The field of resource optimization and energy efficiency is rapidly evolving, with a focus on developing innovative solutions to improve the performance and sustainability of complex systems. Recent research has explored the application of deep reinforcement learning, physics-informed neural networks, and hybrid learning frameworks to optimize resource allocation, reduce energy consumption, and enhance overall system efficiency. Notably, the use of customized techniques, such as two-stage frameworks and feature extraction modules, has shown promising results in improving the accuracy and speed of resource optimization algorithms. Furthermore, the integration of thermal modeling and optimal allocation of safety-critical tasks on heterogeneous MPSoCs has led to significant reductions in temperature and energy consumption. Overall, the field is moving towards the development of more sophisticated and adaptive optimization strategies that can effectively balance performance, energy efficiency, and reliability. Noteworthy papers include: Towards VM Rescheduling Optimization Through Deep Reinforcement Learning, which achieves a performance comparable to the optimal solution but with a running time of seconds. CPINN-ABPI: Physics-Informed Neural Networks for Accurate Power Estimation in MPSoCs, which reduces the mean absolute error by 84.7% and 73.9% for CPU and GPU, respectively. SealOS+: A Sealos-based Approach for Adaptive Resource Optimization Under Dynamic Workloads for Securities Trading System, which achieves an average CPU utilization of 78% and reduces transaction response time to 105ms.