The field of multimodal understanding and safety perception is rapidly evolving, with a focus on developing more accurate and robust models for real-world applications. Recent research has highlighted the importance of incorporating human-centered approaches, such as eye-tracking systems and explainable AI, to better understand how people perceive safety in various environments. Additionally, there is a growing emphasis on improving the performance of large language models in domain-specific tasks, particularly in low-resource languages. The development of new benchmarks and datasets, such as those for multimodal question answering and disability parking detection, is also driving progress in the field. Notably, some papers have made significant contributions to the field, including the introduction of a computational framework for decoding human safety perception and the development of a novel pipeline for detecting disability parking from aerial imagery. For example, the paper on Human vs. AI Safety Perception presents a framework that uses eye-tracking systems and deep learning approaches to quantify human attention and identify key visual elements that influence safety perceptions. The paper on Where Can I Park introduces a deep learning pipeline called AccessParkCV for detecting disability parking and inferring quality characteristics from aerial imagery.