets used to train generative music systems comes from non-Western genres. They discovered that 94% of the music represented Western music, while Africa, the Middle East, and South Asia accounted for only 0.3%, 0.4%, and 0.9% respectively. The team also tested whether parameter-efficient fine-tuning with adapters could improve generative music systems on underrepresented styles, presenting their findings at NAACL. Why it matters: This research highlights the critical need for more diverse datasets in AI music generation to better serve global musical traditions and audiences.
MBZUAI is conducting research to improve cross-cultural understanding using AI, including studying LLM limitations in recognizing cultural references. They developed "Culturally Yours," a tool that helps users comprehend cultural references in text, and the "All Languages Matter Benchmark" (ALM Bench) to evaluate multimodal LLMs across 100 languages. MBZUAI has also developed LLMs tailored to low-resource languages like Jais (Arabic), Nanda (Hindi), and Sherkala (Kazakh). Why it matters: These initiatives promote inclusivity and ensure AI systems are culturally aware and can serve diverse populations effectively, particularly in the Middle East's multicultural context.
MBZUAI researchers release JEEM, a new benchmark dataset for evaluating vision-language models on Arabic dialects. The dataset covers image captioning and visual question answering tasks using images from Jordan, UAE, Egypt, and Morocco. Results show models struggle with cultural understanding and relevance despite fluent language generation.
MBZUAI researchers, in collaboration with over 70 researchers, have created the Culturally diverse Visual Question Answering (CVQA) benchmark to evaluate cultural understanding in multimodal LLMs. The CVQA dataset includes over 10,000 questions in 31 languages and 13 scripts, testing models on images of local dishes, personalities, and monuments. Testing of several multimodal LLMs on the CVQA benchmark revealed significant challenges, even for top models. Why it matters: This benchmark highlights the need for AI models to better understand diverse cultures, promoting fairness and relevance across different languages and regions.
MBZUAI researchers have released ALM Bench, a new benchmark dataset for evaluating the performance of multimodal LLMs on cultural visual question-answer tasks across 100 languages. The dataset includes over 22,000 question-answer pairs across 19 categories, with a focus on low-resource languages and cultural nuances, including three Arabic dialects. They tested 16 open- and closed-source multimodal LLMs on it, revealing a significant need for greater cultural and linguistic inclusivity. Why it matters: The benchmark aims to improve the inclusivity of multimodal AI systems by addressing the underrepresentation of low-resource languages and cultural contexts.