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.
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.
The Technology Innovation Institute (TII) will host the GlobalEM high-power electromagnetics conference in Abu Dhabi from November 13–17. The conference, organized by TII’s Directed Energy Research Center (DERC), will feature sessions on sources, antennas, IEMI threats, high energy lasers, and critical infrastructure impacts. GlobalEM brings together experts to discuss challenges and opportunities in electromagnetics. Why it matters: The event strengthens the advanced directed energy domain in the UAE and supports Abu Dhabi's goal of becoming a hub for innovation in mitigating electromagnetic risks.
Saee, a KAUST-supported Saudi startup based in Jeddah, aims to disrupt last-mile delivery in Saudi Arabia by addressing challenges like delays and cash-on-delivery (COD) issues. They offer services such as fast COD consolidation, flexible dispatch to freelancers, and warehouse management. Saee also introduces a new concept called Cash Before Delivery (CBD) to improve cash flow for e-commerce platforms. Why it matters: This aims to solve a key bottleneck for e-commerce growth in Saudi Arabia, where COD is prevalent and creates financial inefficiencies.
Mae AlMansoori from TII's Directed Energy Research Center won the Young Scientist Award at URSI Kleinheubacher Tagung 2020. Her paper introduced a correlation metric to evaluate the influence of random variables on high-power electromagnetic sources, specifically a Vircator model. The study combined extreme value theory and descriptive statistics to analyze peak output power variability and identify dominant factors. Why it matters: The research offers a framework for managing uncertainties in high-power electromagnetic sources and optimizing their efficiency, relevant for defense and energy applications in the region.
Ceer and KAUST are partnering to research and develop innovations in smart mobility, including connectivity and autonomous driving technologies. The collaboration aims to transform research into economically productive innovations, supporting Saudi-based SMEs and the automotive manufacturing industry. Ceer and KAUST will share facilities and conduct collaborative activities to foster knowledge exchange and create jobs. Why it matters: This partnership signifies a crucial step towards diversifying Saudi Arabia's economy and advancing technological capabilities in the electric vehicle sector within the Kingdom.
Todd Nims, a filmmaker born in Saudi Arabia, premiered his film "Joud" at KAUST's 2018 Winter Enrichment Program. The film, set in Saudi Arabia, explores the cycle of life in reverse and the meaning of "Joud" (generosity in the face of scarcity). Nims describes Saudi Arabia as a "magical place" due to its rich storytelling tradition. Why it matters: The article highlights KAUST's role in showcasing cultural works and supporting Saudi artists, though the AI relevance is limited.
KAUST startups, including Sadeem (a wireless sensor system for flood monitoring), have won top prizes in entrepreneurship competitions such as GITEX Future Stars, ArabNet Riyadh, and the Oqal Best Entrepreneur Award. Sadeem won Best Global Startup at GITEX and took home $100,000. KAUST supports early-stage startups through accelerator programs like TAQADAM and 9/10ths, as well as the KAUST Research & Technology Park. Why it matters: This success highlights KAUST's role in fostering innovation and entrepreneurship in Saudi Arabia, contributing to the Kingdom's growing startup ecosystem and economic diversification efforts.