The study compares deep learning models trained via transfer learning from ImageNet (TII-models) against those trained solely on medical images (LMI-models) for disease segmentation. Results show that combining outputs from both model types can improve segmentation performance by up to 10% in certain scenarios. A repository of models, code, and over 10,000 medical images is available on GitHub to facilitate further research.
Researchers developed a data-driven toolkit for short-term traffic forecasting using high-resolution traffic data from urban road sensors. The method models forecasting as a matrix completion problem, mapping inputs to a higher-dimensional space using kernels and adaptive boosting. Validated using real-world data from Abu Dhabi, UAE, the method outperforms state-of-the-art algorithms.
A KAUST team led by Xin Gao developed an AI model for COVID-19 detection from CT scans, addressing limitations of existing methods. The model incorporates a novel embedding strategy, a CT scan simulator, and a 2.5D deep-learning algorithm. Tested at King Faisal Specialist Hospital, the model demonstrated high accuracy in detecting COVID-19 cases. Why it matters: This research provides a valuable tool for rapid and accurate COVID-19 diagnosis in the region, especially in early-stage infections, improving healthcare outcomes.
The paper proposes a method for causal inference using satellite image time series to determine the impact of interventions on climate change, focusing on quantifying deforestation due to human causes. The method uses computer vision and deep learning to detect forest tree coverage levels over time and Bayesian structural causal models to estimate counterfactuals. The framework is applied to analyze deforestation levels before and after the hyperinflation event in Brazil in the Amazon rainforest region.
KAUST researchers collaborated with the Paris Observatory and the National Astronomical Observatory of Japan (NAOJ) to develop advanced Extreme-AO algorithms for habitable exoplanet imaging. The new algorithms, powered by KAUST's linear algebra code running on NVIDIA GPUs, optimize and anticipate atmospheric disturbances. The implemented Singular Value Decomposition (SVD) algorithm won an award at the PASC Conference 2018 and is used at the Subaru Telescope in Hawaii. Why it matters: This advancement enhances the ability to image exoplanets, potentially leading to breakthroughs in the search for habitable planets using ground-based telescopes.
KAUST researchers led by Dr. Niveen Khashab have developed thermosensitive liposomes for controlled drug release, particularly in cancer therapies. The liposomes are designed to release drugs only when they reach heated tumor tissue, minimizing systemic side effects. Cholesterol moieties are used as anchors to create a "nail" or "comb" effect, enabling temperature-triggered drug release inside cells. Why it matters: This targeted drug delivery system could significantly improve the efficacy and reduce the toxicity of cancer treatments.
KAUST researchers developed a new single-molecule imaging method called the cumulative-area (CA) method. This method allows for simultaneous characterization of size, shape, and conformational dynamics of individual molecules, along with accurate determination of diffusion kinetics. The researchers demonstrated the CA method's effectiveness on nano- and micro-sized objects, extracting quantitative information about size, diffusion, and relaxation time. Why it matters: This advancement expands the capabilities of molecule imaging techniques in the region and has potential applications in polymer dynamics research and the study of molecular mechanisms within cells.
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 faculty and students presented 31 papers at the 2022 Conference on Computer Vision and Pattern Recognition (CVPR), including 6 oral presentations. Professors Fahad Khan and Shijian Lu had 9 and 8 papers accepted respectively. Researchers collaborated with 57 institutions across 16 countries. Why it matters: MBZUAI's strong showing at a top-tier CV conference demonstrates the rapid growth and international collaboration of AI research in the UAE.
MBZUAI researchers are refining AI techniques to improve cancer diagnosis for colorectal and breast cancer, both common in the Middle East. They are using "few-shot tissue image generation," in which AI generates data for training AI models to recognize lesions, addressing the challenge of limited training data. The developed framework improves the efficiency of radiologists in breast cancer diagnosis, leading to better detection of breast lesions and timely treatment interventions. Why it matters: These advancements in AI-aided diagnostics can lead to earlier and more accurate cancer detection, ultimately improving patient outcomes in the region and beyond.
MBZUAI has published 674 papers in 2023 and holds a global ranking of 18 in AI, CV, ML, NLP, and robotics according to CSRankings. The university presented 30 papers at ICCV, will present 53 papers at NeurIPS, and has 44 papers at EMNLP 2023. MBZUAI was also awarded its first patent by the US Patent Office for a system and method for handwriting generation. Why it matters: This demonstrates the rapid growth and increasing prominence of MBZUAI as a leading AI research institution in the region and globally.
MBZUAI researchers have developed an AI program using vision transformers that can learn a person's handwriting style and generate text in that style. The US Patent and Trademark Office recently granted a patent for this technology, which could aid individuals with writing impairments. The system overcomes limitations of previous GAN-based approaches by processing long-range dependencies in handwriting. Why it matters: This patented AI tool enhances personalized text generation and has potential applications in assistive technology and improving handwriting recognition models.
MBZUAI researchers, in collaboration with TUM, developed Open-YOLO 3D, a new method for open-vocabulary 3D instance segmentation. Open-YOLO 3D enables robots to detect and differentiate individual objects in a 3D scene without being limited to predefined object categories, using both camera images and lidar-generated 3D point clouds. The new system was shown to be more accurate and significantly faster than previous approaches. Why it matters: This advancement enhances robots' ability to understand and interact with dynamic, real-world environments, bringing robots closer to being useful in everyday life.
MBZUAI researchers won second place at the AgentX Competition at UC Berkeley for their benchmark measuring AI agents' reasoning across images, comparisons, and video. The Agent-X dataset includes 828 tasks across six domains, requiring agents to use 14 executable tools without explicit instructions. Agent-X analyzes the agent's full reasoning trajectory, unlike typical evaluations that focus only on final answers. Why it matters: The benchmark exposes limitations in current multimodal AI agents and provides a more rigorous evaluation framework for real-world applications in the region and beyond.
MBZUAI researchers received high honors at EMNLP 2025 for two research papers, placing them in the top 2% of accepted work. One paper, MAviS, is a multimodal AI system that identifies bird species by combining images, sounds, and text. The other award-winning paper focuses on uncertainty in LLM-as-a-Judge. Why it matters: The recognition highlights MBZUAI's growing influence in NLP and multimodal AI research, particularly in domain-specific applications like biodiversity conservation.
MBZUAI researchers have developed MAviS, a new multimodal dataset, benchmark, and chatbot for fine-grained bird species recognition. MAviS includes images, audio, and text to help models identify subtle differences between species, especially rare and regional varieties. The related study was presented at EMNLP 2025 and selected as a "Senior Area Chair Highlight". Why it matters: This work addresses a key limitation in AI's ability to support biodiversity conservation and ecological monitoring in the region and globally.
MBZUAI researchers have introduced SURPRISE3D, a benchmark for evaluating 3D spatial reasoning in AI systems, along with a 3D Spatial Reasoning Segmentation (3D-SRS) task. The benchmark includes over 900 indoor scenes and 200,000 language queries paired with 3D masks, emphasizing spatial relationships over object naming. A companion paper, MLLM-For3D, explores adapting 2D multimodal LLMs for 3D reasoning. Why it matters: This work addresses a key limitation in current AI, pushing towards embodied AI that can understand and act in 3D environments based on human-like spatial reasoning.
MBZUAI researchers introduced CausalVerse, a new benchmark for causal representation learning (CRL) presented at NeurIPS 2025. CausalVerse combines high-fidelity visual complexity with access to underlying causal variables and graphs, featuring 200,000 images and 300 million video frames across 24 sub-scenes in four domains. It aims to provide a realistic and precise testbed to evaluate whether CRL methods can truly learn the right causes. Why it matters: By bridging the gap between toy datasets and real-world data, CausalVerse can drive advances in AI systems capable of understanding causality in complex scenarios.
MBZUAI researchers developed a new AI-generated image detection method called 'consistency verification' (ConV). Instead of training on labeled real and fake images, ConV identifies structural patterns unique to real photos using a data manifold concept. The system modifies images and uses DINOv2 to measure the difference between original and transformed representations, classifying images based on their proximity to the manifold. Why it matters: This approach offers a more robust way to detect AI-generated images without needing training data from every image generator, addressing a key limitation in the rapidly evolving landscape of AI image synthesis.
MBZUAI researchers tackled the challenge of AI-powered waste detection in messy, real-world recycling facilities. They fine-tuned modern object detection models on real industrial waste imagery and combined this with a semi-supervised learning pipeline. Fine-tuning more than doubled performance and their semi-supervised pipeline outperformed fully supervised baselines. Why it matters: This research offers a practical path for open research that can rival proprietary systems while reducing the need for costly manual labeling in waste management, a problem of global importance.
MBZUAI's Institute of Foundation Models (IFM) has developed PAN, a novel interactive world model capable of long-horizon simulation. PAN uses a Generative Latent Prediction (GLP) architecture, coupling internal latent reasoning with generative supervision in the visual domain. The model evolves an internal latent state conditioned on history and natural language actions, then decodes that state into a video segment using a Causal Swin-DPM mechanism for smooth transitions. Why it matters: PAN represents a significant advancement in AI's ability to simulate and predict evolving environments, enabling more steerable and coherent long-term video generation and opening new possibilities for interactive AI systems.
MBZUAI researchers have developed MorphDiff, a diffusion model that predicts cell morphology from gene expression data. MorphDiff uses the transcriptome to generate realistic post-perturbation images, either from scratch or by transforming a control image. The model combines a Morphology Variational Autoencoder (MVAE) with a Latent Diffusion Model, enabling both gene-to-image generation and image-to-image transformation. Why it matters: This could significantly accelerate drug discovery and biological research by allowing scientists to preview cellular changes before conducting experiments.
A new paper at ICCV 2025, co-authored by MBZUAI Ph.D. student Dmitry Demidov, introduces Dense-WebVid-CoVR, a 1.6-million sample benchmark for composed video retrieval (CoVR). The benchmark features longer, context-rich descriptions and modification texts, generated using Gemini Pro and GPT-4o, with manual verification. The paper also presents a unified fusion approach that jointly reasons across video and text inputs, improving performance on fine-grained edit details. Why it matters: This work advances video search capabilities by enabling more human-like queries, which is crucial for creative and analytic workflows that require nuanced video retrieval.
MBZUAI Ph.D. candidate Muhammad Maaz has been awarded the 2025 Google Ph.D. Fellowship in Machine Perception. Maaz is the first student from MBZUAI and the first from the Gulf region to receive this recognition, which includes funding, mentorship, and $50,000. He has published extensively in top-tier CV/NLP venues and has over 4,500 citations. Why it matters: This award highlights the growing prominence of MBZUAI and the increasing quality of AI research in the Gulf region on the global stage.
Researchers from MBZUAI, IBM, and ServiceNow introduced GEOBench-VLM, a benchmark for evaluating vision-language models on Earth observation tasks using satellite and aerial imagery. The benchmark includes over 10,000 human-verified instructions across 31 sub-tasks spanning object classification, localization, change detection, and more. GEOBench-VLM addresses the gap in current VLMs' ability to perform spatially grounded reasoning and change detection in satellite imagery. Why it matters: This benchmark will drive progress in AI's ability to analyze satellite data for critical applications like disaster response, climate monitoring, and urban planning in the Middle East and globally.
MBZUAI researchers presented DEFUSE-MS at MICCAI 2025, a novel AI system for analyzing changes in MRI scans of multiple sclerosis (MS) patients. DEFUSE-MS uses a deformation field-guided spatiotemporal graph-based framework to identify new lesions by reasoning about how the brain has changed. The model constructs graphs of small regions within baseline and follow-up MRIs, linking them across time with edges enriched with learned embeddings of the deformation field. Why it matters: DEFUSE-MS reframes the task from simple "spot the difference" to understanding structural changes, potentially improving the speed and accuracy of MS diagnosis and treatment monitoring.
MBZUAI and Cleveland Clinic Abu Dhabi jointly showcased AI applications to address UAE healthcare challenges, focusing on diabetes, heart disease, and vision loss. The event highlighted AI-powered robots for surgical tool guidance, autonomous ultrasound systems, and AI-based eye diagnostics for early disease detection. Researchers presented advances in AI-powered intrabody navigation, robotic ultrasound imaging, and AI-based oculomics. Why it matters: These AI innovations have the potential to transform healthcare delivery in the UAE, improving surgical outcomes, democratizing access to diagnostic imaging, and enabling earlier disease detection.
MBZUAI researchers are introducing MedNNS, a system to be presented at MICCAI 2025, that recommends the best AI architecture and initialization for medical imaging tasks. MedNNS addresses the challenge of inefficient trial-and-error in building medical imaging AI by reframing model selection as a retrieval problem. The system employs a Once-For-All ResNet-like model and a learned meta-space of 720k model-dataset pairs, using dataset embeddings to predict optimal model performance. Why it matters: By automating model selection, MedNNS promises to significantly reduce the time and resources required to develop effective AI solutions for healthcare, particularly in medical imaging.
MBZUAI researchers developed a new approach called Multimodal Optimal Transport via Grounded Retrieval (MOTOR) to improve the accuracy of vision-language models for medical image analysis. MOTOR combines retrieval-augmented generation (RAG) with an optimal transport algorithm to retrieve and rank relevant image and textual data. Testing on two medical datasets showed that MOTOR improved average performance by 6.45%. Why it matters: This technique addresses the challenges of limited specialized medical datasets and computational costs associated with training AI models for medical image interpretation, offering a more efficient and accurate solution.
MBZUAI Ph.D. students Muhammad Maaz and Hanoona Rasheed interned at Meta, developing a vision encoder for images and videos. They created PerceptionLM, a multimodal language model, to generate synthetic video-caption data to train the Perception Encoder. The team addressed the challenge of limited labeled video data by building a multimodal language model called PerceptionLM to understand video's spatial and temporal aspects. Why it matters: This highlights MBZUAI's strength in computer vision and provides students opportunities to contribute to cutting-edge research at global tech firms.
MBZUAI researchers developed AirCast, a novel AI model for improved air pollution forecasting, which won the best paper award at the TerraBytes workshop during ICML. AirCast fuses weather and chemistry data using a Vision Transformer and frequency-weighted MAE to better predict extreme events like Saharan dust storms. In tests across the Middle East and North Africa, AirCast reduced PM2.5 error by 33% compared to a persistence baseline and outperformed the CAMS physics model. Why it matters: Accurate air pollution forecasting is critical for public health in the GCC region, and this research demonstrates a significant advancement using AI to address this challenge.
MBZUAI researchers, in collaboration with Monash University, have introduced ArEnAV, a new dataset for deepfake detection featuring Arabic-English code-switching. The dataset comprises 765 hours of manipulated YouTube videos, incorporating intra-utterance code-switching and dialect variations. Experiments showed that code-switching significantly reduces the performance of existing deepfake detectors. Why it matters: This work addresses a critical gap in AI's ability to handle linguistic diversity, particularly in regions where code-switching is prevalent, enhancing the reliability of deepfake detection in real-world scenarios.
MBZUAI Ph.D. student Raza Imam and colleagues presented a new benchmark called MediMeta-C to test the robustness of medical vision-language models (MVLMs) under real-world image corruptions. They found that top-performing MVLMs on clean data often fail under mild corruption, with fundoscopy models particularly vulnerable. To address this, they developed RobustMedCLIP (RMC), a lightweight defense using few-shot LoRA tuning to improve model robustness. Why it matters: This research highlights the critical need for robustness testing in medical AI to ensure reliability in clinical settings, particularly in resource-constrained environments where image quality may be compromised.
Researchers from Carnegie Mellon University and MBZUAI have developed a new method called ConceptAligner for precise image editing using AI. The system decomposes text embeddings into independent building blocks called atomic concepts, allowing users to make targeted tweaks without generating entirely new images. Their approach ensures that each latent factor maps to a specific user-controllable dial, enabling accurate concept-level modifications. Why it matters: This research addresses a major limitation in AI image generation, enhancing its usefulness in industries where precise control is crucial, such as advertising and medicine, and improving the reliability of AI-driven creative tools.
MBZUAI researchers have developed GeoPixel, a new multimodal model for pixel grounding in remote sensing images. GeoPixel associates individual pixels with object categories, enabling detailed image analysis by linking language to objects at the pixel level. The model was trained on a new dataset and benchmark, outperforming existing systems in precision. Why it matters: This advancement enhances the utility of remote sensing data for critical applications like environmental management and disaster response by providing more granular and accurate image interpretation.
MBZUAI researchers developed Mobile-VideoGPT, a compact and efficient multimodal model for real-time video understanding on edge devices. The system uses keyframe selection, efficient token projection, and a Qwen-2.5-0.5B language model. Testing showed that Mobile-VideoGPT is faster and performs better than other models while being significantly smaller, and the model and code are publicly available. Why it matters: This research enables on-device AI processing for video, reducing reliance on remote servers and addressing privacy concerns, which can accelerate the adoption of AI in mobile and embedded applications.
Researchers from MBZUAI and other institutions have developed a new framework called STEREO to improve the safety of text-to-image diffusion models. STEREO uses a two-stage approach: STE (Search Thoroughly Enough) based on adversarial training and REO (Robustly Erase Once) for batch concept erasure. This framework aims to enhance safety without significantly impacting the model's performance on normal queries. Why it matters: The framework addresses vulnerabilities in AI image generation, reducing the creation of inappropriate images while preserving performance on harmless queries.
MBZUAI researchers developed FarSight, a plugin to reduce hallucinations in Multimodal Large Language Models (MLLMs). FarSight addresses the issue where MLLMs generate inaccurate text by losing focus on relevant image details, leading to snowball hallucinations. Testing on models like LLaVA-1.5-7B showed FarSight's effectiveness in reducing initial mistakes, thereby minimizing overall hallucinations. Why it matters: Improving the reliability of MLLMs is crucial for applications requiring high accuracy, enhancing their utility in various real-world scenarios.
Researchers at MBZUAI, IBM Research, and other institutions have developed EarthDial, a new vision-language model (VLM) specifically designed to process geospatial data from remote sensing technologies. EarthDial handles data in multiple modalities and resolutions, processing images captured at different times to observe environmental changes. The model outperformed others on over 40 tasks including image classification, object detection, and change detection. Why it matters: This unified model bridges the gap between generic VLMs and domain-specific models, enabling complex geospatial data analysis for applications like disaster assessment and climate monitoring in the region.
MBZUAI and Corniche Hospital researchers have developed FetalCLIP, a foundation model for analyzing fetal ultrasound images to detect congenital conditions. FetalCLIP outperformed other foundation models on ultrasound analysis tasks. The AI model aims to improve the early diagnosis of ailments like congenital heart defects. Why it matters: This innovation has the potential to dramatically improve health outcomes for millions of children annually by providing physicians with better insights into fetal health.
MBZUAI researchers introduced Web2Code, a new dataset suite, at NeurIPS to enhance multimodal LLM performance in web page analysis and HTML generation. The suite includes a fine-tuning dataset and two benchmark datasets. Instruction tuning with Web2Code improved performance on specialized tasks without affecting general capabilities. Why it matters: This contribution addresses a key limitation in current multimodal LLMs, potentially boosting productivity in web design and development by providing targeted training data.
MBZUAI doctoral student Mai A. Shaaban and colleagues developed MedPromptX, a system that analyzes chest X-rays and patient data to aid lung disease diagnoses. MedPromptX uses multimodal large language models with visual grounding and few-shot prompting, trained on a new dataset of 6,000 patient records (MedPromptX-VQA) derived from MIMIC-IV and MIMIC-CXR. The system addresses the challenge of incomplete electronic health records by leveraging the knowledge embedded in large language models to interpret lab results. Why it matters: This research advances AI-driven medical diagnostics by integrating diverse data sources and addressing data gaps, potentially leading to quicker and more accurate diagnoses.
MBZUAI researchers developed a method to adapt Meta's Segment Anything Model (SAM) for medical image segmentation, addressing its performance gap with natural images. Their approach improves SAM's accuracy without requiring extensive retraining or large medical image datasets. The research, led by Chao Qin, was nominated for the Best Paper Award at the MICCAI conference in Marrakesh. Why it matters: This offers a more efficient and effective way to leverage foundation models in specialized medical imaging applications, potentially improving diagnostic accuracy and reducing the need for large-scale, domain-specific training data.
MBZUAI, in partnership with IBM Research, is developing GeoChat+, a vision-language model (VLM) for multi-modal, temporal remote sensing image analysis. GeoChat+ builds on the previous GeoChat model, enhancing capabilities with multi-modal images from various Earth observation systems like Sentinel-1, Sentinel-2, Landsat, and high-resolution imagery. GeoChat+ will integrate data from multiple satellites at different times to detect environmental changes and analyze the impact on soil quality, air quality, and erosion. Why it matters: This advancement promises to revolutionize geographic data analysis, providing detailed reports for high-risk regions and aiding reforestation efforts.
MBZUAI researchers presented EXAMS-V, a new benchmark dataset for evaluating the reasoning and processing abilities of vision language models (VLMs). EXAMS-V contains over 20,000 multiple-choice questions across 26 subjects and 11 languages, including Arabic. The dataset presents the questions within images, testing the VLM's ability to integrate visual and textual information. Why it matters: This dataset fills a gap in VLM evaluation, providing a valuable resource for assessing and improving the multimodal reasoning capabilities of these models, particularly in diverse languages like Arabic.
MBZUAI researchers found that ImageNet performance isn't always indicative of real-world task performance for computer vision models. The study analyzed four popular model configurations, revealing variations in behavior on specific image types despite similar overall ImageNet accuracy. It indicates that certain model configurations are better suited for particular tasks, even with lower ImageNet scores. Why it matters: This challenges the reliance on ImageNet as a sole benchmark and highlights the need for task-specific evaluations in computer vision.
MBZUAI M.Sc. graduate Mohammad Hanan Ghani developed new techniques to improve text-to-image generation from long text prompts, combining large language models and diffusion models. Advised by Dr. Salman Khan, Ghani published three papers at ICLR, BMVC, and NeurIPS, with the ICLR paper focusing on generating images that accurately reflect detailed text descriptions. The new system improves upon existing techniques to generate images that closely follow the details of the input text. Why it matters: This research addresses a key limitation in current T2I models and advances the field of multimodal AI, potentially improving the capabilities of robots and autonomous devices.
MBZUAI researchers developed FeSViBS, a new federated split learning technique for vision transformers that addresses data scarcity and privacy concerns in healthcare image classification. The method combines federated learning and split learning to train models collaboratively without sharing sensitive patient data directly. It overcomes limitations of traditional centralized training and vulnerabilities in federated learning. Why it matters: This approach enables the development of AI-powered healthcare applications while adhering to stringent data privacy regulations, unlocking the potential of machine learning in medical imaging.
MBZUAI and Sheikh Shakbout Medical City researchers developed PECon, a deep learning method for pulmonary embolism detection using CT scans and electronic health records. PECon uses neural networks and contrastive learning to encode and align image and text data. The method aims to improve diagnosis accuracy and speed, potentially saving lives. Why it matters: This research demonstrates AI's potential to enhance medical diagnostics in the UAE, addressing a critical healthcare challenge.
MBZUAI faculty and students will present 53 papers at NeurIPS 2023. Key faculty include Eric Xing, Kun Zhang, and Tongliang Liu, each contributing to nine papers. One paper was selected for oral presentation and two for spotlight presentations. Why it matters: MBZUAI's strong presence at NeurIPS highlights its growing influence in the global AI research community and its focus on high-impact AI research.