emirates7 - Since its inception in 2019, Mohamed bin Zayed University of Artificial Intelligence (MBZUAI) has been dedicated to pushing the frontiers of AI research and innovation, focusing on real-world challenges in areas such as healthcare, sustainability, education, and smart cities.
This week, attendees at two significant technology events in the UAE will have the chance to explore the advancements made by MBZUAI faculty, students, and researchers over the past five years in various AI-driven domains, along with a preview of some exciting new projects the university is fostering.
Visitors at the IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS 2024) in Abu Dhabi and the annual GITEX Global technology event in Dubai will discover new initiatives from MBZUAI, including:
FRAPPE (FRAming, Persuasion, and Propaganda Explorer)
FRAPPE is a system designed to analyze media content for framing, persuasion, and propaganda techniques. It offers real-time evaluations of individual articles, focusing on their genre, framing methods, and persuasive strategies, allowing for comparisons across different media outlets, countries, and languages. This innovative tool is intended for researchers, media professionals, and the general public, promoting media literacy by uncovering techniques used to influence opinions and emotions. By enhancing understanding of media manipulation, FRAPPE fosters informed decision-making and supports transparent information dissemination worldwide.
BiMediX2 – AI Virtual Arabic Doctor
BiMediX2 is the first bilingual medical large language model (LLM) and large multimodal model (LMM) capable of understanding and generating content across various data types, including text, images, audio, and video. It facilitates a wide array of medical interactions, including multi-turn conversations and question-answering, and can even analyze medical images like X-rays, CT scans, and MRIs. The research team anticipates that this service will be especially beneficial in remote areas where doctors conduct telemedicine consultations. They are currently enhancing BiMediX2's functionality to help patients locate nearby physicians and facilities and schedule appointments for further medical care.
Dog Bot and Immersive Command Center for Robot Dog Navigation
‘Dog Bot’ is a visual analytics generative pre-trained transformer (GPT) that integrates image, text, and audio capabilities, enabling speech and vision recognition and comprehension, as well as audio responses. This solution has potential applications in various sectors, including inspecting work sites for health and safety compliance, assisting law enforcement through area patrols, and surveying agricultural sites for crop performance. Essentially, Dog Bot could serve as the eyes and ears for vital public services, even supporting healthcare by interacting with patients in remote locations and transmitting diagnostic images to physicians.
GeoChat+
MBZUAI researchers are also addressing key sustainability issues with GeoChat+, a vision-language model (VLM) designed for multi-modal, temporal remote sensing (RS) image analysis. This project aims to monitor deforestation to help mitigate biodiversity loss, protect endangered species, and limit ecological disruption. GeoChat+ seeks to transform geographic data analysis by incorporating various satellite and imaging sources, utilizing data from prominent Earth observation systems like Sentinel-1, Sentinel-2, Landsat, and high-resolution images from WorldView and Google Earth. The goal is to provide comprehensive geospatial insights for applications ranging from environmental monitoring to disaster response and urban planning.
This week, attendees at two significant technology events in the UAE will have the chance to explore the advancements made by MBZUAI faculty, students, and researchers over the past five years in various AI-driven domains, along with a preview of some exciting new projects the university is fostering.
Visitors at the IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS 2024) in Abu Dhabi and the annual GITEX Global technology event in Dubai will discover new initiatives from MBZUAI, including:
FRAPPE (FRAming, Persuasion, and Propaganda Explorer)
FRAPPE is a system designed to analyze media content for framing, persuasion, and propaganda techniques. It offers real-time evaluations of individual articles, focusing on their genre, framing methods, and persuasive strategies, allowing for comparisons across different media outlets, countries, and languages. This innovative tool is intended for researchers, media professionals, and the general public, promoting media literacy by uncovering techniques used to influence opinions and emotions. By enhancing understanding of media manipulation, FRAPPE fosters informed decision-making and supports transparent information dissemination worldwide.
BiMediX2 – AI Virtual Arabic Doctor
BiMediX2 is the first bilingual medical large language model (LLM) and large multimodal model (LMM) capable of understanding and generating content across various data types, including text, images, audio, and video. It facilitates a wide array of medical interactions, including multi-turn conversations and question-answering, and can even analyze medical images like X-rays, CT scans, and MRIs. The research team anticipates that this service will be especially beneficial in remote areas where doctors conduct telemedicine consultations. They are currently enhancing BiMediX2's functionality to help patients locate nearby physicians and facilities and schedule appointments for further medical care.
Dog Bot and Immersive Command Center for Robot Dog Navigation
‘Dog Bot’ is a visual analytics generative pre-trained transformer (GPT) that integrates image, text, and audio capabilities, enabling speech and vision recognition and comprehension, as well as audio responses. This solution has potential applications in various sectors, including inspecting work sites for health and safety compliance, assisting law enforcement through area patrols, and surveying agricultural sites for crop performance. Essentially, Dog Bot could serve as the eyes and ears for vital public services, even supporting healthcare by interacting with patients in remote locations and transmitting diagnostic images to physicians.
GeoChat+
MBZUAI researchers are also addressing key sustainability issues with GeoChat+, a vision-language model (VLM) designed for multi-modal, temporal remote sensing (RS) image analysis. This project aims to monitor deforestation to help mitigate biodiversity loss, protect endangered species, and limit ecological disruption. GeoChat+ seeks to transform geographic data analysis by incorporating various satellite and imaging sources, utilizing data from prominent Earth observation systems like Sentinel-1, Sentinel-2, Landsat, and high-resolution images from WorldView and Google Earth. The goal is to provide comprehensive geospatial insights for applications ranging from environmental monitoring to disaster response and urban planning.