The field of multiscale modeling and simulation is experiencing significant developments, driven by the increasing need to accurately predict complex phenomena in various domains. A major direction in this field is the integration of data-driven and physics-based approaches, enabling the creation of more accurate and efficient models. Researchers are exploring new methods to combine machine learning techniques with traditional modeling approaches, allowing for the acceleration of simulations and the reduction of computational costs. Another area of focus is the development of novel mathematical models for complex systems, such as open membranes in Stokes flow, which are essential for understanding biological phenomena. The use of energy variational frameworks, phase-field models, and uncertainty-driven approaches is becoming increasingly popular, as these methods provide a powerful tool for modeling and simulating complex systems. Noteworthy papers in this area include:

• A study on kernel-learning parameter prediction and evaluation in algebraic multigrid methods, which demonstrates the effectiveness of Gaussian Process Regression in optimizing parameters and reducing computational time.
• A paper on mixing data-driven and physics-based constitutive models using uncertainty-driven phase fields, which introduces an adaptive mixture approach that enables significant reductions in data collection time.