Reduced order models (ROMs) address the significant computational costs associated with solving differential equations using full-order models (FOMs), and provide efficient and reliable approximations essential for simulations in science and engineering, particularly in real-time simulations and multi-query applications. Combining ROMs with advanced computational techniques such as domain decomposition and machine learning further enhances their ability to tackle complex problems. This mini-symposium aims to delve into these integrated methodologies, exploring their potential and advancements. Key topics include domain decomposition for surrogate modeling, dynamic mode decomposition, and machine learning-based ROMs. These topics will be discussed in the context of various linear and nonlinear problems, such as multi-component systems, multi-physics problems, dynamical systems with random inputs, and data assimilation. By bringing together researchers and practitioners from diverse fields, this symposium will facilitate the exchange of innovative methodologies and the identification of future research directions, thereby advancing the fields of reduced order modeling.