This study evaluated the adversarial robustness of five state-of-the-art Arabic Language Models against various Arabic adversarial attacks at character, word, and sentence levels. It found that diacritic insertion could reduce model accuracy by up to 92%, while manipulating Arabic conjunctions led to a 58% accuracy degradation, and paraphrasing reduced performance by an average of 76%. While adversarial training improved overall resilience, particularly for MARBERT and AraBERT, challenges against character-level noise persist. Why it matters: These findings are crucial for understanding and mitigating security vulnerabilities in Arabic AI, guiding the development of more robust and safe Arabic NLP systems.
The paper introduces VENOM, a text-driven framework for generating high-quality unrestricted adversarial examples using diffusion models. VENOM unifies image content generation and adversarial synthesis into a single reverse diffusion process, enhancing both attack success rate and image quality. The framework incorporates an adaptive adversarial guidance strategy with momentum to ensure the generated adversarial examples align with the distribution of natural images.
This paper introduces a novel black-box adversarial attack method, Mixup-Attack, to generate universal adversarial examples for remote sensing data. The method identifies common vulnerabilities in neural networks by attacking features in the shallow layer of a surrogate model. The authors also present UAE-RS, the first dataset of black-box adversarial samples in remote sensing, to benchmark the robustness of deep learning models against adversarial attacks.
Faris Almalik, an MBZUAI alumnus, discusses his journey from mechanical engineering to becoming a senior data scientist at the Department of Government Enablement – TAMM in Abu Dhabi. Almalik emphasizes his passion for continuous learning and knowledge sharing, which has driven his career across defense, education, finance, and government sectors. He published his first paper on adversarial attacks on medical imaging at MICCAI during his time at MBZUAI. Why it matters: This highlights the success of MBZUAI in producing impactful AI professionals who contribute to both research and practical applications in key sectors within the UAE.
MBZUAI launched the AI Quorum, a winter series from October 2022 to March 2023, to stimulate AI research. The first session, led by Professor Michael Jordan, focused on collaborative learning with around 20 research experts. Discussions covered the use of edge devices like cell phones and hospitals providing data to build large models, as well as risks like free-riding and adversarial attacks. Why it matters: The AI Quorum initiative positions MBZUAI as a hub for global AI collaboration, addressing key challenges and opportunities in collaborative learning for real-world applications.
This article discusses adversarial training (AT) as a method to improve the robustness of machine learning models against adversarial attacks. AT aims to correctly classify data and ensure no data fall near decision boundaries, simulating adversarial attacks during training. Dr. Jingfeng Zhang from RIKEN-AIP will present on improvements to AT and its application in evaluating and enhancing the reliability of ML methods. Why it matters: As ML models become more prevalent in real-world applications in the GCC region, ensuring their robustness against adversarial attacks is crucial for maintaining their reliability and security.