Topics ›
50 articles RSS ↗
A new approach to composed video retrieval (CoVR) is presented, which leverages large multimodal models to infer causal and temporal consequences implied by an edit. The method aligns reasoned queries to candidate videos without task-specific finetuning. A new benchmark, CoVR-Reason, is introduced to evaluate reasoning in CoVR.
The paper introduces SectEval, a new benchmark to evaluate sectarian biases in LLMs concerning Sunni and Shia Islam, available in English and Hindi. Results show significant inconsistencies in LLM responses based on language, with some models favoring Shia responses in English but Sunni in Hindi. Location-based experiments further reveal that advanced models adapt their responses based on the user's claimed country, while smaller models exhibit a consistent Sunni-leaning bias.
MBZUAI researchers introduce DuwatBench, a new benchmark for multimodal understanding of Arabic calligraphy. The dataset contains 1,272 samples across six calligraphic styles with detailed annotations to evaluate visual-text alignment. Evaluation of 13 multimodal models reveals challenges in processing calligraphic variations and artistic distortions, highlighting the need for culturally grounded AI research.
A national survey in Saudi Arabia of 330 participants reveals that 93% are actively using Generative AI, primarily for text-based tasks, while awareness and understanding remain uneven. Participants recognize benefits like productivity but caution against risks such as privacy, misinformation, and ethical misuse. The study highlights the need for AI literacy, culturally aligned solutions, and stronger frameworks for responsible deployment in Saudi Arabia.
This paper introduces a hybrid deep learning and machine learning pipeline for classifying construction and demolition waste. A dataset of 1,800 images from UAE construction sites was created, and deep features were extracted using a pre-trained Xception network. The combination of Xception features with machine learning classifiers achieved up to 99.5% accuracy, demonstrating state-of-the-art performance for debris identification.
arXiv · · Research Healthcare
This paper introduces an explainable machine learning framework for early-stage chronic kidney disease (CKD) screening, specifically designed for low-resource settings in Bangladesh and South Asia. The framework utilizes a community-based dataset from Bangladesh and evaluates multiple ML classifiers with feature selection techniques. Results show that the ML models achieve high accuracy and sensitivity, outperforming existing screening tools and demonstrating strong generalizability across independent datasets from India, the UAE, and Bangladesh.
A new benchmark, LongShOTBench, is introduced for evaluating multimodal reasoning and tool use in long videos, featuring open-ended questions and diagnostic rubrics. The benchmark addresses the limitations of existing datasets by combining temporal length and multimodal richness, using human-validated samples. LongShOTAgent, an agentic system, is also presented for analyzing long videos, with both the benchmark and agent demonstrating the challenges faced by state-of-the-art MLLMs.
A new study compares vision-language models (VLMs) to YOLOv8 for wastewater treatment plant (WWTP) identification in satellite imagery across the MENA region. VLMs like Gemma-3 demonstrate superior zero-shot performance compared to YOLOv8, trained on a dataset of 83,566 satellite images from Egypt, Saudi Arabia, and UAE. The research suggests VLMs offer a scalable, annotation-free alternative for remote sensing of WWTPs.
Researchers at MBZUAI introduce "Interactive Video Reasoning," a new paradigm enabling models to actively "think with videos" by performing iterative visual actions to gather and refine evidence. They developed Video CoM, which reasons through a Chain of Manipulations (CoM), and constructed Video CoM Instruct, an 18K instruction tuning dataset for multi-step manipulation reasoning. The model is further optimized via reinforcement learning with reasoning aware Group Relative Policy Optimization (GRPO), achieving strong results across nine video reasoning benchmarks.
Researchers at MBZUAI have introduced Video-R2, a reinforcement learning approach to improve the consistency and visual grounding of reasoning in multimodal language models. Video-R2 combines timestamp-aware supervised fine-tuning with Group Relative Policy Optimization (GRPO) guided by a Temporal Alignment Reward (TAR). The model demonstrates higher Think Answer Consistency (TAC), Video Attention Score (VAS), and accuracy across multiple benchmarks, showing improved temporal alignment and reasoning coherence for video understanding.
This paper analyzes the energy consumption and carbon footprint of LLM inference in the UAE compared to Iceland, Germany, and the USA. The study uses DeepSeek Coder 1.3B and the HumanEval dataset to evaluate code generation. It provides a comparative analysis of geographical trade-offs for climate-aware AI deployment, specifically addressing the challenges and potential of datacenters in desert regions.
Researchers at MBZUAI have introduced EvoLMM, a self-evolving framework for large multimodal models that enhances reasoning capabilities without human-annotated data or reward distillation. EvoLMM uses two cooperative agents, a Proposer and a Solver, which generate image-grounded questions and solve them through internal consistency, using a continuous self-rewarding process. Evaluations using Qwen2.5-VL as the base model showed performance gains of up to 3% on multimodal math-reasoning benchmarks like ChartQA, MathVista, and MathVision using only raw training images.
This paper introduces Cross-Document Topic-Aligned (CDTA) chunking to address knowledge fragmentation in Retrieval-Augmented Generation (RAG) systems. CDTA identifies topics across documents, maps segments to topics, and synthesizes them into unified chunks. Experiments on HotpotQA and UAE legal texts show that CDTA improves faithfulness and citation accuracy compared to existing chunking methods, especially for complex queries requiring multi-hop reasoning.
This study compares AI uptake in the UAE and Kuwait, analyzing how constitutional, collective-choice, and operational rules shape AI implementation and its impact on citizen centricity and public value creation. It finds that the UAE's concentrated authority and pro-innovation environment enable scaling AI initiatives, while Kuwait's dispersed governance and cautious approach limit progress despite similar resources. The research highlights the importance of vertical rule coherence over wealth in determining AI's public-value yield.
A new method is proposed to reduce the verbosity of LLMs in step-by-step reasoning by retaining moderately easy problems during Reinforcement Learning with Verifiable Rewards (RLVR) training. This approach acts as an implicit length regularizer, preventing the model from excessively increasing output length on harder problems. Experiments using Qwen3-4B-Thinking-2507 show the model achieves baseline accuracy with nearly twice shorter solutions.
Researchers introduce ALARB, a new benchmark for evaluating reasoning in Arabic LLMs using 13K Saudi commercial court cases. The benchmark includes tasks like verdict prediction, reasoning chain completion, and identification of relevant regulations. Instruction-tuning a 12B parameter model on ALARB achieves performance comparable to GPT-4o in verdict prediction and generation.
The study analyzes over 1,000 images generated by ImageFX, DALL-E V3, and Grok for 56 Saudi professions, finding significant gender imbalances and cultural inaccuracies. DALL-E V3 exhibited the strongest gender stereotyping, with 96% male depictions, particularly in leadership and technical roles. The research underscores the need for diverse training data and culturally sensitive evaluation to ensure equitable AI outputs that accurately reflect Saudi Arabia's labor market and culture.
The researchers introduce KAU-CSSL, the first continuous Saudi Sign Language (SSL) dataset focusing on complete sentences. They propose a transformer-based model using ResNet-18 for spatial feature extraction and a Transformer Encoder with Bidirectional LSTM for temporal dependencies. The model achieved 99.02% accuracy in signer-dependent mode and 77.71% in signer-independent mode, advancing communication tools for the SSL community.
This paper presents a UI-level evaluation of ALLaM-34B, an Arabic-centric LLM developed by SDAIA and deployed in the HUMAIN Chat service. The evaluation used a prompt pack spanning various Arabic dialects, code-switching, reasoning, and safety, with outputs scored by frontier LLM judges. Results indicate strong performance in generation, code-switching, MSA handling, reasoning, and improved dialect fidelity, positioning ALLaM-34B as a robust Arabic LLM suitable for real-world use.
This paper introduces Saudi-Dialect-ALLaM, a LoRA fine-tuned version of the Saudi Arabian foundation model ALLaM-7B-Instruct-preview, designed to improve the generation of Saudi dialects (Najdi and Hijazi). The model is trained on a private dataset of 5,466 synthetic instruction-response pairs, with two variants explored: Dialect-Token and No-Token training. Results indicate that the Dialect-Token model achieves superior dialect control and fidelity compared to generic instruction models, although the dataset and model weights are not released.
Researchers introduce UnsafeChain, a new safety alignment dataset designed to improve the safety of large reasoning models (LRMs) by focusing on 'hard prompts' that elicit harmful outputs. The dataset identifies and corrects unsafe completions into safe responses, exposing models to unsafe behaviors and guiding their correction. Fine-tuning LRMs on UnsafeChain demonstrates enhanced safety and preservation of general reasoning ability compared to existing datasets like SafeChain and STAR-1.
This paper introduces Absher, a new benchmark for evaluating LLMs' linguistic and cultural competence in Saudi dialects. The benchmark comprises over 18,000 multiple-choice questions spanning six categories, using dialectal words, phrases, and proverbs from various regions of Saudi Arabia. Evaluation of state-of-the-art LLMs reveals performance gaps, especially in cultural inference and contextual understanding, highlighting the need for dialect-aware training.
A new benchmark, ViMUL-Bench, is introduced to evaluate video LLMs across 14 languages, including Arabic, with a focus on cultural inclusivity. The benchmark includes 8k manually verified samples across 15 categories and varying video durations. A multilingual video LLM, ViMUL, is also presented, along with a training set of 1.2 million samples, with both to be publicly released.
Researchers from MBZUAI have introduced VideoMolmo, a large multimodal model for spatio-temporal pointing conditioned on textual descriptions. The model incorporates a temporal module with an attention mechanism and a temporal mask fusion pipeline using SAM2 for improved coherence across video sequences. They also curated a dataset of 72k video-caption pairs and introduced VPoS-Bench, a benchmark for evaluating generalization across real-world scenarios, with code and models publicly available.
MBZUAI researchers introduce VideoMathQA, a new benchmark for evaluating mathematical reasoning in videos, requiring models to interpret visual information, text, and spoken cues. The dataset spans 10 mathematical domains with videos ranging from 10 seconds to over 1 hour, and includes multi-step reasoning annotations. The benchmark aims to evaluate temporal cross-modal reasoning and highlights the limitations of existing approaches in complex video-based mathematical problem solving.
MBZUAI introduces Agent-X, a benchmark for evaluating multi-step reasoning in vision-centric agents across real-world, multimodal settings. Agent-X includes 828 tasks with diverse visual contexts and spans six environments, requiring tool use and stepwise decision-making. Experiments show that current LLMs struggle with multi-step vision tasks, achieving less than 50% success, highlighting areas for improvement in LMM reasoning and tool use.
MBZUAI researchers release 'Fann or Flop', a new benchmark for evaluating Arabic poetry understanding in LLMs. The benchmark covers 12 historical eras and 14 poetic genres, assessing semantic understanding, metaphor interpretation, and cultural context. Evaluation of state-of-the-art LLMs reveals challenges in poetic understanding despite strong performance on standard Arabic benchmarks.
MBZUAI researchers introduce ARB, the first comprehensive benchmark for evaluating step-by-step multimodal reasoning in Arabic across textual and visual modalities. The benchmark spans 11 diverse domains and includes 1,356 multimodal samples with 5,119 human-curated reasoning steps. Evaluations of 12 state-of-the-art LMMs revealed challenges in coherence, faithfulness, and cultural grounding, highlighting the need for culturally aware AI systems.
Researchers from MBZUAI have introduced UrduFactCheck, a new framework for fact-checking in Urdu, along with two datasets: UrduFactBench and UrduFactQA. The framework uses monolingual and translation-based evidence retrieval to address the lack of Urdu resources. Evaluations using twelve LLMs showed that translation-augmented methods improve performance, highlighting challenges for open-source LLMs in Urdu.
The paper introduces SaudiCulture, a new benchmark for evaluating the cultural competence of LLMs within Saudi Arabia, covering five major geographical regions and diverse cultural domains. The benchmark includes questions of varying complexity and distinguishes between common and specialized regional knowledge. Evaluations of five LLMs (GPT-4, Llama 3.3, FANAR, Jais, and AceGPT) revealed performance declines on region-specific questions, highlighting the need for region-specific knowledge in LLM training.
MBZUAI researchers introduce LLMVoX, a 30M-parameter, LLM-agnostic, autoregressive streaming text-to-speech (TTS) system that generates high-quality speech with low latency. The system preserves the capabilities of the base LLM and achieves a lower Word Error Rate compared to speech-enabled LLMs. LLMVoX supports seamless, infinite-length dialogues and generalizes to new languages with dataset adaptation, including Arabic.
A new culturally inclusive and linguistically diverse dataset called Palm for Arabic LLMs is introduced, covering 22 Arab countries and featuring instructions in both Modern Standard Arabic (MSA) and dialectal Arabic (DA) across 20 topics. The dataset was built through a year-long community-driven project involving 44 researchers from across the Arab world. Evaluation of frontier LLMs using the dataset reveals limitations in cultural and dialectal understanding, with some countries being better represented than others.
Researchers introduce a benchmark to evaluate the factual recall and knowledge transferability of multilingual language models across 13 languages. The study reveals that language models often fail to transfer knowledge between languages, even when they possess the correct information in one language. The benchmark and evaluation framework are released to drive future research in multilingual knowledge transfer.
Researchers introduce TimeTravel, a benchmark dataset for evaluating large multimodal models (LMMs) on historical and cultural artifacts. The benchmark comprises 10,250 expert-verified samples across 266 cultures and 10 historical regions, designed to assess AI in tasks like classification and interpretation of manuscripts, artworks, inscriptions, and archaeological discoveries. The goal is to establish AI as a reliable partner in preserving cultural heritage and assisting researchers.
arXiv · · LLM Healthcare
MBZUAI releases BiMediX2, a bilingual (Arabic-English) Bio-Medical Large Multimodal Model, along with the BiMed-V dataset (1.6M samples) and BiMed-MBench evaluation benchmark. BiMediX2 supports multi-turn conversation in Arabic and English and handles diverse medical imaging modalities. The model achieves state-of-the-art results on medical LLM and LMM benchmarks, outperforming existing methods and GPT-4 in specific evaluations.
MBZUAI researchers release OpenFactCheck, a unified framework to evaluate the factual accuracy of large language models. The framework includes modules for response evaluation, LLM evaluation, and fact-checker evaluation. OpenFactCheck is available as an open-source Python library, a web service, and via GitHub.
MBZUAI researchers introduce Web2Code, a new large-scale dataset and evaluation framework for training and benchmarking multimodal LLMs on webpage understanding and HTML code generation. The dataset includes webpage images, HTML code, and QA pairs about webpage content. Experiments demonstrate the dataset's utility in webpage understanding, code generation, and general visual domain tasks, with code and data available on Github.
Researchers introduce two new benchmarks, derived from the Qiyas exam, to evaluate mathematical reasoning and language understanding in Arabic. They tested ChatGPT-3.5-turbo and ChatGPT-4, which achieved 49% and 64% accuracy respectively. The new benchmarks aim to address the lack of resources for evaluating Arabic language models.
MBZUAI researchers introduce VideoGPT+, a novel video Large Multimodal Model (LMM) that integrates image and video encoders to leverage both spatial and temporal information in videos. They also introduce VCGBench-Diverse, a comprehensive benchmark for evaluating video LMMs across 18 video categories. VideoGPT+ demonstrates improved performance on multiple video benchmarks, including VCGBench and MVBench.
Researchers from MBZUAI have introduced the Complex Video Reasoning and Robustness Evaluation Suite (CVRR-ES) for assessing Video-LLMs. The benchmark evaluates models across 11 real-world video dimensions, revealing challenges in robustness and reasoning, particularly for open-source models. A training-free Dual-Step Contextual Prompting (DSCP) technique is proposed to enhance Video-LMM performance, with the dataset and code made publicly available.
arXiv · · Research Healthcare
Researchers at MBZUAI have developed DynaMMo, a dynamic model merging method for efficient class incremental learning using medical images. DynaMMo merges multiple networks at different training stages using lightweight learnable modules, reducing computational overhead. Evaluated on three datasets, DynaMMo achieved a 10-fold reduction in GFLOPS compared to existing dynamic methods with a 2.76 average accuracy drop.
Researchers from MBZUAI have released MobiLlama, a fully transparent open-source 0.5 billion parameter Small Language Model (SLM). MobiLlama is designed for resource-constrained devices, emphasizing enhanced performance with reduced resource demands. The full training data pipeline, code, model weights, and checkpoints are available on Github.
Researchers introduce PALO, a polyglot large multimodal model with visual reasoning capabilities in 10 major languages including Arabic. A semi-automated translation approach was used to adapt the multimodal instruction dataset from English to the target languages. The models are trained across three scales (1.7B, 7B and 13B parameters) and a multilingual multimodal benchmark is proposed for evaluation.
arXiv · · LLM Healthcare
MBZUAI researchers introduce BiMediX, a bilingual (English and Arabic) mixture of experts LLM for medical applications. The model is trained on BiMed1.3M, a new 1.3 million bilingual instruction dataset and outperforms existing models like Med42 and Jais-30B on medical benchmarks. Code and models are available on Github.
This paper investigates the intrinsic self-correction capabilities of LLMs, identifying model confidence as a key latent factor. Researchers developed an "If-or-Else" (IoE) prompting framework to guide LLMs in assessing their own confidence and improving self-correction accuracy. Experiments demonstrate that the IoE-based prompt enhances the accuracy of self-corrected responses, with code available on GitHub.
MBZUAI researchers introduce M4GT-Bench, a new benchmark for evaluating machine-generated text (MGT) detection across multiple languages and domains. The benchmark includes tasks for binary MGT detection, identifying the specific model that generated the text, and detecting mixed human-machine text. Experiments with baseline models and human evaluation show that MGT detection performance is highly dependent on access to training data from the same domain and generators.
Researchers from the National Center for AI in Saudi Arabia investigated the sensitivity of Large Language Model (LLM) leaderboards to minor benchmark perturbations. They found that small changes, like choice order, can shift rankings by up to 8 positions. The study recommends hybrid scoring and warns against over-reliance on simple benchmark evaluations, providing code for further research.
Researchers introduce Arabic Mini-ClimateGPT, a tailored Arabic LLM for climate change and sustainability. The model is fine-tuned on the Clima500-Instruct dataset and uses vector embedding retrieval during inference. Evaluations show the model outperforms baseline LLMs and is preferred by experts in 81.6% of cases.
Researchers at MBZUAI have developed GeoChat, a new vision-language model (VLM) specifically designed for remote sensing imagery. GeoChat addresses the limitations of general-domain VLMs in accurately interpreting high-resolution remote sensing data, offering both image-level and region-specific dialogue capabilities. The model is trained on a novel remote sensing multimodal instruction-following dataset and demonstrates strong zero-shot performance across tasks like image captioning and visual question answering.
MBZUAI researchers introduce PG-Video-LLaVA, a large multimodal model with pixel-level grounding capabilities for videos, integrating audio cues for enhanced understanding. The model uses an off-the-shelf tracker and grounding module to localize objects in videos based on user prompts. PG-Video-LLaVA is evaluated on video question-answering and grounding benchmarks, using Vicuna instead of GPT-3.5 for reproducibility.