Combining vision and language comprehension is critical for educational applications, especially for interpreting visual content and answering questions about it. The goal of educational visual question answering (Edu-VQA) systems is to support learners by providing educated answers to questions regarding images, diagrams, illustrations, and charts, thereby augmenting interactivity in learning environments. However, most current models encounter challenges when there is not good alignment between either visual regions and textual questions, resulting in incomplete or incorrect answers. Many models implement a purely global representation of an image without fine-grain cross-modal interactions required for interpreting complex educational content. In this paper, we develop an Attention-Guided Vision-Language Transformer (AG-VLT) framework to confront these significant challenges. AG-VLT uses dual encoders to embed images and textual questions into a shared embedding space, employing a cross-modal attention mechanism to selectively attend to relevant regions of the image in response to tokens in a question. Multi-layer transformer modules further refine these embeddings so that the model can learn the complex relationships between visual comprehension and reasoning with language. Our experimental results demonstrate that AG-VLT outperforms current methods in terms of accuracy as well as interpretability, bridging a gap between visual comprehension and reasoning with language in educational applications.
The rapid growth of social media use through modern communication methods has led to the spread of fake news globally. For the sake of mitigating this problem, several researchers in Artificial Intelligence have explored the use of machine learning and natural language processing (NLP) techniques to distinguish fake news from believable information. Despite significant progress in detecting fake news in the English language, still the research on detecting fake news for the Arabic Language remaining limited, highlighting evident gaps that need robust solutions. This study comprehensively explores the current detection methods for Arabic-language-based in automating fake news, summarizing the main approaches, findings, and most used datasets. Many research papers were analyzed to represent current trends and identify existing methodological limitations. Based on these insights, this study suggests applying two BERT based models (ARBERT and AraBERT) using two categorized Arabic datasets to evaluate news credibility. This study showed that the proposed detection models achieved a superior accuracy of approximately 85%. It highlights an elevated potential for detecting fake news in the Arabic language and confirms the importance of continuing research to enhance multilingual credibility assessment geared towards specialized fields.
Iris recognition on head-mounted displays (HMDs) is gaining attention for securing Extended Reality (XR) systems, but practical HMD imagery is low-resolution, off-axis, and prone to specular highlights. We present a 1:1 verification pipeline for periocular/iris images acquired by an AR/HMD device that combines deterministic segmentation with deep metric learning. Classical computer vision is used to preprocess the eye region, detect pupil and iris ellipses, and unwrap the iris into a normalized strip, which is then fed to a ResNet-50 backbone with a 128-dimensional embedding head trained using triplet loss. Verification is performed by thresholding the Euclidean distance between L2-normalized embeddings. On a subset of the public PolyU AR/VR iris dataset comprising 92 subjects, we evaluate under a balanced one-session protocol with 2,000 genuine and 2,000 impostor pairs. The proposed verifier achieves ROC–AUC = 0.9803 and EER = 6.40%. At a Youden operating point, accuracy and balanced accuracy are 94.03% with FAR = 7.55% and FRR = 4.40%, while at a looser security budget of FAR = 10% the system attains TAR = 97.20%. These results show that a simple metric-learning encoder coupled with deterministic segmentation provides a strong, transparent baseline for iris verification in HMD-based XR authentication.
This is the newest technique to make sure that smart contracts are always secure in very important domains, like world banking. Quantum computers can’t break blockchain rules, hence this is conceivable. The major purpose of this research is to see whether post-quantum cryptography can be combined with blockchain technology to make it safer to move money across borders. Quantum computers might easily get into older blockchain systems since they employ well-known encryption methods like ECC and RSA. This makes it easy for other people to hurt them. As quantum computing becomes more common, the safety of smart contracts used in international banking is becoming more and more at danger. We think that the Lattice-Based Cryptographic Integration Framework (LBCIF) might help with these issues. This system leverages lattice-based post-quantum algorithms like Kyber and Dilithium. These algorithms may keep cryptography safe against quantum assaults. You can use these algorithms to sign smart contracts, prove who you are, and agree on blockchains. The LBCIF is an excellent instrument for trade finance since it uses smart contracts to make settlements and letters of credit with banks all around the world automatic.The framework ensures secure key exchange, digital signature validation, and interoperable transactions across regulatory environments. Findings show that LBCIF maintains performance while significantly improving resistance to quantum threats. It enables secure, scalable, and regulation-compliant smart contract execution for global financial operations.
Named Entity Recognition (NER) is a fundamental task in Natural Language Processing (NLP), and its significance is amplified in the biomedical field, especially for resource-constrained languages like Arabic. This study proposes a rule-based system for the identification of disease-related entities in Arabic biomedical texts. The method accurately finds relevant terms by using dictionary matching along with a set of linguistic rules. The system uses fuzzy matching based on the Levenshtein similarity measure to deal with the common changes in spelling and terminology. This lets it find entities that are spelled differently but are still the same. The model achieved strong evaluation results, with an F1-score of 90.8%, a recall of 90.8%, and a precision of 90.7%. These outcomes demonstrate that competitive performance can be achieved through a well-structured rule-based framework enhanced by similarity techniques, without the need for advanced machine learning methods or external resources.
